Messengers From The Cosmos: Francesco Riggi

UNITEXT for Physics
Francesco Riggi
Messengers
from the Cosmos
An Introduction to the Physics of Cosmic
Rays in Its Historical Evolution
UNITEXT for Physics
Series Editors
Michele Cini, University of Rome Tor Vergata, Roma, Italy
Attilio Ferrari, University of Turin, Turin, Italy
Stefano Forte, University of Milan, Milan, Italy
Guido Montagna, University of Pavia, Pavia, Italy
Oreste Nicrosini, University of Pavia, Pavia, Italy
Luca Peliti, University of Napoli, Naples, Italy
Alberto Rotondi, Pavia, Italy
Paolo Biscari, Politecnico di Milano, Milan, Italy
Nicola Manini, University of Milan, Milan, Italy
Morten Hjorth-Jensen, University of Oslo, Oslo, Norway
Alessandro De Angelis , Physics and Astronomy, INFN Sezione di Padova,
Padova, Italy
UNITEXT for Physics series publishes textbooks in physics and astronomy,
characterized by a didactic style and comprehensiveness. The books are addressed
to upper-undergraduate and graduate students, but also to scientists and researchers
as important resources for their education, knowledge, and teaching.
Francesco Riggi
Messengers from the Cosmos

An Introduction to the Physics of Cosmic
Rays in Its Historical Evolution
Francesco Riggi
Department of Physics and Astronomy
University of Catania and INFN
Catania, Italy
ISSN 2198-7882 ISSN 2198-7890 (electronic)

UNITEXT for Physics
ISBN 978-3-031-24761-3 ISBN 978-3-031-24762-0 (eBook)
https://doi.org/10.1007/978-3-031-24762-0
© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature
Switzerland AG 2023
This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether
the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse
of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and
transmission or information storage and retrieval, electronic adaptation, computer software, or by similar
or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or
the editors give a warranty, expressed or implied, with respect to the material contained herein or for any
errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
All areas of scientific investigation may be considered, especially in the eyes of those
directly involved in that activity, as an exciting adventure. If we looked for a definition
of “adventure”, we would find that it is described as a singular and extraordinary
story, or as a risky but fascinating and attractive undertaking, especially for what is
unexpected and unknown, as pointed out by the very same Latin root of the word,
adventura, what will happen. With these premises, it is not easy to find a discipline
that presents, more than others, these characteristics, such as the physics of cosmic
rays.
The study of the radiation of cosmic origin presented, since its inception, pecu-
liarities capable of involving researchers in experimental activities that went beyond
the boundaries of their research laboratories, pushing them to carry out their studies
and their measurements in the most disparate environments of our planet. The debate
about the possible nature of this radiation, whether it was linked to the phenomena of
terrestrial radioactivity, which somehow began to be known already a century ago,
or it was rather the result of processes that took place outside the Earth, in the whole
cosmos, put the object of the investigations in a specific perspective and opened up
the horizon of places where to look for clues that could have given an answer to this
question.
The history of cosmic ray physics, especially in the first decades that followed the
first evidence of the existence of this radiation, shows us an incessant activity by the
researchers involved, who did not disdain to engage themselves in the most disparate
working conditions: We will see them climbing mountains, boarding balloons, pres-
surized capsules or aircrafts to go to high altitudes, placing their equipment under the
water of seas and lakes, or in the deepest mines, performing scientific expeditions in
all the globe, from equatorial to polar regions, all to study the nature of this radiation
and characterize its behavior. We must consider how this happened about a century
ago, with the tools and difficulties of the time, which makes these activities even
more full of adventure. We may say this is an attitude and a scientific practice that
in some way still goes on today, since the research in this area, more generally in
the field that we call today astroparticle physics, has reached the boundaries of space
v
vi Preface
outside the Earth and, even on our planet, makes use of environments that are among
the least comfortable, with huge and complex detection apparatuses.
The first evidence of the existence of radiation of extraterrestrial origin refers,
as it is well known, to the first experiences, carried out with electroscopes, of
the phenomenon of air ionization. The observation of the apparently spontaneous
discharge of these instruments indicated the existence of a source capable of ionizing
the air inside them. Was it possible to explain this phenomenon solely on the basis of
the radiation emitted by radioactive substances present in the Earth’s soil? For a long
time, this remained an open problem, to which the observations made by numerous
experimenters gave decisive contributions, until the reasonable certainty was reached
that the cause was rather to be found in a radiation coming from outside the Earth, a
radiation of cosmic origin.
In the following decades, the nature of this radiation, its characteristics, and its
links with the nascent physics of elementary particles became more and more defined.
The experimental observations benefited from the development of equipment and
detection techniques that allowed not only more detailed and precise studies of the
known phenomena, but also of highlighting qualitatively new phenomena, such as
the existence of enormous showers of particles in the Earth’s atmosphere, produced
by the arrival of a single particle of very high energy.
Looking at the events that characterized the development of this discipline in
a historical key is particularly useful from an educational point of view. Indeed,
it allows us to better understand how knowledge developed, and through which
steps—and also which errors, or approximate knowledge—we have achieved a more
correct point of view. But it is also important to see how it was possible to make
a remarkable series of observations, corroborated by interpretations that turned out
to be in most cases correct, although the tools and observational techniques of the
past were enormously simpler than the current ones. For this reason, it was useful
to discuss, even at an introductory level, what were the observational tools of the
cosmic radiation, those which became from that moment on the detectors of particles
and radiation.
There is another important aspect from an educational point of view in dealing
with the main phenomena relating to the physics of cosmic rays. In fact, a remarkable
variety of projects and teaching activities has developed around this area, especially
in recent decades, in the awareness that the physics of cosmic rays can be an excellent
means of introducing students and science enthusiasts to the problems and results
of modern physics. This was accompanied by the possibility of using even simple
detection devices, such as the Geiger counter in the first instance—but also more
complex ones—to organize a multiplicity of experiments that can be carried out
at low cost but with considerable potential in terms of possible scientific activities
and outcomes. The author has been involved for more than twenty years in various
educational projects, local or international, related to cosmic ray physics. In many
cases, these activities have given rise to what is now often referred to as “citizen
science”, i.e., the possibility that the general public can also carry out scientific
activities by learning something directly about the way in which scientific research
proceeds.
Preface vii
If this is the spirit in which these pages were written, it must be said that it is always
difficult to understand which target, which audience to turn to when writing a book.
An attempt has been made to provide as broad a picture as possible of the problems
and historical developments of this discipline, maintaining an understandable level,
for most of the topics, even by the passionate science reader. At the same time,
however, attempts have been made to present many phenomena in a quantitative way,
often making use of the original results, in order to provide non-generic information
on the phenomena in question. In this direction, we must consider several plots that
report the data published in the original articles, in some cases slightly reworked to
make them more readable.
The idea of this text is therefore to provide a working tool to those who wish
to get more involved into this field of physics, whether they are students attending
undergraduate courses in physics or simple science enthusiasts. To achieve this aim,
an extensive bibliography has been collected, which includes, in addition to the list
of the major textbooks published on the subject, also many articles of a historical
and biographical nature, as well as a long list of detailed references, over 500, to
the original articles cited along the text. The invitation for those wishing to further
deepen their knowledge of this sector is therefore to start from this text as a guide for
further reading. Another aspect that may contribute to consider this text as a working
tool is the presence of a certain number of Appendices, where results obtained by
numerical calculations are also presented, with an invitation to readers to implement
the procedures themselves and get additional results.
Of course, although we may try to present a broad phenomenology, cosmic ray
physics is strictly related to several other aspects of scientific knowledge that have
only marginally been mentioned here, such as the overall picture of elementary
particle physics, the physics of neutrinos, that of gravitational waves, and in general
many important problems in astroparticle physics.
Instead, we wanted to give, in a specific chapter, also a picture of the different
application aspects in which muons, as a penetrating charged component of the
secondary cosmic radiation, play an increasingly important role, and which in recent
years have given rise to numerous projects in the most diverse sectors, from the
prevention of illicit transport of radioactive materials in containers to the study of
the internal structure of volcanoes, to monitoring the stability of civil buildings, to
name only those in which the author has been involved.
As personally engaged in various educational and research activities related to
the physics of cosmic rays, I would like to thank many people with whom I have
interacted in these decades. Among these, the undergraduate and Ph.D. students who
collaborated to these activities, many of them are now being enrolled in research or
academic Institutions. I particularly thank all the colleagues of the Extreme Energy
Events (EEE) Collaboration, in which the author has participated since its begin-
ning, with whom it was a privilege and a pleasure to collaborate. The interactions
with the thousands of high-school students and teachers participating in the EEE
project in recent years have provided an important framework to place many of the
considerations set out in this text. Critical reading of parts of this manuscript by
viii Preface
several colleagues is warmly acknowledged. A special thank to the Springer staff

for help in editing the manuscript. Finally, I would also like to thank the members
of my family, my wife, my sons and daughter, together with our grandchildren,
with whom I have sometimes tried to share the passion for investigating the world
around us.
Catania, Italy Francesco Riggi

October 2022
Contents
1 The Discovery of the Cosmic Radiation . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Terrestrial Radioactivity and First Experiences
with Electroscopes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 Investigations in the Atmosphere . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.4 Victor Hess and the Evidence for an Extraterrestrial
Radiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
1.5 Towards a Confirmation of Hess’s Results . . . . . . . . . . . . . . . . . . 22
2 Confirmation of the Existence of a Cosmic Radiation . . . . . . . . . . . . . 27
2.1 Further Investigations in Europe During and After
the First World War . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.2 Robert Millikan and the First American Contributions
to Cosmic Ray Physics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
2.3 The Acceptance of the Idea of a Cosmic Radiation . . . . . . . . . . . 37
3 The Nature of the Cosmic Radiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.1 The Influence of the Earth’s Magnetic Field . . . . . . . . . . . . . . . . . 41
3.2 Campaigns for Measuring the Intensity of the Cosmic
Radiation in Various Geographical Locations . . . . . . . . . . . . . . . . 47
3.3 The Debate on the Corpuscular or Radiative Nature
of the Cosmic Radiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
3.4 Further Contributions in Europe and Other Countries
for Understanding the Nature of the Cosmic Radiation . . . . . . . . 61
3.5 Protons as an Essential Component of Primary Radiation? . . . . 72
4 New Particles and Their Links with the Cosmic Radiation . . . . . . . . 75
4.1 The Discovery of New Particles and the Links
with the Understanding of Cosmic Radiation . . . . . . . . . . . . . . . . 75
4.2 Properties of µ-Mesons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
4.3 The Discovery of the Pion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
4.4 The Discovery of the Neutron . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
ix
x Contents
5 Developments of the Techniques for the Detection

of Cosmic Rays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
5.2 From Wulf’s Electroscopes to Automatic Recording
Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
5.3 Ionization Chambers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
5.4 Proportional Counters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
5.5 Wilson Cloud Chamber . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
5.6 Geiger-Müller Counters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
5.7 Electronics and Coincidence Techniques . . . . . . . . . . . . . . . . . . . . 105
5.8 Nuclear Emulsions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
5.9 Detectors Based on Scintillators . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
6 The Interaction of Primary Cosmic Rays in the Atmosphere . . . . . . 117
6.1 The First Evidence of Nuclear Interactions of Cosmic Rays . . . 117
6.2 Interactions in the Atmosphere and First Evidence
of a Complex Primary Radiation . . . . . . . . . . . . . . . . . . . . . . . . . . 120
6.3 Production of Other Particles in Nuclear Interactions . . . . . . . . . 122
6.4 The Role of High-Altitude Laboratories . . . . . . . . . . . . . . . . . . . . 123
7 Extensive Air Showers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
7.1 Secondary Processes and Local Showers . . . . . . . . . . . . . . . . . . . . 127
7.2 First Evidence of the Existence of Atmospheric Showers . . . . . . 129
7.3 An “Operational” Definition and the First Properties
of Extensive Air Showers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
7.4 Towards a More Complete Description of the Formation
of Extensive Air Showers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
7.5 The Study of Atmospheric Showers Since the 1940s . . . . . . . . . . 140
7.6 The Longitudinal Development of an Extensive Air Shower . . . 141
7.7 The Transverse Development of an Extensive Air Shower . . . . . 147
7.8 The Time Profile of an Extensive Air Shower . . . . . . . . . . . . . . . . 152
8 The Detection of Extensive Air Showers . . . . . . . . . . . . . . . . . . . . . . . . . 155
8.1 Direct and Indirect Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
8.2 Arrays of Particle Detectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
8.3 Detector Arrays Based on the Cherenkov Effect . . . . . . . . . . . . . 159
8.4 Fluorescence Detectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
8.5 Detection of Radio Waves Associated with Extensive Air
Showers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
8.6 An Example of Reconstruction of Extensive Air Showers
in the 1950s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
8.7 Arrays for the Reconstruction of Extensive Air Showers . . . . . . 166
9 The Primary Cosmic Radiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
9.2 The Hadronic Component and the Energy Spectrum . . . . . . . . . . 174
9.3 Composition of the Hadronic Component . . . . . . . . . . . . . . . . . . . 179
Contents xi
9.4 Electrons and Positrons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

9.5 Other Components in the Primary Radiation . . . . . . . . . . . . . . . . 187
9.6 The Intensity of the Primary Radiation at Different
Altitudes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
9.7 Possible Anisotropy Effects in the Primary Radiation . . . . . . . . . 192
10 The Secondary Cosmic Radiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
10.1 Composition of the Secondary Radiation . . . . . . . . . . . . . . . . . . . 195
10.2 Muons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
10.3 Electrons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
10.4 Gammas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
10.5 Charged Hadrons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
10.6 Neutrons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
10.7 Nuclei . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
11 The Influence of the Earth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
11.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
11.2 The Interaction with the Atmosphere and Meteorological
Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
11.3 The Influence of the Earth Magnetic Field . . . . . . . . . . . . . . . . . . 219
11.4 Angular Distribution of Muons and the East–West Effect . . . . . . 224
11.5 The Latitude Effect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
11.6 Other Influences on the Cosmic Ray Flux Due to the Earth
Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227
12 The Secondary Cosmic Radiation and the Influence of the Sun . . . . 229
12.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
12.2 Periodic Phenomena in the Sun and Solar Cycles . . . . . . . . . . . . 230
12.3 Modulation of the Cosmic Ray Flux Due to the Sun . . . . . . . . . . 231
12.4 Forbush Variations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
12.5 Other Effects Related to Solar Activity . . . . . . . . . . . . . . . . . . . . . 239
13 Interaction of Muons with Matter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
13.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
13.2 Energy Loss of Muons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
13.3 Range of Muons in Matter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
13.4 Multiple Scattering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
14 Cosmic Radiations Underground, Under Water and Under
the Ice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
14.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
14.2 Underground Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250
14.3 Measurements Underwater and Under the Ice . . . . . . . . . . . . . . . 255
xii Contents
15 The Origin of Cosmic Rays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257

15.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
15.2 Some Historical Considerations About the Acceleration
Mechanisms and the Origin of Cosmic Rays . . . . . . . . . . . . . . . . 260
15.3 Fermi Acceleration Mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . . 262
15.4 The Role of Supernovae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264
15.5 The High-Energy Extragalactic Component . . . . . . . . . . . . . . . . . 268
16 The Impact of Cosmic Rays in Applications and in Daily Life . . . . . 271
16.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
16.2 Production of Radioactive Isotopes by Cosmic Rays
and Dating Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272
16.3 Cosmic Ray Dating Outside the Earth . . . . . . . . . . . . . . . . . . . . . . 275
16.4 The Radiation Dose Produced by Cosmic Rays on Earth
and in the Solar System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
16.5 Electronics and the Effect of Cosmic Radiations . . . . . . . . . . . . . 281
16.6 Muons and the Origin of Tomographic Techniques . . . . . . . . . . . 283
16.7 Tomographic Techniques Based on Cosmic
Muon Absorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
16.8 Muon Tomography and Scattering from High-Z Materials . . . . . 288
16.9 Imaging Techniques Based on Secondary
Particle Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
16.10 Monitoring of the Stability of Buildings by Tracking
Cosmic Muons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292
16.11 Additional Possible Applications of Muon Tomography . . . . . . . 294
16.12 The Impact of Cosmic Rays on Cloud Formation . . . . . . . . . . . . 295
16.13 Use of Extensive Air Showers in Time Synchronization . . . . . . . 295
Appendix A: A Calculation of the Flux at the Top of the Eiffel

Tower Due to Soil Radioactivity . . . . . . . . . . . . . . . . . . . . . . . . 297
Appendix B: The Absorption Coefficient in Water
and the Directionality of Cosmic Rays . . . . . . . . . . . . . . . . . . 301
Appendix C: Geographic and Geomagnetic Latitude . . . . . . . . . . . . . . . . . 303
Appendix D: The Magnetic Rigidity of Particles . . . . . . . . . . . . . . . . . . . . . 309
Appendix E: The Energy Loss of Charged Particles
and the Estimate of the Muon Mass . . . . . . . . . . . . . . . . . . . . 313
Appendix F: List of High-Altitude Observation Stations
in the Mid-1950s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317
Appendix G: An Estimate of the Particle Density in an Extensive
Air Shower . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319
Appendix H: The Relationship Between Altitude and Atmospheric
Depth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321
Contents xiii
Appendix I: Gaisser-Hillas Parameterization of the Longitudinal

Profile of a Shower . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323
Appendix J: The Thickness of Air Crossed by a Particle
in the Atmosphere . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
Appendix K: Evaluation of the Shower Direction from the Relative
Timing of Several Detectors . . . . . . . . . . . . . . . . . . . . . . . . . . . 327
Appendix L: Parameterizations of the Muon Spectrum at Sea Level . . . 329
Appendix M: The Flux of Underground Muons . . . . . . . . . . . . . . . . . . . . . . 333
Appendix N: Detection of Bit-Flip Errors Originated by
Cosmic Rays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335
Notes to the Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 341
Chapter 1
The Discovery of the Cosmic Radiation
Abstract The balloon ascent made by Victor Hess in 1912, up to an altitude

of 5300 m, is generally associated to the discovery of the cosmic rays. Related
phenomena at that time were the terrestrial radioactivity, the absorption of the gamma
radiation in air and the air ionization, all aspects extensively studied by the use of
electroscopes. Even before the famous flight by Hess, several people contributed to
these studies: Theodore Wulf with his measurements on the top of the Eiffel Tour,
Julius Elster and Hans Geitel who carried out measurements of the air ionization
in different locations, Albert Gockel and Karl Bergwitz performing high altitude
experiments, and Domenico Pacini with his underwater measurements. The chapter
reviews all these contributions and analyzes in detail, by tables and plots, the original
data obtained by Victor Hess in 1911–1912 and by Werner Kolhörster in 1913–1914
at even higher altitudes, which definitely demonstrated the existence of an extrater-
restrial radiation. Appendix A also provides examples and suggestions on how to
perform numerical calculations of the expected gamma ray flux at the top of the
Eiffel Tour, originating from ground radioactivity.
1.1 Introduction
August 7th, 1912, early morning: a young Austrian physicist, Victor Franz Hess
(1883–1964), at that time in his thirties, is ready to leave aboard a hydrogen-filled
balloon for a scientific expedition, from a location called Aussig at the time (now
Ústí nad Labem, in the Czech Republic). It is not the first of his balloon flights: in
the previous months he has already made other excursions, reaching almost 3000 m
above sea level. Today’s flight, the seventh in the series, will take him even higher,
to an altitude of over 5000 m.
His aim is to verify whether the ionization of air at high altitude, measured by the
discharge of the electroscopes he is carrying with him, increases significantly with
altitude, confirming what he has already observed in several previous flights, carried
out at lower altitudes. A few hours later Hess will have confirmation of what he is
looking for: his flight, the seventh in the series, which brought him to a maximum
altitude of 5350 m, will allow to establish that the cause of the electroscope discharge
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 1

F. Riggi, Messengers from the Cosmos, UNITEXT for Physics,
https://doi.org/10.1007/978-3-031-24762-0_1
2 1 The Discovery of the Cosmic Radiation
phenomenon is due to something whose intensity undoubtedly increases with altitude

and therefore cannot be linked to the radioactivity of terrestrial rocks.
This event has been generally associated to the discovery of the so-called cosmic
radiation, a term that was coined many years later by the American physicist Robert
Millikan (1868–1953), according to the commonly used interpretation, but which
had actually been used many years before him.
Many discoveries in physics are objectively difficult to associate with defined
dates or with single events, even if—often a posteriori—it becomes evident how that
experiment, that formulation, that model represented a real novelty compared to the
status of previous knowledge, such as to assign to them a specific role, a turning point
in the history of that discipline. It is also the case of the discovery of cosmic rays,
which presupposes a series of previous events, intuitions, verifications, significant
contributions given by other authors, which inevitably intertwine with those events
that are usually classified as “discovery”.
In the specific case of the discovery of cosmic radiation and its subsequent develop-
ments many physical phenomena have played an important role, such as air conduc-
tivity and natural radioactivity, and the interpretation of the results concerned with
the study of cosmic radiation involved at the same time other branches of physics,
such as the knowledge of subatomic particles.
1.2 Terrestrial Radioactivity and First Experiences

with Electroscopes
The phenomenon of the spontaneous discharge of electroscopes had already been

investigated by Charles-Augustin de Coulomb (1736–1806) in 1785 [Coulomb1785].
An electroscope, in its simplest version, is a device capable of detecting the presence
of an electrical charge. It may consist of two thin metal foils (for example two gold
leaves), placed in contact at the upper end, through a support leaving them free
to separate at the bottom; the leaflets are positioned inside an insulating container
and electrically connected to an external conductor. It is a known experience that
by touching the external conductor with an electrically charged body, or simply
approaching the charged body to it, by induction, the leaflets diverge, both being
charged with the same sign and therefore repelling each other. At the beginning of
1900 it had already been known for a long time that an electroscope gradually lost
its charge, even if isolated from an electrical point of view and in the absence of any
direct contact with other bodies. The speed with which the electroscope discharges
depends on the pressure of the air inside it, as the observations of William Crookes
(1832–1919), British physicist and chemist, had shown, suggesting that the cause of
the discharge was linked to the ionization of the air [Crookes1878] (Fig. 1.1).
In those years, Charles Vernon Boys (1855–1944) had also observed the same
phenomenon, concluding that this loss was due to convection processes through the
air [Boys1889]. At the end of the 1800s, however, the conductivity of air was the
1.2 Terrestrial Radioactivity and First Experiences with Electroscopes 3
Fig. 1.1 Williams Crookes (1832–1919). Source AIP Emilio Segrè Visual Archives
subject of very extensive studies by numerous scientists. Although the knowledge

of the detailed atomic structure was yet to come, the fact that matter consisted of
atoms, each corresponding to a different chemical species, and that atoms could form
molecules, was already known. As well as the fact that ordinary matter is neutral,
containing an equal amount of positive and negative charge. The discharge of an
electroscope could therefore be interpreted based on the fact that the air is always
slightly ionized, that is, it contains a small fraction of electrons and positive ions,
which could be attracted by the electroscope leaves charged with the opposite sign.
Subsequently, the knowledge of the phenomena related to natural radioactivity,
the discovery of cathode rays and that of X-rays had allowed to establish that these
causes certainly contributed to the conductivity of the air and therefore could explain
the progressive loss of electric charge by the electroscope (Fig. 1.2).
Charles Thomson Rees Wilson (1869–1959), an English physicist who had started
his career working in the laboratory of Joseph Thomson (1856–1940), had already
conducted, in 1896, his first observations with the prototype of what would later
be the cloud chamber, investigating the conductivity of the air exposed to X-rays
[Wilson1896, Wilson1897]. Further studies carried out in the following years had
allowed to understand different aspects of the effect that X-rays had on the ioniza-
tion of the air and on the condensation produced in the gases by these ionizing
Fig. 1.2 Charles Thomson Rees Wilson (1869–1959). Source AIP Emilio Segrè Visual Archives
agents [Wilson1898, Wilson1899, Wilson1900a, Wilson1900b, Wilson1901a]. In

the same period, similar results on the phenomenon of the spontaneous loss, even
in pure air, of the charge possessed by a charged conductor were obtained by Julius
Elster (1854–1920) and Hans Geitel (1855–1923), two German physics teachers who
had conducted experiments using their private laboratory [Elster1900a, Elster1900b,
Geitel1901].
The overall results of this activity, carried out by numerous experimenters around
the end of 1800, could therefore be summarized in the fact that a charged conductor
lost its charge in the air regardless of the environmental conditions (light or dark),
and that this loss was proportional to the pressure of the air, as it had been verified by
experiments conducted inside sealed vessels. From a quantitative point of view, this
loss of charge corresponded to about 20 ions per second for each cubic centimeter
of air. It was therefore necessary to admit that an equivalent number of ions were
continuously produced in the air.
Wilson had also conducted experiments expressly to establish whether this
production of ions in the air could be attributed to radiation from sources extra-
neous to the atmosphere, such as X-rays or cathode rays, but of greater penetrating
power. These experiments involved, for example, the comparison between what was
observed on the ground and what could be observed underground, at a certain depth,
in a railway tunnel [Wilson 1901b]. However, since there was no appreciable differ-
ence between the results of the two measurements, Wilson concluded that ionization
was an intrinsic property of air itself. The term “ionization” began to be used right
from the works of this author.
We will discuss Wilson’s role and contribution to cloud chamber observations
later on. Instead, we still remain on the crucial issue of this period: the ionization of
air and the causes that produce it.
As we have already seen, it was clear that different sources of radiation could
produce the creation of ions in the air, thus discharging the electroscopes: both the
X-rays and the products emitted by radioactive decay were capable of producing such
effects. The prevailing opinion about the cause of this ionization was that the main
source of air ionization was linked to the radioactivity of the soil rocks, in particular to
the gaseous emanations emitted by certain minerals present in the rocks. In addition
to those already cited, further studies by Elster and Geitel contributed to this opinion
[Elster1899, Elster1901].
However, already in those years, there were also studies of the shielding effect
on the discharge of electroscopes, in the underlying hypothesis that the cause was a
radiation coming from above [McLennan1903a, McLennan1903b, Rutherford1903].
For example, various metal layers were used around electroscopes, to evaluate their
effect on the phenomenon of discharge, and in the case of a considerable thickness
(a few cm of lead) some sensitive effect was observed. Apparently, an additional
amount of lead, even considerable, did not further influence the discharge of the
electroscope, an aspect incomprehensible at that time but which is consistent with
the modern vision of cosmic radiation.
Another phenomenology was represented by measurements made in cellars, caves
or underground locations, again by Elster and Geitel [Elster1900b]. In some cases,
an even faster discharge of electroscopes was observed than that observed on the
surface in the open air, especially if the underground environments had been closed
for several days. It is a phenomenology that today is related to the emission of radon
gas produced by the radioactive decay of certain nuclei contained in the rocks.
All these measurements and related considerations revolved around the use of
electroscopes, as the main tool for evaluating the ionization of air. The use of elec-
troscopes was in fact very common at that time, for the study of issues related to elec-
tricity in the air, meteorology and for the understanding of atmospheric phenomena
in everyday life [DeAngelis2014]. Many of the experimenters involved in these
early studies of cosmic radiation personally contributed to the construction and
improvement of these instruments, which in some way represented the first particle
detectors.
Compared to the original leaf electroscopes, significant improvements
[Wulf1909a] were made by Theodor Wulf (1868–1946), a German physicist, teacher
in Walkenburg (Holland), a Jesuit priest, who introduced two thin quartz fibers in
place of leaflets, with the possibility of observing the deviation through a micro-
scope with a graduated scale (Fig. 1.3). Further improvements were subsequently
introduced by Victor Hess and the German Werner Kolhörster (1887–1946), so much
so that the company in charge of building such devices had five different types of
electroscopes in its catalog, according to the needs of the different experimenters
[Walter2012]. Elster and Geitel had also dedicated themselves to improving the char-
acteristics of these electroscopes, achieving better sensitivity [Fricke2017]. Today
many of these original instruments can be viewed in collections of ancient instru-
ments, for example at the European Center for the History of Physics in Austria
[Kahlert2011], in many university or school collections, or on sites showing exten-
sive virtual collections of electroscopes and electrometers, for example that viewable
at the site of the Oak Ridge Associated Universities [ORAU].
The most common explanation of the phenomenon of the spontaneous discharge
of electroscopes, linked to the conductivity of the air, was therefore associated, as we
have seen, with the presence of ionizing radiation sources, in the form of radioactive
impurities existing in common materials placed near the electroscope.
Although some ideas about the possibility of highly penetrating radiation of
extraterrestrial origin were already circulating at that time, and it seems that even
Marie Curie (1867–1934) had speculated, in 1898–1899 [Curie1898, Curie1899],
about the possibility of a highly penetrating radiation coming from the Sun, no
quantitative results had been obtained that could confirm these hypotheses.
In the following decade a series of quantitative investigations was carried out
by various experimenters, both in Germany and in Great Britain, USA, Switzer-
land, Austria and also in Italy, by Domenico Pacini (1878–1934). The improvement
of electrometers, by Wulf, allowed more precise measurements of air ionization,
performed in different environments, on the land, on the surface of water, on lakes
Fig. 1.3 Left: schematic of a Wulf electroscope, in which the divergence of the fine quartz fibers
can be quantitatively observed using a microscope and a graduated scale. On the right a photo of
a real model of an electroscope, used by Wulf around 1912. Source Wikimedia Commons, A. De
Angelis and C. Guaita, from the collection of R. Fricke
and on glaciers. The idea that guided these observations was linked to the fact that if
this ionization had been caused by any material present in the Earth’s soil, it would
have had to be smaller on the surface of the water, which contained fewer radioactive
impurities. The results, however, indicated only a slight decrease compared to the
values observed on the ground, and therefore did not allow a clear interpretation.
One of the important consequences of these investigations was the evaluation of
the absorption coefficient of gamma rays in air, water and solid ground [Eve1906,
Eve1907], due to the English physicist Arthur Stewart Eve (1862–1948) (Fig. 1.4),
who worked mainly in Canada. Gamma rays, made up of high-energy photons, typi-
cally produced in radioactive phenomena involving nuclei, have the greatest penetra-
tion capacity among the radiations emitted by radioactive substances present in the
soil. It can be assumed as a first approximation that the gamma flux capable of crossing
a given material thickness x is governed by an exponential relation, of the type
I = I0 e−µx
where I 0 and I represent the original flux and that measured after a thickness x, and µ
is the linear absorption coefficient, expressed for example in cm−1 if the thickness is
expressed in cm. The following table shows the value of some absorption coefficients,
extracted from the original work of Eve [Eve1906] for some materials, compared with
the values commonly accepted today for air at different gamma energies, tabulated
on the website of the National Institute of Standards and Technology [NIST].
Based on the absorption coefficients shown in Table 1.1, the reduction of the
gamma ray flux emitted in a specific direction would result in an attenuation of about
50% after a distance of 150 m of air crossed. On the other hand, these absorption
coefficients are much higher in solid rocks: a 50% reduction in the gamma ray flux
under the same conditions would be obtained after only 7 cm of rock crossed. From
this it can be estimated that only the most superficial part of the ground can contribute
significantly to the radiation present in the air.
If the emitting source were point-like, with isotropic emission, and we wanted
to consider the effect produced by the gamma at a certain distance d in the air, for
example the number of ions per second per unit of volume at a distance d from the
source, this would be proportional to exp (−µ d)/d2 , where the exponential term
accounts for the absorption in the material and the term 1/d2 , inversely proportional
to the square of the distance, takes into account the subtended solid angle. The overall
result is that the effect would diminish very quickly with distance.
In the case of radioactivity emitted from the Earth’s soil, however, the emitting
source is not point-like, but can be considered as a first approximation to be uniformly
diffused, both horizontally and vertically, below the ground. Of course, the radioac-
tive material present beyond a certain depth and beyond a certain distance from the
place where the ionization is considered contributes little to this effect, both because
there is an attenuation due to the absorption in the material (rock or air) and because
the radiation is emitted isotropically in all directions and therefore is attenuated with
a trend inversely proportional to the square of the distance. The problem is therefore
Fig. 1.4 Arthur Stewart Eve (1862–1948)
Table 1.1 Absorption coefficients for gamma rays coming from the ground, in cm−1 , for different
materials
Material Absorption coefficient (cm−1 )
Air 0.000044
Water 0.034
Solid rock 0.092
Air (E = 1 MeV) 0.000078
Air (E = 2 MeV) 0.000054
Air (E = 3 MeV) 0.000044
Water (E = 3 MeV) 0.039
The values for air, water and solid rock are extracted from [Eve1906]. The other values, corre-
sponding to fixed gamma energies, are current values, taken from the site of the National Institute
of Standards and Technology [NIST]
more complex than the absorption of the gammas emitted by a point source that
travels a single path, of length x, up to the point where the flux is measured.
The situation around 1909 could be summarized by the conclusions reached up
to that moment by Wulf: the result of the experiments shows that the penetrating
radiation is produced by radioactive substances present in the subsoil, up to about
one meter deep. If a certain fraction of this radiation comes from the atmosphere, it is
however negligible compared to that due to the substances present in the subsoil and
cannot be detected with the tools available at the time [Wulf1909b]. In these conclu-
sions, as appears evident, it is not excluded that there may be an additional radiation,
in addition to that due to terrestrial radioactivity, but it is quantitatively assessed
that it cannot be clearly identified, given the experimental working conditions and
the precision of the instruments, an absolutely reasonable and modern attitude to
proceed. How then to take a step forward? We will see in the next section the results
obtained in 1909 at the top of the Eiffel Tower by Wulf himself.
Even before these measurements, however, as early as 1907, the Italian physicist
Domenico Pacini was also conducting experiments using electrometers (similar to
those used by Wulf), in different environments (in the mountains, on the surface of a
lake or sea, and finally also under a certain depth of water). In one of these measure-
ments, conducted in a boat at a certain distance from the mainland, and with a water
thickness of about 4 m under the apparatus, Pacini had observed values that were not
too different (about 2/3) compared to those measured on the mainland [Pacini1910].
The absorption coefficient of gamma rays in water, measured previously, shows that
already a thickness of one meter of water typically reduces the gamma ray flux
to a few percent of the initial value. A thickness of a few meters therefore makes
the contribution due to the radioactivity of the seabed absolutely negligible on the
surface. Pacini’s conclusions, even following further investigations, were clearly in
favor of a significant cause of ionization due to the atmosphere, independent of the
radioactivity of the soil [Pacini1912]. It would therefore seem that an important role
in the awareness of the existence of a radiation of extraterrestrial origin is to be
assigned to the Italian physicist, as documented in detail by Carlson and De Angelis
[Carlson2011, DeAngelis2010, DeAngelis2012], although his contribution has not
been adequately recognized (Fig. 1.5).
Beyond the recognition of the specific role of the many people who contributed
to these studies, it can be said that around 1910 various evidence seemed to indi-
cate the presence of something not attributable to the radioactivity of the soil or of
the air circulating near the ground. The experimental working conditions of these
experiments (shielding of the devices, degree of contamination of the materials …)
were however not sufficiently defined and the quantitative results obtained were not
precise enough to allow certain conclusions. Despite the difficulty of reconciling the
results obtained with the hypothesis of radioactivity due to the Earth, a difficulty
recognized by many, there was not however any real alternative, which referred to
other known causes and allowed to explain in a quantitative way what was observed.
Fig. 1.5 Domenico Pacini (1878–1934). Courtesy of Pacini family, by A. De Angelis
1.3 Investigations in the Atmosphere
After a long series of measurements performed in the terrestrial (or marine) envi-
ronment, it was inevitable to think of obtaining more detailed information on what
was happening at altitudes higher than sea level, or in the atmosphere. These are
indeed different conditions: in the first case, although moving to higher altitudes,
for example over the top of a high mountain, one is always close to the Earth’s soil,
with the presence of rocks capable of emitting radiation. If you want to investigate
the behaviour of air ionization away from the presence of the soil, it is necessary to
move away from it instead. Measurements of both the first and the second type had
in fact been carried out by several experimenters in those years.
Elster and Geitel had compared the ionization measurements of the air at sea
level and at an altitude of 3000 m, finding a slight increase at 3000 m above sea
level [Elster1900a]. High altitude measurements had also been carried out by Albert
Gockel (1860–1927) together with Wulf, starting from 1908. The two had carried out
measurements of the ionization of the air at various altitudes, in the Alps, from 650 m
up to 3000 m [Gockel1908]. However, these measurements had not allowed to high-
light a clear influence of altitude on ionization, and the experimenters had therefore
concluded that any cosmic radiation, if it ever existed, should have contributed to the
phenomenon with a negligible intensity. Some have correctly noted [Walter2012]
1.3 Investigations in the Atmosphere 11
how the term “cosmic radiation” has already been used in this context, well before
what is commonly believed, based on the term used by Robert Millikan in 1926.
But how to get away from the ground, so as to make negligible, or at least
strongly reduce, the influence of radioactivity due to the terrestrial soil? Wulf, after
a campaign of measurements conducted with electroscopes in various European
locations [Wulf1909b], had the opportunity in 1910 to carry out measurements at
the top of the Eiffel Tower, about 300 m above the ground [Wulf1910]. The Eiffel
Tower, opened to the public about twenty years earlier, in 1889, was at that time the
tallest building in existence, far surpassing the Washington Monument, a 169-m-high
obelisk. In addition, it had the advantage of having a very light, almost transparent
structure, and therefore greatly reducing the presence of potential disturbing mate-
rial under the devices used for measuring air ionization. The measurements were
repeated by Wulf on different days and times, obtaining different values, measured
both at the top of the Eiffel Tower and on the ground (Table 1.2; Fig. 1.6).
As can be seen from this table, the values obtained by Wulf at the top of the
Eiffel Tower are slightly lower (about 15% less) than those obtained on the ground,
on average 15.7 ions/(cm3 s) at the top against a value of 17.9 (in the same units)
on the ground. Although with large fluctuations from one measurement to another,
the uncertainty associated with the average value of the set of four measurements
conducted at the top of the Eiffel Tower, about 0.3 ions/(cm3 s), is small enough to
consider as significantly different the values obtained in the two conditions.
The main problem, however, is that the observed reduction is not compatible
with what would be expected based on the absorption of gammas if they came from
radioactive substances present in the soil. Such a calculation, depending on the height
above the ground, is discussed in Appendix A, assuming the values of the absorption
coefficients for air and rock shown in Table 1.1.
Figure 1.7 shows the average values obtained by Wulf on the ground and at the
top of the Eiffel Tower, together with the trend of the expected values, as a function
of the distance from the ground, according to what can be expected if the cause of
this ionization at a given height was to be attributed to the gamma radiation emitted
Table 1.2 Values of the air ionization rate, obtained by Wulf during the measurements conducted
on the Eiffel Tower [Wulf1910]
Date Location Ionization rate (ions cm−3 s−1 )
29.03.1910 Ground 17.5
30.03.1910 300 m 16.2
31.03.1910 300 m 14.4
01.04.1910 300 m 15.0
02.04.1910 300 m 17.2
03.04.1910 Ground 18.3
Average Ground 17.9
Average 300 m 15.7
Fig. 1.6 Theodore Wulf (1868–1946)
by radioactive nuclei uniformly distributed in the ground, up to a depth of one meter

and within a distance of 1000 m from the Eiffel Tower. Some further details of
the calculation are discussed in Appendix A. The curve has been normalized to the
mean of the values measured at the top. Compared to this value, as can be seen, the
ionization of the air should decrease with the height above the ground, at least by a
factor of 10 for a height equal to that of the Eiffel Tower, while the experimentally
obtained value is only 15% smaller.
It is clear, at least to us who read these data in the light of current knowledge, that
they indicate a further cause of ionization, not related to the radioactivity of the soil,
and capable of compensating for the attenuation due to the distance from the ground.
These data, however, were not sufficient to prove the existence of an extraterrestrial
radiation, and the discrepancies were generally attributed to the influence of causes
still attributable to radioactivity of terrestrial origin (residues of radioactive material
in the devices, influence of radioactivity present in the material of construction of
the Eiffel Tower itself, …).
How to move further away from the ground and verify if something different really
happened at great heights above the Earth’s surface? In fact, at the beginning of the
1900s, several possibilities were already available to courageous experimenters.
Fig. 1.7 Average values of the ionization rate measured by Wulf on the ground and at the top of the
Eiffel Tower (blue triangles). The solid red line shows the expected trend based on the contribution
generated by gamma radioactive isotopes dispersed in the soil surrounding the Eiffel Tower (within
a radius of 1000 m) and up to a depth of one meter underground. Further details of the calculation
are given in Appendix A
The first hot-air balloons capable of carrying people on board date back to 1783,
over a century earlier, and from the very beginning they had also allowed scien-
tific explorations. A hot air balloon, or “Mongolfiera” in Italian (named after the
two French brothers Joseph-Michel and Jacques-Étienne Montgolfier) consists of a
balloon made of suitable fabric, with a hole in the lower part, and whose air contained
inside can be heated, allowing an upward thrust capable of carrying the balloon even
to great heights. Modern hot air balloons have even reached higher altitudes than
commercial flights. The hot air can be produced either at the beginning, by heating
the air inside the balloon with a source that remains on the ground, or by carrying
with it the necessary heat source (a burner), which allows for further maneuvers to
climb in quote (Fig. 1.8).
Unlike hot-air balloons, gas balloons are filled with a gas with a lower density than
air (Hydrogen or Helium) but at the same temperature. They must be hermetically
sealed, so as not to let the gas escape spontaneously. To climb to a higher altitude, it
is necessary to lose ballast (sand), while to go back down it is possible to control the
escape of a certain quantity of gas. They too, like hot air balloons, have a long history,
and were available for over a century at the time of the first cosmic ray investigations
[Pfotzer1972, Ziegler1989].
We also recall that in 1900 the first dirigibles were also designed, by the German
von Zeppelin, huge oblong-shaped balloons, filled with gas and capable of carrying a
large number of people even for civil flights, although this possibility was exploited
only for a few years, also due to the numerous accidents. What about the first planes?
After the first flight by the Wright brothers in 1903, which lasted just a dozen seconds,
Fig. 1.8 Image of a hot air balloon (circa 1850–1860). Source Wikimedia Commons, Metropolitan
Museum of Arts Collection, Creative Commons CC0 1.0 Universal Public Domain Dedication
progress had come rapidly, and by 1910 several flights had reached considerable
altitudes, remaining in the air for hours.
In the early 1900s, therefore, there was no lack of means to transport measuring
devices at large altitudes, an idea that in fact was pursued by numerous experimenters,
who, among the various means of ascension, chose aerostatic balloons. Victor Hess’s
balloon ascents, remembered at the beginning, were therefore not a total novelty. The
first measurements of the conductivity of air at high altitudes, aboard balloons, were
made by Franz Linke (1878–1944), a German geophysicist and meteorologist, in
1902–1903, although in the context of investigations on the electric field generated
by the Earth [Linke1904]. Linke had used a gold leaf electroscope up to an altitude of
5500 m, observing that around 1000 m of height the charge loss of the electrometer
Fig. 1.9 Albert Gockel (1860–1927)
was less than that observed on the ground, between 1000 and 3000 m it became
similar to this, and that finally, above 3000 m, became greater, up to a factor of 4
higher, at the maximum altitude reached of 5500 m, compared to what was observed
on the ground. These results are strikingly similar to those observed by Hess a few
years later. However, even though the activity had been conducted in collaboration
with other physicists involved in such activities, such as Gockel (Fig. 1.9), it appears
that they did not receive any form of recognition.
In 1908 another balloon flight was made by Karl Bergwitz (1875–1958)
(Fig. 1.10), up to a height of 1300 m [Bergwitz1910], during which, even if in
a much more limited interval, the same trend was noted, a decrease followed by a
subsequent increase in ionization. Other balloon flights were carried out by Hermann
Ebert between June 1900 and January 1901, up to a height of 3770 m, with unclear
results, due to weather conditions [Ebert1901].
Subsequently, between 1909 and 1911, other flights aboard balloons were carried
out by Gockel himself, at different altitudes, up to a maximum altitude of 4500 m:
once again a qualitatively similar result was observed, a decrease with altitude, but
Fig. 1.10 Karl Bergwitz (1875–1958)
much less than expected [Gockel1910, Gockel1911]. A detailed analysis of the results
obtained by Gockel during these flights has been discussed by Lacki [Lacki2014].
An interesting detail, reported by Lacki, concerns the fact that in one of the
flights Gockel could have used hydrogen, instead of the usual illuminating gas. The
more expensive hydrogen was promised to Gockel due to the planned dismantling
of an airship, and would have allowed, as the lightest gas, to reach an altitude of
7000 m. Unfortunately, this event did not occur, and Gockel was able to carry out
measurements only at moderate altitudes, for which the effect of the increase in
ionization was more difficult to observe. One can only speculate on what the course
of events would have been if this flight at a height of 7000 m had already been carried
out by Gockel a couple of years before the ascents by Victor Hess, but that was not
the case.
1.4 Victor Hess and the Evidence for an Extraterrestrial Radiation 17
1.4 Victor Hess and the Evidence for an Extraterrestrial

Radiation
When Victor Hess in turn planned to make balloon ascents for a quantitative study
of the ionization of air at high altitudes and to estimate the absorption of gamma
rays from the ground into the air, the problem had therefore already been addressed
by several other authors, as seen above. Studies of the absorption of gammas in
various materials, including air, had already been carried out and a numerous series of
measurements with electroscopes, even at high altitudes, with balloons, had already
been performed. However, even if many considerations about the existence of an
extraterrestrial radiation, as a possible interpretation of the data obtained, had been
proposed by several authors, the quality of the data collected and the quantitative
information extracted from these data were not sufficient to draw definite conclusions,
and above all they were unable to give a quantitative estimate of the contribution due
to this hypothetical source of ionization linked to the atmosphere. Indeed, according
to Millikan’s perhaps too drastic comment many years later, “Until 1910 there was no
data about the penetrating radiation coming to the Earth from the outside. I cannot
find any information on existence though any idea, even distantly concerning the
phenomenon which we connect now with the term cosmic rays” [Millikan1935a].
Victor Franz Hess, born in Austria, had obtained his doctorate in Physics from
the University of Graz in 1910. He had therefore received his doctorate just a year
before when he made his first two balloon flights, in 1911, to study absorption in air
of the gammas coming from the radioactivity of the terrestrial soil. Hess was aware
of Wulf’s results and Eve’s estimates of gamma absorption coefficients and was
therefore aware of the incompatibility of the two results. His first measurements,
carried out on the ground with a high intensity gamma source (2.6 1010 Bq) at
distances between 10 and 90 m, were therefore aimed at a further study of the
gamma absorption coefficients, using sources of known intensity, and had given
results compatible with Eve’s estimates, confirming the discrepancy observed in
Wulf’s measurements, and suggesting further measures to be taken aboard balloons
to clarify the issue [Hess1911].
The first two flights, carried out in 1911, confirmed that the ionization present at
still relatively low altitudes was comparable to that measured on the ground and that
there was apparently no difference between what was measured during the day and
during the night.
The next program also included flights at higher altitudes, with some improve-
ments to the electroscopes to be brought on board (in particular to keep the pressure
constant inside them). The idea was to maintain the same altitude for a certain dura-
tion of the flight, to verify the influence of other factors, at the same altitude with
respect to the ground, and observe any variations during measurements of a certain
duration. One of the three instruments was also equipped with much thinner walls,
to be sensitive to electrons. The program also included some measurements during
Table 1.3 Summary of balloon flights performed by Hess in 1912 [Hess1912]

Date Max height (m) Notes
17.04.1912 2750 Measurements made also during a partial solar eclipse, no
effect observed. Slight rise around 2000 m
26–27.04.1912 2100 Measurements carried out also during the night; no
appreciable difference observed compared to the day. Values
reported up to an altitude of 1800 m
20–21.05.1912 1200 Night flight, values reported up to an altitude of 500 m
03–04.06.1912 1900 Short duration night flight due to weather conditions. Values
reported up to an altitude of 1100 m
19.06.1912 950 Solo flight, only one instrument used
28–29.06.1912 360 Low altitude flight at night, with long duration
measurements at the same altitude. Slight decrease observed
relative to the ground
07.08.1912 5350 High-altitude flight, with hydrogen balloon. Detailed results
reported in the text and in Table 1.4
the night, to get further information about the possible role of the Sun. It was there-
fore a consistent scientific program, carefully planned, to give possibly quantitative
answers to different aspects of the problem.
This program was carried out over the course of seven different flights, between
April and August 1912 (Table 1.3), supported by the Austrian Academy of Sciences,
the first six departing from Vienna. A detailed description of the original article
[Hess1912], through its English translation and related commentary, has been made
available by De Angelis and Schultz [DeAngelis2018]. The scientific figure of Victor
Hess was outlined on the occasion of the centenary of this event in [Schuster2014].
We may see in one of the historical archive photos the scene of one of these flights
(Fig. 1.11). A small crowd, mostly children—we count about fifteen—look partly at
the photographer, partly at the rectangular basket, in which two people are already
on board. Two of the spectators are caught by the photographer’s lens while they
are observing in amazement the huge balloon that rises above their heads. Around
the basket, weight bags marked with “Aero Club Wien” are hooked to the edges, to
be used in case of need. The “light” clothing worn to climb to an altitude of several
thousand meters, a normal jacket with a waistcoat and a light cap, more typical of a
short hike in the hills or a picnic in the countryside, amazes in these photos.
Following the observations obtained during the first six flights of 2012, it was
confirmed that the ionization rate was independent of the presence of the Sun:
measurements carried out during the night and, during the first flight, even on the
occasion of a solar eclipse, substantially gave the same result. As regards the depen-
dence on altitude, the results obtained down to modest altitudes confirmed only a
slight decrease in ionization with altitude, in accordance with what was found qualita-
tively by other authors (Linke, Bergwitz, Gockel, Wulf), and therefore suggested, as
already hypothesized, that there should be an additional source of ionizing radiation
beyond that linked to the emission of gamma rays from the Earth’s soil.
Fig. 1.11 Victor F. Hess, in the center, on the occasion of one of the flights made on board a balloon
Except for the last flight, all the others used illuminating gas to fill the balloons,
which did not allow reaching altitudes beyond a certain limit. The last flight, on the
contrary, used hydrogen, and reached a remarkable altitude of over 5000 m, with
results that opened a new chapter in the history of cosmic ray physics.
The ascent began on the morning of 7 August, at 6:12, from the locality of Aussig
(Ústí nad Labem), with the presence, together with Hess, also of the pilot (W. Hoffory)
and a meteorologist (E. Wolf), and had a relatively short duration: already around
noon the crew had landed about fifty km from Berlin, after reaching the maximum
altitude of 5350 m, and having traveled about 200 km. In the first part of the flight, up
to an altitude of 1000 m, a slight decrease in ionization was observed, in accordance
with what was already known, while at higher altitudes the values began to increase,
reaching and exceeding the values observed on the ground. At altitudes between
3000 and 4000 m the values were significantly higher, between 20 and 40% higher
than those obtained on the ground. But it was above 4000 m that the measurements
with the two gamma-sensitive devices reached values that were clearly 100% higher
than those observed on the ground.
Table 1.4 shows a summary of the measurements obtained during this flight. The
third device, the one equipped with thin walls, sensitive to electrons, unlike the other
Table 1.4 Summary of the average values (expressed in number of ions per cm3 per second)
obtained from measurements made during the August 1912 flight from Hess, up to an altitude of
5350 m [Hess1912]
Altitude Average for instrument Average for instrument Average for Instrument
#1 #2 #3 (normalized value)
0 16.3 19.6 11.8
0–200 m 15.4 11.1 19.1
200–500 m 15.5 10.4 18.8
500–1000 m 15.6 10.3 20.8
1000–2000 m 15.9 12.1 22.2
2000–3000 m 17.3 13.3 31.2
3000–4000 m 19.8 16.5 35.2
4000–5200 m 34.4 27.2 –
two, was not pressure-tight; the air pressure, therefore, was equivalent to the external
one and, especially at higher altitudes, the measured values needed to be renormalized
for the estimated pressure difference. Each of the averages derives from a more or
less high number of different readings, which however were taken as the balloon
moved, during the ascending or descending phase. For this reason, each of these
values refers to an altitude range, during which multiple readings were taken.
Figure 1.12 shows the average values measured by the three devices as a function
of the height with respect to the ground, reported numerically in Table 1.4. The
devices 1 and 2 (values shown with the blue and red symbols) were more shielded,
therefore not very sensitive to electrons, while the device 3 (green symbols) was
equipped with thin walls, which electrons could traverse. The results show for all
the instruments used a slight decrease up to an altitude of around 1000 m, then a
rise that becomes more and more consistent with the height. The horizontal bar for
each measurement indicates the interval within which the measurements were made
obtaining an average. As for the uncertainty in the measured ionization values, Hess
discusses some estimates for the different instruments, reporting values between 5
and 10%.
Since the different instruments, having different calibrations, provided different
values measured on the ground, it is possible to report the ionization rate measured
by each instrument with respect to the value measured on the ground (Fig. 1.13).
This trend is very similar for instruments 1 and 2, of the same type, while it shows a
greater increase with the altitude for instrument 3, sensitive to electrons.
Between 4000 and 5000 m, the measured ionization was more than double that
measured on the ground. The conclusions of Hess at the end of this investigation
[Hess1912, Hess1913] report the evidence of the existence of a highly penetrating
radiation, present in the upper part of the atmosphere and whose effects are however
visible down to the ground.
Regarding the origin of this radiation, Hess concludes that it cannot be associated
with the Sun, at least if it is assumed that it consists of gamma rays that propagate in a
Fig. 1.12 Evolution of the ionization rate measured at different distances from the ground, during
the balloon flight performed by Hess on August 7, 1912, up to a maximum altitude of 5350 m.
The symbols in blue and in red show the values measured by devices 1 and 2 (shielded), while the
symbols in green show the values obtained by device 3 (also sensitive to electrons). The values
were taken from the original article by Hess [Hess1912] and the horizontal bars show the altitude
ranges within which the different measurements were made. Device 3 did not take data at the highest
altitudes
Fig. 1.13 Values of the ionization rate, normalized to those measured on the ground, for each of
the three devices used by Hess during the August 1912 flight, as a function of the distance from
the ground. The different colours, relating to the three measuring devices, correspond to those also
used in the previous figure
straight line. The fact that this radiation was made up of gammas was, moreover, the
most plausible assumption, since of the three radiations known at that time (alpha,
beta and gamma), gammas were those with a greater penetrating power.
1.5 Towards a Confirmation of Hess’s Results
Contrary to what one might believe, the interpretation of Hess’s results about the
notable increase in air ionization with altitude did not immediately receive general
agreement. Many physicists, perhaps most of them, continued to doubt that the cause
of this ionization was due to a highly penetrating radiation of extraterrestrial origin.
Indeed, the same experimental results were the subject of debate and required, in the
opinion of most, further independent confirmation. This attitude, which might seem
unusual, represents on the contrary what is most common in the way science usually
proceeds. As the historian of science Thomas Kuhn [Kuhn1962] has pointed out in
describing the structure of scientific revolutions, science is generally characterized
by periods of “normal science”, in which a paradigm is able to guide the research,
allowing to improve the paradigm itself, to proceed to more accurate measurements,
to look at applications. It may happen that during this work, however, some anomalies
are noticed, aspects that do not completely agree with the current paradigm, and that
require further reflection. As Kuhn points out, the existence of such anomalies does
not in itself produce the abandonment of the existing paradigm, without which even
the normal scientific practice would not be possible. On the contrary, scientists try,
as far as possible, to bring the anomalies within the current paradigm, through small
corrections, generalizations, extensions, which do not distort their nature. Only after
an incessant work of this kind that has not given results, and especially after new
hypotheses have arisen and have had the opportunity to be verified in detail, will the
scientific environment be ready to abandon the old paradigm in favor of the new one,
which in the meantime has been better clarified and defined. Kuhn calls this process
a scientific revolution.
In some ways, we can look at the events related to the difficulty of reconciling
the observations on the ionization of air at high altitudes with terrestrial radioactivity
as a change in the current paradigm, a real revolution. As such, the anomalies about
the ionization values, observed in the first measurements, gave rise to attempts at
interpretations that ultimately did not upset the dominant vision, in terms of a cause
linked to radioactivity, of the soil or air. This even after the results obtained by Hess.
It was difficult to accept that the cause was something still unknown, and it seemed
more reasonable to continue to believe that the radioactivity of natural substances,
in various forms, was still the cause. This is how various considerations can be
interpreted about the possibility that radioactive substances could be concentrated in
the upper part of the atmosphere, or that the cause could be attributed to the elec-
trical discharges produced during thunderstorms. Moreover, the possible influence
of weather conditions had also been considered, and excluded, by Hess.
1.5 Towards a Confirmation of Hess’s Results 23
One of the physicists initially unconvinced not only of the interpretation but of the
validity of the very same results provided by Hess was the German Werner Kolhörster
(Fig. 1.14). At the time of Hess’s achievements, in 1911–1912, Kolhörster was also
very young, having just received his doctorate. Attracted by the studies of Wulf
and Hess, he also had contributed to improving electroscopes for measurements
to be conducted at high altitudes, in particular by making them suitable for with-
standing differences in pressure between inside and outside up to 500 Torr, therefore
capable of carry out measurements at constant pressure up to an altitude of 9000 m
[Kolhörster1913a]. The devices also had greater sensitivity, so much so that they
could make much more frequent measurements than those conducted by Hess, and
therefore more correlated to the actual altitude at the time of the measurement. In the
summer of 1913 Kolhörster made three balloon flights, reaching approximate alti-
tudes of 3500, 4000 and 6200 m [Kolhörster1913b]. During this last flight, Hess’s
results were confirmed, with values measured even at higher altitudes. Further flights,
carried out in 1914, then allowed to reach the maximum altitude of 9300 m, at which
the ionization was enormously greater [Kolhörster1914]. The results obtained by
Kolhörster are shown in Fig. 1.15, after subtracting the measured value at ground
level. They therefore represent the differences with respect to what is observed at
ground level.
No doubt remained at this point about the cause of the ionization, which could not
originate from terrestrial radioactivity, and which instead showed clear evidence for
a radiation coming from above. Assuming that this too was a penetrating radiation,
similar to the gamma radiation emitted by the radioactive substances present on Earth,
and that it was attenuated according to an exponential trend, Kolhörster was able
to estimate, from the ionization values measured at various altitudes, an absorption
coefficient, in a similar way to what had been done for the absorption of the radiations
emitted from the ground. The value derived from these measurements indicated a
much lower absorption coefficient, approximately one tenth, compared to that of the
known gamma radiation. It was therefore something with a much greater penetrating
power than that associated with the known radiation (Fig. 1.15).
Although many years later, illustrious physicists like Millikan continued to ques-
tion the existence of this radiation, to change however their mind shortly afterwards,
a new phase had now entered, in which, having established the existence of this radi-
ation, it was now a question of characterizing it, understanding its nature, organizing
and obtaining quantitative measures. A work that would engage a large part of the
physics community in the following decades.
In September 1913 a scientific congress, with different sections and the partic-
ipation of 7000 scientists, held in Vienna, also saw reports expressly dedicated to
the debate about the origin of the radiation observed in the atmosphere. During
the conference some of the scientists, including Wulf, did not recognize in the data
obtained a reason to admit radiations of extraterrestrial origin, while others, including
Kolhörster and Gockel confirmed Hess’s hypotheses about a radiation coming from
outer space, an opinion that from that moment on could be said to be accepted by
most. From the measurements carried out at different altitudes it was also possible
to estimate that at sea level the contribution to ionization due to radiation originating
Fig. 1.14 Werner Kolhörster (1887–1946)
Fig. 1.15 Difference between the ionization rate measured at different altitudes, compared to that
measured on the ground, based on measurements made by Kolhörster in 1914 [Kolhörster1914]
from outside the Earth was approximately 20%, while the rest was due to natural
radioactivity, which explains the initial difficulties in recognizing the origin of this
radiation source.
The period immediately following the Kolhörster measurements marked a reduc-
tion in research activities, due to the war events of the First World War, with a strong
downsizing of the budget allocated to this activity and a smaller number of publica-
tions of new scientific results. Sporadic activities were continued by Hess, Kolhörster
1.5 Towards a Confirmation of Hess’s Results 25
and others, both with measurements performed in the high mountains and aboard
balloons, as we will see in the next chapter, but without significant improvements, so
that a few researchers, including Wilson himself, continued to verify the possibility
that the ionization of the air, even at high altitudes, was due to other causes.
Chapter 2
Confirmation of the Existence
of a Cosmic Radiation
Abstract Although the working conditions were not optimal, after the First World
War some of the pioneers contributing to the discovery of the cosmic radiation still had
the possibility to go on in Europe with their research on cosmic rays. However, in the
1920s, investigations concerning cosmic ray physics progressively moved towards
the United States. The chapter discusses the important steps of this transition and the
leading role of Robert Millikan to the field, with his new measurements by unmanned
balloons and with the investigation of the radiation intensity under the water of lakes
at various depths. Quantitative data and plots from his original papers are reported
and discussed, with further analysis of the absorption effects in water reported in
Appendix B. The controversial contribution of the American physicist and his first
ideas of the nature of the cosmic radiation are also discussed in connection with
the debate with the original findings by the European researchers, till the general
acceptance of the idea of a radiation originating from the outer space.
2.1 Further Investigations in Europe During and After

the First World War
Despite the difficulties imposed by the world conflict in those years, it cannot be
said that research activities were completely absent or suspended at that time. Some
European nations, in particular Germany, Austria and Switzerland, remained at the
center of further studies concerning the phenomena related to cosmic radiation,
as we will call it from this point on, even if the terminology associated with this
phenomenon represents a long and complex story in itself.
The interest in carrying out further measures of air ionization moved in those
years towards the possibility of organizing experiments on the mountains, at high
altitudes, where, although the contribution due to the radioactivity of the soil was not
negligible, as in the case of high-altitude balloon flights, the relative ease of carrying
out systematic and long-lasting measurements allowed a significant benefit for the
quantitative understanding of the phenomenon.

https://doi.org/10.1007/978-3-031-24762-0_2
28 2 Confirmation of the Existence of a Cosmic Radiation
Which were the roles and contributions from the pioneers of the study of cosmic
radiation at that time? Did they have the opportunity to continue working in this field
and to contribute further results to this study?
Wulf, who was a Jesuit priest from the age of 20, before studying physics with
Nernst in Gottingen, continued to teach physics at the Jesuit College in Valkenburg,
Holland, continuously from 1904 until 1935, except for the parenthesis of the First
World War, and was the author of several educational textbooks in physics, but it
does not seem to have been further interested in research related to cosmic radiation.
Gockel, after the flights carried out up to an altitude of 4500 m, and after having
missed the possibility of a high-altitude flight with a balloon filled with hydrogen,
had no further chance to carry out other balloon flights, also due to the conditions
of his work at the University of Freiburg, where, despite having been appointed
professor, he did not have a lot of funds available. However, he remained active in
this field, contributing to study the effects due to this radiation in various locations
in Switzerland, on the Jungfraujoch (3400 m) and on the Aletsch glacier (2800 m)
[Gockel1915], and also performing measurements under the ice, to evaluate the
absorption of radiation. Despite his direct involvement in a specific activity of char-
acterization of the cosmic radiation, Gockel remained largely skeptical about the
extraterrestrial nature of this radiation, and in 1923, a few years before his disap-
pearance, he still wondered whether the phenomenon of the ionization of air in the
Earth’s atmosphere had to do with a radiation originating from outside the Earth
[Gockel1923].
During the years of the conflict, Kolhörster had served mainly in a meteorolog-
ical unit, and was subsequently sent to Turkey as a member of the German military
mission. He also continued some research activities on atmospheric electricity. At
the end of the war, for a few years, he also had to teach at school, and was only
able to resume research activities related to the study of cosmic radiation in 1922,
carrying out various measurements in the mountains of Switzerland. Subsequently,
from 1928, Kolhörster, together with Walter Bothe (1891–1957), who had become
director of the Physikalische-Technische Reichsanstalt (PTR) in Berlin, and whom
we will discuss in more detail later, began to employ the first Geiger-Müller counters,
even in coincidence, for the detection of the cosmic radiation, in order to study its
nature, whether it consisted of highly penetrating gammas or particles. Kolhörster,
therefore, except for a brief parenthesis during the years of the conflict and in the
immediately following years, always remained active in research related to the study
of cosmic radiation, contributing in various ways to the understanding of its prop-
erties. Indeed, his activity was highly appreciated especially in the German context
until the end of the Second World War, as we will see in the following sections.
What about Pacini, whose intuitions seemed to have preceded the hypothesis of
a radiation of extraterrestrial origin? Unfortunately, for various reasons, analysed
in detail by De Angelis [DeAngelis2010, DeAngelis2012], Pacini did not have the
opportunity to continue this activity with significant roles in the following period.
After graduating from university, he got a teaching position at the University of
Rome and, subsequently, in 1928, he obtained the professorship at the University of
Bari. Although he remained active in the field of experimental physics, his works
2.1 Further Investigations in Europe During and After the First World War 29
after 1913, published in Italian, were mainly devoted to spectrophotometric studies

of sky light and other properties of the atmosphere. In 1913 Pacini did not have the
opportunity to participate in the Conference where he could have known the latest
results on the observations regarding the causes of air ionization and contribute to the
debate. To the period immediately following the end of the conflict (1920) dates back
the exchange of correspondence between Pacini and Hess, in which Pacini expresses
his disappointment for the fact that his works on the absorption of radiation, carried
out on the surface of the sea and under water, had not been mentioned. Unfortunately,
Pacini’s contribution was unable to be adequately exploited even in the following
years.
In 1913 Hess had considered as an important task the possibility to carry out
simultaneous measurements in different locations, in order to observe any correlated
variation in the intensity of the radiation. Some of these measurements were actually
organized, with the collaboration of several colleagues, by carrying out true “mea-
surement working shifts” of 12 h every day [Schuster2014], perhaps the precursors
of the current measurement shifts for long-term physics experiments, such as those
at particle accelerators or astronomical observatories.
Measurements lasting about one year, in continuous mode, were also organized by
Hess and Kofler in the Alps, between 1913 and 1914, at an altitude of about 2000 m,
with electroscopes positioned outdoor, in a container with wooden walls. Being able
to measure ionization over such a long period of time, and at repeated intervals
throughout each day, makes this experiment perhaps the first long-duration cosmic
ray flux monitoring activity. The results made it possible to evaluate any correlation
between the observed values and various environmental parameters, such as temper-
ature, pressure, time along the day [Hess1917]. For example, it was established that
there was no appreciable difference between the values observed during the day and
the night, and that the same daily variations were much less than those observed at
sea level.
In the years following his balloon flights, Hess mainly devoted himself to activities
related to natural radioactivity, also evaluating the number of alpha disintegrations
per second produced by a gram of radium (3.7 × 1010 ), which allowed to establish
a unit of measure of the activity of a sample, the Curie, exactly equal to the activity
of one gram of radio in radioactive equilibrium, a unit still in use today. Between
1915 and 1917 he also carried out research activities related to the use of X-rays in
the medical field. In 1920 Hess had become a professor at the University of Graz,
but soon after he moved to the USA for a period of two years, where he still made
contributions related to radioactivity and its applications.
Only a few years later, in 1923, did he return to Graz, resuming his research
on cosmic rays. Despite the experimental results achieved through balloon flights
at higher altitudes, both by Hess himself and by Kolhörster, the acceptance of the
extraterrestrial origin of this radiation still posed some problems. As we will see in the
next section, a great deal of opposition initially came from American circles, linked
to the figure of Millikan, and the same priority of the evidence of this radiation was
also called into question in the context of European physics, due to the contribution
of Kolhörster. In the second half of the 1920s, Hess continued to contribute to the
Fig. 2.1 An apparatus for measuring cosmic radiation (ionization chamber), installed in the labora-
tory for high-altitude cosmic rays on Mount Hafelekar, 1931. Source Wikimedia Commons, Creative
Commons Attribution 3.0 Unported License
study of the cosmic radiation, through quantitative investigations of the variations

observed in the mountains, at high altitudes, up to the creation in 1931 of a perma-
nent observation station at an altitude of 2300 m, on Mount Hafelekar [Hess1932]
(Fig. 2.1).
Among the pioneers who had contributed to the first results on cosmic radiation,
the names of Hess and Kolhörster are certainly the ones that had a greater importance
and continuity even in the following decades. However, starting from the early 1920s,
the study of these phenomena also began in the United States, where the figure of
Robert Millikan initially played a crucial role, with alternating events, in the debate
about the existence and nature of this radiation.
2.2 Robert Millikan and the First American Contributions to Cosmic Ray … 31
2.2 Robert Millikan and the First American Contributions

to Cosmic Ray Physics
In the years in which the first high-altitude balloon ascents for the study of air
ionization were made in Europe (1912–1913), by researchers less than 30 years old,
the name of Robert Millikan was already well known in science in the United States.
Millikan, at that time nearly fifty years old, had already made his experiments on
the electron charge, which ten years later, in 1923, would have given him the Nobel
Prize, and held numerous positions of responsibility in the academic environment
as well as organizational and managerial roles related to science. In the early 1920s
he had moved from Chicago to California, where he would later establish and direct
Caltech (California Institute of Technology).
Millikan’s personality was complex, and together with his great scientific and
organizational skills, there are some aspects of his figure, his character and above all
his scientific activity that have created many problems to his biographers. One of the
last steps in this area is for example the recent decision (2021) by the Caltech Manage-
ment to remove the name of Millikan, along with that of numerous other colleagues,
from the dedications that auditoriums, buildings, or other structures had received in
the past. For what reason? For the positions that Millikan had taken at that time,
belonging to the eugenics movement through the Human Betterment Foundation
(HBF). The Caltech Board of Trustees assessed that “In making its recommenda-
tions regarding Millikan, the committee also considered his stances on gender, race,
and ethnicity, finding them sexist, racist, xenophobic, and inexcusable by any stan-
dard” [CALTECH]. One may wonder if it is correct to notice these problems almost
a century later, but probably everything is part of the renewed attention that has been
given to these issues in recent years.
Apart from this paradoxical episode, the events concerning the role of Millikan,
and more generally, of American physicists, on the activities concerning the study
of cosmic radiation have always been the subject of attention. Millikan’s attitude
towards European activities concerning the study of cosmic radiation was in fact
always critical, at first denying the validity and importance of these studies and the
conclusions that gradually emerged about the existence of a source of radiation of
extraterrestrial nature, subsequently, when he realized, through his own experiments,
the real existence of this radiation, trying to minimize the European contributions in
favour of the American ones. Still in 1935, that is many years after the acceptance
of the ideas concerning cosmic radiation, Millikan wrote [Millikan1935a] that there
was no trace until 1910 of any idea even remotely linked to what could be called
“cosmic rays” [Xu1987].
Millikan, therefore, in the early 1920s, began a series of experiments in the United
States related to the study of this radiation. The first measurements were carried out
in March and April 1922 by Millikan and Bowen, by means of balloon flights without
human intervention, up to a height of about 15 km. These results were communicated
at a meeting held in May 1923 [Millikan1923]. Further measurements carried out
on board airplanes, balloons and in the high mountains, were communicated during
the same meeting by his collaborators [Otis1923a, Otis1923b]. However, the full
results of this activity were only published in 1926, in a series of three articles
which constituted an important step in the study of cosmic radiation [Millikan1926a,
Millikan1926b, Millikan1926c].
In the first of these articles [Millikan1926a], Millikan refers in the introduction
to the existing situation, quoting the works of Gockel, Hess and Kolhörster (but not
those of Pacini, which perhaps were not even known to him). The results of Hess
and Kolhörster are expressly reported, and it is explicitly admitted that these results
had immediately raised curiosity and interest in further measures to be carried out
personally. The situation originating from the world conflict, although America was
geographically far from the theatre of the battles, and the decision to build automatic
recording devices, which avoided the physical presence of people on board the large
balloons, had delayed this activity until the years 1921–1922. At that time Millikan
had succeeded in having four small electroscopes built, capable of recording, by
means of photographic films, the indications regarding the divergence of the fibers of
the electroscope as well as the pressure and temperature. A more detailed description
of these tools is given in Chap. 5.
The qualitative result obtained from these flights was that indeed, at high altitudes,
the ionization was greater than that measured on the ground, in qualitative agreement
with what European researchers had already found. However, from a quantitative
point of view, Millikan found no congruence with the trend measured by Hess and
especially by Kolhörster. The estimate of the value obtained on average at heights
above 5 km was much lower than that obtained by Kolhörster. For the balloon that
had reached an altitude of 15.5 km, the value obtained above 5 km was about 3 times
that obtained on the ground, while Kolhörster had already reported an increase of a
factor of 7 at an altitude of 9 km. The extrapolation to altitudes of 15 km, starting
from the data obtained by Kolhörster, gave values enormously higher than those
measured by Millikan. According to Millikan, the interpretation of these measures
seems to nullify the possibility of radiation of extraterrestrial origin: “The results …
constitute definite proof that there exists no radiation of cosmic origin having such
characteristics as we had assumed” [Millikan1926a].
But in the final sentences of the same article, Millikan anticipates that new results,
which will be published in subsequent articles (evidently partly ready), will present
unambiguous evidence of cosmic radiation with extreme penetrating power. This
first article, although it discusses the results obtained in the course of measurements
made a few years earlier, appears to have been sent with the date of December 24,
1925, and published in April 1926.
In the following article, signed by Millikan and Otis, sent on March 1, 1926 and
published in June of the same year [Millikan1926b], the results of a new series of
observations, also made in the years 1922–1923, are discussed, both with balloons
and in the high mountains or on board of airplanes, in different locations. In partic-
ular, some flights by plane, up to altitudes of about 5000 m, made it possible to
measure values once again in qualitative, but not quantitative, agreement with what
was obtained in Europe, i.e., a decrease in ionization up to moderate altitudes, around
2500 m, then a slight increase with altitude—but with absolute values well below
those reported by Hess and Kolhörster. Regarding the possible dependence of the
intensity of this radiation on the particular time along the day, a further series of
measurements were made in the high mountains, at Mt. Whitney (4130 m) and at
Pikes Peak (4300 m), showing that this intensity was constant within the limits of
observations, during the whole 24-h period and which did not depend on the position
of the celestial bodies or on the Milky Way itself.
Further detailed measurements were made with different types of lead shielding
around the electroscopes, the results of which allowed to understand that this radiation
did not come from any privileged geographic direction (North, South). Beyond the
details, the main conclusions are once again that the causes of air ionization in the
various conditions are mainly attributable to local causes: “Both lines of evidence
presented above point, then, to the conclusion that there is on mountain peaks a
copious radiation of local origin and of a hardness not greater than that of the gamma
rays of radium or thorium, but they reveal thus far no penetrating radiation of cosmic
origin. If such a radiation exists at all it can at the most produce but a small part of
the ionization observed in electroscopes on mountain peaks” [Millikan1926b].
Nothing new, therefore, about the conclusions on the origin of this radiation, not
even in this second work? In fact, the details of the measurements reported are valu-
able elements that add to the knowledge existing up to that moment. But there is
another aspect, which Millikan expressly mentions. Many considerations made in
this field, and the same discrepancy with the values reported by other authors are
based on the assumption of a gamma radiation with a given penetrating power, and
Millikan is aware of the fact that assuming that cosmic radiation exists, then it must
be less intense on the ground than previously assumed, or much more penetrating
than previously assumed. These statements therefore leave room for further attempts
to understand the phenomenon, which will come from an additional series of experi-
ments, reported in the third of these articles of 1926, signed by Millikan and Cameron
[Millikan1926c]. The article was sent to the Journal on August 7 and published in
November 1926, and marks a significant step in understanding the radiation and
especially in Millikan’s change of opinion.
In the introductory part of this work Millikan summarizes the situation regarding
the possible existence of this radiation and its penetrating power, defined by the
absorption coefficient, which Millikan believes he has evaluated based on his previous
work through balloon flights. Millikan also discusses the new results obtained by
Kolhörster under the surface of the water (but not, again, Pacini’s results). With
these considerations, Millikan denies that on the basis of what has been observed so
far, including his previous works, the existence of extraterrestrial radiation can be
assumed, unless it is even more penetrating. This justifies a series of further measures,
reported in this article, which should finally shed light on the possible existence of the
radiation itself. The measurements in question concern experiments to be carried out
under the surface of lakes, at different altitudes, in particular in lakes fed by melting
snow, as measures conducted in the past have shown the existence of an appreciable
radioactivity even in water. Millikan is therefore looking for non-radioactive lakes,
located in the high mountains, where cosmic rays, if they exist, have a considerably
higher intensity than at ground.
The measurements were carried out in late summer of 1925 at Lake Muir, about
3400 m above sea level, a large and very deep lake near Mount Whitney, the highest
peak in the United States, excluding Alaska, and subsequently near Lake Arrowhead,
at an altitude of about 1570 m, therefore significantly lower than the first, in order to
compare the results obtained in the two conditions.
In Lake Muir the electroscopes were immersed to a depth of about 20 m, demon-
strating a decrease to a depth of 15 m. Taking into account that the thickness of
atmospheric air above the surface of the lake corresponded to about 7 m of water, this
meant that the radiation measured by the electroscopes, if actually coming from the
outside, would be able to cross the equivalent of 15 + 7 = 22 m of water before being
absorbed. This radiation, always assumed as gamma radiation, was therefore enor-
mously more penetrating than any other known radiation. Millikan, in comparing this
penetration capacity to that of X-rays, used in hospitals, which are arrested approxi-
mately by 1 cm of lead, realizes that the radiation observed underwater would instead
be able to cross almost 2 m of the same material.
For the other observation site, Lake Arrowhead, the equivalent thickness of water,
due to the atmosphere above the lake’s water, differed from the first by about 1.80 m;
it is as if each reading made in Lake Arrowhead therefore corresponds to a reading
made in Lake Muir at a depth 1.80 m greater. The results obtained, reported in Table
2.1 and Fig. 2.2, actually show an agreement with this hypothesis. The top plot in
Fig. 2.2 shows the two data series reported as a function of the depth of water actually
measured in the two cases, while in the bottom graph the values originating from
Lake Arrowhead (located at a lower altitude) have been shifted by the equivalent
amount of water corresponding to the difference in air thickness. As it can be seen,
after this shift the two series of values appear as belonging to a single set of data,
that is, they actually lie on a single curve.
From this comparison Millikan is ready to draw the important conclusion “that
the rays do come in definitely from above, and that their origin is entirely outside
the layer of atmosphere between the levels of the two lakes”. Moreover, Millikan
had already used the term “cosmic” with regard to penetrating radiation in a work
published in Nature in 1925 [Millikan1925].
Measurements in the lakes were also carried out in the following years in the
southern hemisphere, in Bolivia and Peru, exploiting the lakes Miguilla (4570 m)
Table 2.1 Values obtained by Millikan and Cameron with one of the two electroscopes immersed
at a certain depth below the water of lakes Muir and Arrowhead [Millikan1926c]
Depth (m) 0.0 0.45 0.7 1.0 1.1 2.8 3.0 5.0 10 15 20
Average ionization 13.25 9.7 – 7.75 – 5.9 5.45 4.75 4.0 3.6 3.65
(ions/cm3 s), Lake
Muir
Average ionization 7.0 – 5.8 5.5 5.15 – 4.9 4.4 – 3.7 –
(ions/cm3 s), Lake
Arrowhead
Fig. 2.2 Average ionization values obtained by Millikan and Cameron with one of the electroscopes
immersed at a certain depth in the waters of lakes Muir (dot symbols in red) and Arrowhead (square
symbols in blue), taken from [Millikan1926c]. In the graph above, the values are shown as a function
of the real depth in each of the two lakes. In the bottom one, the values obtained in Lake Arrowhead
(located at a lower altitude) have been shifted by an amount equal to the difference in equivalent
altitude in water (see text)
and Titicaca (3820 m), to verify that the results were compatible with those found in
the northern hemisphere, in the United States [Millikan1928a].
To extract the absorption coefficient from these data, Millikan and Cameron use
for the first time the assumption that cosmic radiation comes from every direction and
not only along the vertical direction. Absorption at a given depth therefore depends
on the thickness crossed along each direction (see Appendix B). The analysis of these
curves also shows that the absorption cannot be represented with a single exponential
curve, but rather shows two components: one that is absorbed in a smaller thickness,
the other that requires larger amount of material, with two different absorption coef-
ficients, which are estimated at 0.0030 cm−1 and 0.0018 cm−1 respectively, after
including the thickness of the atmosphere in equivalent meters of water. In other
words, the radiation is not homogeneous but has a certain distribution in energy
(spectral composition), resulting on average more penetrating for the components
that manage to reach a certain depth. The term “soft radiation” is even used to denote
the component that is most easily absorbed.
Regarding the nature of radiation, given that the most natural hypothesis was to
assume that it consisted of gamma, the “soft” component is associated with electrons,
in terms of the recently formulated Compton scattering process [Compton1923].
These electrons have a penetration power in the air, based on their energy, even tens
of meters. We will resume later the question of the origin of cosmic radiation, on
which Millikan will initially support hypotheses that he will be forced once again to
abandon later.
It was this set of measurements, therefore, rather than the measurements carried
out at high altitudes, which convinced Millikan of the real existence of radiation with
an extraterrestrial origin. Indeed, the presentation of the results seems to suggest that
without the latter evidence, obtained by Millikan, all the previous experiments would
be of little significance, and could not be used as unambiguous proof of the existence
of this radiation. In other words, Millikan tends to bring to himself the irrefutable
proof of the existence of cosmic rays.
We recall at this point that Millikan had obtained the Nobel Prize just three years
earlier, in 1923, for the measurement of the electron charge and for the study of the
photoelectric effect. He was undoubtedly, at that time, the most famous physicist
in the United States, so his views had a relevant social weight, especially if they
could support American supremacy in science. His name, about the claims related
to cosmic radiation, was in major newspapers, such as the New York Times, and Time
magazine had dedicated several articles to him, writing that Millikan had announced
the existence of a new radiation, the Millikan rays, capable of traversing a six-foot
Fig. 2.3 Robert Millikan discusses with Marie Curie in Rome, on the occasion of the Conference
(October 1931) dedicated to nuclear physics. Source Wikimedia Commons
2.3 The Acceptance of the Idea of a Cosmic Radiation 37
thickness of lead. The cover of Time of April 25, 1927 portrays him for example while
observing under the microscope, with the caption “Dr. Robert Andrews Millikan…
detected the cosmic pulse”. In the same period also the Science magazine used the
term “Millikan rays”. It must be said that Millikan did not do much to disprove these
claims.
Naturally, Millikan’s statements contained in the last of his works of 1926, and
above all the use that the press made of them, created many problems for European
researchers. Hess in the same year 1926 wrote that “the recent affirmation of the
existence and penetrating power of this radiation (Höhenstrahlung) by Millikan and
collaborators gave the inspiration to journals such as Science and Scientific Monthly
to propose the term “Millikan-rays” for them. Since this is only a confirmation
and an extension of the results of the ionization measurements carried out in the
balloon by Gockel, myself and Kolhörster in the period between 1910 and 1913, this
nomenclature must be rejected as inappropriate and unauthorized” [Hess1926]. And
Kolhörster himself intervened, recalling that the so-called “Millikan rays” are none
other than the “Höhenstrahlung” known for some time [Kolhörster1926].
When Hess and others expressed their disappointment at this appropriation of
roles, Millikan pointed out that on no occasion had he taken the initiative to appro-
priate the discovery, and that the important thing was that with the contribution of
all the nature of this was finally clarified, as a radiation of extraterrestrial origin
[Xu1986]. It is illustrative in this context to read an article by Millikan himself,
History of Research in Cosmic rays [Millikan1930a], published in Nature in July
1930, in which the American physicist reports his general considerations about the
events that gave rise to the study of cosmic radiation and the relationship between
European and American science. Despite the controversy that characterized the rela-
tionship between Millikan’s school and the environment of European physicists in
those years, later in 1936, when the Nobel Prize was awarded to Hess for the discovery
of cosmic rays, Millikan, in congratulating him for the assignment, he publicly wrote
that Gockel, Hess and Kolhörster had been the pioneers until 1914 and that further
work had appeared only ten years later, when the study of these processes had entered
a new era (Fig. 2.3).
2.3 The Acceptance of the Idea of a Cosmic Radiation
Starting from 1927 it can therefore be said that the idea of a cosmic radiation, that
is, of a radiation of extraterrestrial origin, had been universally accepted, with the
exception of a few physicists who continued to have doubts about the results reported.
Apart from the initial attitude of Millikan and in general of the American scientific
community, it must be said that a notable contribution to the understanding of the
phenomenology of observations regarding the cosmic radiation certainly came from
those works carried out by Millikan and collaborators around the mid-1920s, which
are believed to constitute: (i) a finally unambiguous proof of the existence of radiation
of extraterrestrial origin, which cannot be explained on the basis of the radioactivity
of terrestrial substances; (ii) a demonstration of the fact that this radiation has a
spectral composition, with several components, subject to different absorption effects
in matter; (iii) an interpretation of the soft component as constituted by energetic beta
rays (electrons); (iv) a more precise estimate of the ionization present at ground level
and due to cosmic radiation, evaluated in 1.4 ions/(cm3 s).
From this moment a new phase begins, which will see an increasing number of
experimenters contribute to the detailed knowledge of this radiation, especially in
the United States. If before that period most of the articles devoted to the study
of these phenomena were published in German language journals, since 1930 the
situation is reversed, and most of the scientific literature concerning cosmic radiation
can be found in American journals, particularly Physical Review. As an example,
Fig. 2.4 shows the number of articles published in Physical Review, containing the
word “cosmic” in the title or abstract of the article, in each five-years period since
1920. As the figure shows, there is literally an explosion of articles on the subject
starting in the 1930s and in the following decades. It is also possible to note a decline
in scientific production during the period of the Second World War. In more recent
years, the term “cosmic” has been associated in scientific papers also with topics other
than those strictly related to cosmic ray physics and therefore it is not significant for
the purposes of these considerations.
This new period in the study of cosmic radiation is also characterized, as we will
see in more detail in the following chapters, by the introduction and use of new
experimental techniques, such as the Geiger counter and the cloud chamber, which
will allow to achieve far superior quantitative results to those obtainable with elec-
troscopes, and from the use of coincidence techniques involving several detectors,
Fig. 2.4 Number of publications that appeared in the Physical Review since 1920, containing the
word “cosmic” in the title or abstract of the article, grouped by five-year periods
2.3 The Acceptance of the Idea of a Cosmic Radiation 39
which permit to understand many aspects precluded to the use of inclusive measure-
ments. These tools will allow far-reaching measurement campaigns, in the most
disparate locations of the Earth, and, therefore, to highlight, in a quantitative way,
the existence of specific effects that will also help in understanding the nature and
the overall properties of both primary and secondary radiation. In fact, starting from
the 1930s, extensive measurement campaigns of the cosmic flux and its dependence
on geomagnetic phenomena began, such as the dependence on latitude and longi-
tude, the East–West effect, and the connection with the Earth’s magnetic field. These
measurement campaigns will also allow answering one of the questions that remained
open for at least a decade after the acceptance of the existence of cosmic radiation,
namely its composition, both as regards the primary and secondary components. The
discoveries concerning the existence of new particles, many of which observed in
cosmic rays, and the detailed study of their properties were not extraneous to this.
Finally, the experimental evidence of the existence of extensive air showers and the
study of the propagation and development of these showers in the atmosphere helped
to clarify the overall picture of the phenomenology of cosmic radiation, establishing
a link that still lasts today between the study of this radiation and the high-energy
nuclear collision processes at particle accelerators.
Chapter 3
The Nature of the Cosmic Radiation
Abstract Understanding the nature of the cosmic radiation received a strong boost
from the study of the cosmic ray intensity in different locations of Earth. The influence
of the magnetic field of the Earth on the cosmic ray flux is the main topic of the first
part of the chapter. Magnetic field effects on cosmic rays were investigated in a
series of famous worldwide measurements campaigns, carried out by the European
physicists Clay, Bothe and Kolhörster, and in an extended way, by the American
teams leaded by Millikan and Compton. Such effects were also theoretically studied
by Lemaitre and Vallarta. As a result, new insights about the corpuscular nature of
the cosmic radiation were obtained. Measurements by Compton are reported in their
original form, with a discussion on the main points concerned with the nature of
the radiation. Related concepts of geomagnetic coordinates and magnetic rigidity
of charged particles are discussed in Appendices C and D. The second part of the
chapter summarizes other contributions made in that period in various parts of the
world (in different European countries, Japan, India and Soviet Union) to clarify the
nature of the established extraterrestrial radiation, thus arriving to the idea of protons
as the main component of the primary radiation.
3.1 The Influence of the Earth’s Magnetic Field
Millikan’s measurements, performed even at various depths under the water of lakes,
had made possible to quantitatively evaluate the absorption coefficients of cosmic
radiation. The evidence that this radiation had a penetrating power extremely higher
than the known gamma radiation suggested by analogy that they too were extremely
high-energy gamma radiations. More generally, it was natural to ask what the nature
of these radiations was: those known up to that moment, following the studies on
radioactivity, were—in addition to gamma radiations—beta rays, i.e. electrons, and
alpha rays, nuclei of 4 He. The latter, being absorbed very easily by matter, could
therefore certainly not be a crucial component of the cosmic radiation observed at
sea level.
To understand the question of the nature of cosmic radiation it must be remem-
bered that while the gamma radiation is not deflected or influenced by electric or

https://doi.org/10.1007/978-3-031-24762-0_3
42 3 The Nature of the Cosmic Radiation
magnetic fields, electrons, on the contrary, undergo the action of these fields. A
moving electric charge, in fact, is equivalent to an electric current, and is subject to
the same electromagnetic effects. For a phenomenon, such as that of cosmic radiation,
which involves the entire Earth’s atmosphere, it is natural to think of the existence
of the Earth’s magnetic field.
The Earth, as well as some of the other planets of the Solar System and the Sun
itself, is provided with a magnetic field, which extends over large distances from the
Earth’s surface. As a first approximation we know that the Earth’s magnetic field can
be represented by a dipole, similar to the field created by a magnetized bar. However,
the magnetic field generated by the Earth is more complex: the center of this dipole
is displaced with respect to the center of the Earth, its axis does not coincide with
the axis of rotation and the position of the magnetic poles changes over time. We
know today that even the direction of the Earth’s magnetic field can reverse, and
that this phenomenon has occurred countless times throughout the history of our
Planet (Appendix C). The extension of the Earth’s magnetic field is enormous, even
if the intensity of this magnetic field is weak. Therefore, it can also act on charged
particles at large distances from the Earth. Leaving aside the question of the origin
and detailed description of the magnetic field inside the Earth, we are interested in
this context to understand the importance of the magnetic field effect for the purpose
of understanding the nature of the cosmic radiation.
The Northern Lights phenomenon was already known in ancient times, so called
from the end of 1500: a brightness in the sky sporadically visible in regions of
high latitudes, which can take on different shapes and colours. The light emitted
during these events is characterized by a spectrum that includes lines and bands,
and is due to phenomena of ionization and excitation of the atoms that make up the
upper part of the atmosphere. As early as the late 1800s it had been suggested by
Birkeland that this phenomenon could be due to charged particles emitted by the
Sun, predictably electrons, interacting with atoms in the atmosphere. The action of
the Earth’s magnetic field was able to deflect these particles, preferentially conveying
them towards the polar regions. Experimental studies were also done in the laboratory,
using a cathode ray (electron) beam and a model of the structure of the Earth with a
dipolar magnetic field, observing a luminescence like that observed in the Northern
Lights [Rypdal1997] (Fig. 3.1).
A theoretical study of the dynamics of these charged particles under the action
of the Earth’s magnetic field was carried out by the Norwegian mathematician and
geophysicist Carl Størmer (1874–1957), who as early as 1903 analysed the quan-
titative behaviour of possible trajectories of electrons emitted during periods of
intense solar activity, relating them to the observation of the phenomenon of the
Northern Lights [Størmer1907]. Further developments of Størmer’s theory were
published in several subsequent papers [Størmer1930] and in a complete reference
text [Størmer1955] (Fig. 3.2).
The action of the Earth’s magnetic field on any electron is to deviate them from
their trajectory, and this deviation would occur differently in the equatorial regions
and in the polar regions, due to the different magnetic field. It was therefore expected
that if cosmic rays were made up of electrons, they would be subject to the influence
3.1 The Influence of the Earth’s Magnetic Field 43
Fig. 3.1 The image of a Northern Lights. Shows of this kind are visible above all in the regions
of higher latitude, although sporadically they can also be observed in the latitudes of the regions
of Northern Italy. Source Wikimedia Commons, under Creative Commons Attribution 2.0 Generic
License
of the Earth’s magnetic field, which exhibits different values in different geographical
regions. On the contrary, if cosmic rays were made up of photons, their trajectories
would not have to undergo any deviation. Of course, even in the case of electrons
of very high energy the deviation due to the Earth’s magnetic field would be very
small. However, if a part of these electrons had energies lower than some GeV, they
too would be deflected by the magnetic field, and we could expect differences in
intensity in the flux measured at various latitudes.
The study of the behaviour of charged particles in the Earth’s magnetic field
received notable contributions from several other authors, apart from Størmer, espe-
cially from Georges Lemaitre (1894–1966), Belgian astronomer and priest, and from
the Mexican Sandoval Vallarta (1899–1977). Georges Lemaitre is best known for his
cosmological theories on the expansion of the universe and the Big Bang hypothesis.
Fig. 3.2 Fredrik Carl Mülert Størmer (1874–1957). Source Wikimedia Commons, under Creative
Commons Generic 2.0 License from Nasjonalbiblioteket
In the context of this issue, Lemaitre and Vallarta published several papers
[Lemaitre 1933, Lemaitre1936a, Lemaitre1936b], which quantitatively analysed
how low-energy particles would be deflected by the Earth’s magnetic field near the
geomagnetic equator but could reach the Earth near the geomagnetic poles (Figs. 3.3
and 3.4).
Calculating the trajectories of these particles is very complex. Unlike Størmer,
Lemaitre and Vallarta were able to use at the Massachusetts Institute of Technology
(MIT), where they both worked for a certain period, of a calculation tool, a sort of
analog mechanical calculator (the Differential Analyzer developed by Vannevar Bush
(1890- 1974), an electrical engineer from MIT), which allowed to solve differential
equations even of high order. Lemaitre remained interested and intrigued even in the
following decades by the potential of these first computers, which he was able to use
directly in the 1950s.
Considerations on the possible trajectories of particles in the Earth’s magnetic
field placed stringent constraints on the nature of cosmic radiation. If the primary
particles were photons, they would not undergo the action of the magnetic field, so
their intensity should be the same both in the equatorial and polar regions. Even
if photons could produce charged particles by interacting with nuclei in the upper
atmosphere, they would not be significantly deviated from their original direction,
3.1 The Influence of the Earth’s Magnetic Field 45
Fig. 3.3 George Lemaitre (1894–1966). Source American Institute of Physics (AIP), Emilio Segrè
Visual Archives
due to the low intensity of the magnetic field and to the fact that the trajectories
followed would have a length of the order of a few km. If, on the other hand, the
primary particles were charged particles, they would undergo the action of the Earth’s
magnetic field even at large distances from the Earth. Those arriving in the direction
of the poles could reach the Earth, even those with a low energy, while those arriving
in the equatorial plane would be strongly deflected, especially the particles with a
low energy; in this case we should expect a different intensity at different latitudes.
A quantitative analysis of the problem shows what must be the minimum energy that
the particles must have to reach the Earth at any given (geomagnetic) latitude.
In the work of 1933 [Lemaitre1933] Lemaitre and Vallarta showed that the vari-
ations in intensity of the cosmic radiation with latitude could be fully explained if it
was accepted that the radiation consisted, at least in part, of electrons—or protons—
of energy from the order of 10 GeV (1010 eV), arriving on Earth and coming from
all directions.
Figure 3.5 shows the expected trend of the radiation intensity as a function of
geomagnetic latitude, for different values of a parameter (from 0.1 to 1.0) which
is related to the energy of the incident particle considered [Lemaitre1933]. As an
example, a parameter value equal to 0.4 would correspond to 9.54 GeV in the case
of electrons, and 8.61 GeV in the case of protons.
The problem of geomagnetic effects on the intensity of cosmic radiation was
addressed in the 1930s by numerous other authors, as also reported in a review
article of the period [Johnson1938].
Fig. 3.4 Sandoval Vallarta (1899–1977). Source American Institute of Physics (AIP), Emilio Segrè
Visual Archives
Fig. 3.5 Trend predicted by Lemaitre and Vallarta [Lemaitre1933] of the intensity of the cosmic
radiation as a function of geomagnetic latitude (see text). Figure reproduced from the work
of G. Lemaitre and M. S. Vallarta, On Compton’s latitude effect of cosmic radiation, Phys-
ical Review 43(1933)87. Copyright (1933) by the American Physical Society, License No.
RNP/22/MAY/054177
3.2 Campaigns for Measuring the Intensity of the Cosmic Radiation … 47
3.2 Campaigns for Measuring the Intensity of the Cosmic

Radiation in Various Geographical Locations
The considerations made so far show how extremely important was to have a quan-
titative measure of the intensity of the cosmic radiation at different latitudes. This
required measuring campaigns in the most disparate areas of the Earth, which in
those years were indeed carried out, from latitudes close to the equator up to lati-
tudes close to the poles. In the survey concerning the activity organized by his
group of collaborators, which Compton published in 1933 [Compton1933] he recalls,
quoting their works, how measurement campaigns of this kind had been carried out in
previous years by the Dutch physicist Jacob Clay (1882–1955) (Fig. 3.6) [Clay1927,
Clay1928] in a series of trips between Amsterdam and Batavia (Java), reporting lower
intensity values near the equator, by Millikan and collaborators [Millikan1928a,
Millikan1930b], who had measured the intensity of the radiation between California
and Bolivia (observing very small variations) and between California and Churchill,
a locality close to the magnetic North Pole (not observing any variation), from the
Germans Bothe and Kolhörster [Bothe1930a], from Hamburg towards the North
Pole, up to a latitude of 81° N (without observing any variation within the limits
of experimental errors), from Kennedy and Grant [Grant1931] from Australia to
Antarctica (still without appreciable observed changes) and by Corlin [Corlin1930]
in Scandinavia, from latitude 50° N to 70° N (observing small variations, with a
maximum around 55° N). Further measurement campaigns in various geographical
locations had been organized in those years also by the French physicists Pierre
Auger (1899–1993) and Louis Leprince-Ringuet (1901–2000), and by the Indian
Piara Singh Gill (1911–2002) on the occasion of numerous ocean crossings between
Vancouver and Tasmania.
Overall, the set of these observation campaigns seemed to indicate that the inten-
sity of cosmic radiation was approximately constant in the various locations explored
up to that moment, with the exception of the results obtained by Clay. If it had
been shown that the intensity of the observed radiation was actually independent
of the geographic location, in particular of the different Earth’s magnetic field, this
would have supported the hypothesis that the radiation was constituted by gamma (or
possibly by charged particles with energy so high as to not be appreciably deflected
by the Earth’s magnetic field), thus denying the hypotheses that Bothe and Kolhörster
had already made in 1929 [Bothe1929a] about the existence of a corpuscular radia-
tion. On the contrary, the results obtained by Clay made it possible to observe with
certainty the existence of considerable variations in the intensity of the cosmic radi-
ation between these two explored latitudes, variations which were then estimated at
around 15–20% in the observations obtained a few years later.
It was not clear, therefore, whether the intensity of the cosmic radiation really
changed with latitude, because different experiments had given completely different
answers. Although it may seem strange, all the different answers were reasonable,
since, as it became clear later, the dependence of the flux on the geographical region
Fig. 3.6 Jacob Clay (1882–1955). Source Academictree.org
must be related more to the geomagnetic properties of the location (geomagnetic lati-
tude) than to the purely geographical location (see Appendix C). Clay’s measurement
campaign had been conducted along geomagnetic paths that allowed more evident
variations to be observed, while the campaign by the German physicists had been
conducted along paths in which the expected variations were negligible.
Arthur Compton (1862–1962), former Nobel Prize winner in 1927 for the
discovery of the mechanisms of radiation-matter interaction (Fig. 3.7), decided to
enter the research field of cosmic radiation, not convinced by Millikan’s arguments
regarding the nature and origin of this radiation. We will return again to the impor-
tant debate between Millikan and Compton about the corpuscular or photonic nature
of the cosmic radiation. Compton became interested in the nature of the cosmic
radiation only after the mid-1920s, perhaps following a trip to India for a series of
lectures delivered at the University of Punjab, which included a high-altitude scien-
tific expedition to measure the intensity of the cosmic radiation at different altitudes
[Johnston1967, Yodh2013]. In the early 1930s, Compton organized a campaign of
detailed measurements, the results of which were reported in a more systematic form
in 1933 [Compton1933].
It extended over an enormous geographical area (Fig. 3.8). The campaign of
measurements organized by Compton really represented a far-reaching enterprise
in the panorama of the time. Several expeditions, with similar equipment, based on
ionization chambers, travelled the globe, in the same period, in the most disparate
geographical locations. Compton himself, who is portrayed aboard a ship playing on
the deck in one of the archive photos, also participated in some of them.
Although preliminary results had been previously published by Compton himself
and collaborators [Compton1931, Bennett1931, Compton 1932a, Bennett1932a,
Bennett1932b, Compton1932b, Compton1932c], in the 1933 survey he summarizes
the main results obtained from eight different expeditions, whose data were already
available and which required—according to his statements—not to further delay their
publication, probably because in the same period the publication of the results of
another campaign of measures on a worldwide scale, organized by Millikan himself,
was also underway, which will be discussed further on (Fig. 3.8).
In this review, published in Volume 43 of Physical Review, dated March 15, 1933
(but received by the Journal on January 30), Compton, after the description of the
experimental procedures and the corrections necessary for the interpretation of the
Fig. 3.7 Arthur Compton (1862–1962). Source American Institute of Physics (AIP), Emilio Segrè
Visual Archives
Fig. 3.8 Approximate location (blue dots) of the main stations for measuring the intensity of cosmic
radiation, used during the measurement campaign organized by Compton [Compton1933]
results (for example those for the different atmospheric pressure due to the altitude,
for the possible shielding effects offered by the buildings, …), summarizes in a long
table all the results obtained, which are then reported in some of the famous graphs
as a function of the geomagnetic latitude. The data reported by Compton, grouped
into three ranges of altitudes (i.e., atmospheric pressure) and normalized to each
other, are shown in Fig. 3.9. They show, according to Compton’s interpretation, a
clear difference between the intensity measured at geomagnetic latitudes close to the
equator compared to that measured at geomagnetic latitudes above 50° N or S. This
difference amounts to 14% at sea level, but it is greater at high altitudes, reaching
22% at an equivalent altitude of 2000 m and even 33% at an equivalent altitude of
4360 m.
The measured values, reported at an atmospheric pressure of 76 cm Hg, equivalent
to sea level, are here also plotted in a more detailed figure (Fig. 3.10), together with the
results obtained by Clay at different latitudes and with the value obtained by Millikan.
In the original figure of Compton’s article, some values had been averaged, while
Fig. 3.10 shows the values reported in the original table. The agreement between the
different data sets is good, so much that Compton concludes that “From a statistical
standpoint, therefore, the probability of the existence of this latitude effect amounts
practically to a certainty” [Compton1933].
Fig. 3.9 Radiation intensity measured as part of the expeditions organized by Compton
[Compton1933] as a function of geomagnetic latitude, for three different altitude ranges, identified
according to the value of the atmospheric pressure
Fig. 3.10 Radiation intensity at sea level, measured during the campaign organized by Compton
[Compton1933] for different geomagnetic latitudes, and reported together with the values obtained
by Clay and by Millikan. The values reported by Compton are distinct for the Northern and Southern
hemisphere, with two different symbols
In Compton’s original work, dashed curves are also shown, superimposed to the
data, which represent calculations made on the basis of the theory of Lemaitre and
Vallarta [Lemaitre1933], already cited, which hypothesized the arrival of high-energy
electrons (5 and 13 GeV) in the vicinity of the Earth and their deflection following
interaction with the Earth’s magnetic field. Those of higher energy would be able to
get to the poles, but not to latitudes below 50° N. The qualitative similarity of the
curves with the trend of the measured data suggested to Compton that a substantial
fraction of the particles could therefore be electrons of energy intermediate between
these two values. Compton reports the data also as a function of the geographic
latitude, observing an undefined trend, with a much greater dispersion of the data.
The latitude effect, therefore, cannot be merely linked to the geographical latitude.
Furthermore, if the same data are reported as a function of the local magnetic field
(which extends for hundreds of km above the ground level), also in this case a
not well defined distribution of values is observed, as instead occurs in the case of
geomagnetic latitude.
This last result indicates something important, namely that the latitude effect
cannot be due to a deflection of particles at low altitudes, in the Earth’s atmosphere,
but rather to phenomena that occur at much greater altitudes, hundreds of km, in the
case of electrons with energy starting from some GeV. These considerations suggest
that at least a significant fraction of the primary particles that reach the top of the
atmosphere, coming from the outside, are charged particles.
The survey published by Compton was countered by another review of results,
also obtained in various locations around the world, this time by Millikan and collab-
orators, published in April 1936 in Physical Review [Millikan1936], although some
results had been published previously [Millikan1935b]. A striking aspect in this work
by Millikan is the absolute lack of references to the work and results of both Compton
and other authors. In fact, for the most part, the bibliographic references are self-
citations of Millikan’s previous works. This survey of results also required a massive
organization: twelve different expeditions, by sea for the most part, with equiva-
lent detection equipment, a variety of locations and routes explored that covered a
substantial part of the globe, from the northern to the austral hemisphere, in a high
range of geographical (and magnetic) latitudes. The result of this survey is first of
all the confirmation of a latitude effect: the flux measured by the instruments shows
a lower value at geomagnetic latitudes close to the equator, and an increasing value
up to about 45° of geomagnetic latitude North, latitude at which the flux saturates
without showing further increases at even greater latitudes. The observed variation is
around 12%. The measurements also carried out at southern geomagnetic latitudes
also allow us to observe an “equatorial dip”, a valley in the flux values, between
geomagnetic latitudes of about 40° S and 40° N, fairly symmetrical with respect to
the geomagnetic equator.
Millikan’s considerations on these results concern once again, as Compton had
already done, the comparison with the analysis of Lemaitre and Vallarta, and therefore
the conclusion that if we are dealing with primary electrons, they must have an energy
around 6 GeV (within an uncertainty of 20%). The existence of a single geomagnetic
latitude value at which the flux begins to decrease indicates, according to Millikan,
that there cannot be two classes of particles (such as protons and electrons) present
at the same time, otherwise the intensity trend with the geomagnetic latitude should
have shown evidence of a second decrease.
The abundance of results obtained at the different geomagnetic latitudes finally
allows Millikan to draw a map of the intensity in the various regions of the globe,
with lines that connect the locations where the intensity assumes the same value,
therefore of the contour lines of the intensity of cosmic radiation (Fig. 3.11). This
figure clearly shows that these contour lines generally follow a trend parallel to the
geomagnetic equator, with an anomaly along the equator, and in particular in the area
of South America.
Strangely enough, no discussion and interpretation of the results in relation to
the nature of the primary cosmic radiation is present in this work, although in the
Abstract of the article it is stated that these effects related to geomagnetic latitude
derive from an asymmetry of the Earth’s magnetic field, which extend up to large
distances from the Earth, and that these effects are expected both in the case in which
the primary radiation is of a corpuscular nature and in the case of photons. A marked
difference, therefore, with respect to the clear interpretation of Compton, who saw
in this effect linked to the geomagnetic latitude a consistent proof of the fact that the
cosmic radiation was largely of a corpuscular type. This attachment of Millikan to the
Fig. 3.11 Map of the equal intensity curves of cosmic radiation, obtained from the survey by
Millikan and collaborators [Millikan1936]. Figure obtained from the work of R. A. Millikan
and H. Victor Neher, A Precision World Survey of Sea-Level Cosmic-Ray Intensities, Phys-
RNP/22/MAY/053613
interpretation of radiation as essentially due to photons lasted a long time and gave
rise to controversies between the two Nobel laureates, which the press emphasized,
giving ample response to them.
3.3 The Debate on the Corpuscular or Radiative Nature

of the Cosmic Radiation
Millikan’s position on the nature of the cosmic radiation is also linked, according to
the opinion of many historians of science, to the general vision of the world and of
reality that Millikan professed, full of elements including those of religious nature,
which led him to defend, even on principle, certain interpretations. It was also a period
in which questions about the age of the universe, its dimensions, as well as its origin
and evolution, were still in their initial phase. Millikan supported a cosmological
hypothesis of a universe in continuous creation, while he did not look favourably
an interpretation of a universe destined to a thermodynamic “death”. In this context,
the cosmic radiation had to play for Millikan the role of the “cry” of the new atoms
forming in the interstellar space, a proof that the creative process was still in progress.
“Millikan finds Creation still goes on while Creator directs the Universe”, was the
headline of the New York Times, after the communication made by Millikan to the
Assembly of the American Association for the Advancement of Sciences (AAAS).
According to this interpretation, the processes that led to the formation of new atoms
had to produce gamma radiations as a result.
It was also these considerations that led Compton to enter the arena of cosmic ray
physics, to verify for himself whether the corpuscular hypothesis, which he believed
most, could be reflected in the observations of the intensity of the cosmic radiation
at various latitudes. The first measurements made by Millikan in different locations
seemed to indicate that the intensity was constant everywhere. Still in 1932, at the
AAAS Annual Meeting, Millikan argued that there was no latitude effect, presenting
his results in antagonism to those of Compton, discussed in the same Meeting. But
after Compton’s results, in 1933, the evidence seemed in favour of a corpuscular
nature, which Compton announced, and which the press could not help but note, as it
was the opinion of another Nobel laureate. That Millikan, despite his authority, was
wrong? And if he was wrong about the nature of the cosmic radiation, what about
the cosmological visions that were related to it?
The debate between Millikan and Compton also had a following in the non-
specialist press, putting the question of science in the limelight of a broader cultural
context, as the journalist Wolferton has recently described [Wolferton2019], but of
course it was also played out in official contexts, conferences and scientific meetings,
which even at that time were precious occasions for the presentation of one’s results
and for the exchange of opinions with researchers working in the same field.
Figure 3.12 shows one of the historical photos, taken on the occasion of the
International Congress of Nuclear Physics, organized in Rome by Enrico Fermi
3.3 The Debate on the Corpuscular or Radiative Nature of the Cosmic Radiation 55
Fig. 3.12 Participants in the International Congress of Nuclear Physics in Rome, October 1931.
Among the scientists present on this occasion, Millikan and Compton are also recognizable, in the
central part of the photo. Source Wikimedia Commons
in October 1931. Among the leading physicists participating in this Congress are
present, among many others, also Millikan and Compton. It has often been noted
that in this photo there is only one woman, Marie Curie, compared to about forty
men, a proportion that fortunately has changed drastically in recent times, although
many steps remain to be taken in this direction.
Coming back to the question of the origin of cosmic radiation, Millikan supported
for many years hypotheses on which he then had to change his mind. Already in one
of his articles of 1926 [Millikan1926b], Millikan had claimed that this radiation
was emitted by nuclear transmutations, such as the capture of an electron by a light
nucleus, the formation of Helium starting from Hydrogen or other processes such
as the condensation of radiation into atoms, assuming that these cosmic radiations
presumably took place not in stars, but in space and interstellar matter.
In a subsequent article [Millikan1928b], strictly dedicated to the origin of cosmic
radiation, Millikan better clarifies his considerations about the nature of these
processes capable of giving rise to the gamma radiation that he hypothesizes to
constitute cosmic rays. Further precision measurements of the absorption of radi-
ation in water, performed in the same Arrowhead Lake (1570 m) already used for
previous measurements and in Gem Lake (2760 m), contributed to this vision. From
these results Millikan had deduced three different absorption coefficients for the
cosmic radiation: one component with a coefficient of 0.35 per meter of water, one
with a coefficient of 0.08 and finally a third one with a coefficient of 0.04.
The fact that the penetrating radiation had different values of the absorption
coefficient was interpreted as due to the existence of different energy values of the
gamma radiation, produced in nuclear processes with different energy values, such
as quantum jumps. The enormous energy involved for these radiations implied, in
Millikan’s vision, processes of formation of heavier nuclei abundant in nature, such
as helium, oxygen, silicon or iron, starting from the lightest nuclei (hydrogen or
helium). There was also speculation about where these processes could take place,
whether in the interior of stars or in the interstellar medium, concluding that it is the
very low-density interstellar, or intergalactic medium, that is the natural candidate
for the processes at the origin of the cosmic radiation.
In the years following 1933, therefore, even if evidence of a different type accu-
mulated in favour of the corpuscular nature of cosmic radiation, Millikan continued
to support the hypothesis of a highly penetrating gamma radiation, as evidenced by
his works in the following years [Millikan1934, Millikan1936].
In 1936, however, Compton decided to definitely clarify the evidence supporting
the corpuscular hypothesis, in a work published in Physical Review [Compton1936],
summarizing this evidence in three different categories:
(a) those related to the effects of latitude and directional asymmetry
(b) those obtained from coincidence experiments
(c) those coming from asymmetry effects due to the motion of the Earth in the
Galaxy.
Note, once again, the Introduction section in this article, which reports not only the
bibliographic references relating to the works that support the corpuscular hypothesis,
but also, for correctness, those works, especially those of Millikan, which support a
different hypothesis and an alternative interpretation.
Compton basically lists three arguments in favour of the corpuscular hypothesis.
The first concerns precisely the set of geographical variations observed in the
intensity of the cosmic radiation. The fact that these variations follow the course of
the Earth’s magnetic field very closely suggests that the causes of these variations are
of magnetic origin and that therefore at least a non-negligible fraction of the cosmic
radiation observed at sea level must be produced by primary charged particles, since
the effects are not reducible—according to the theory of Lemaitre and Vallarta as
well as that of Størmer, already cited—to those produced by the magnetic field on
secondary particles produced in the atmosphere. The measurement campaigns that
made possible to observe these geomagnetic variations are those organized by Clay,
by Compton himself and by Millikan. According to Clay’s data, which Compton
considers as a model, the reduction in radiation intensity is 17% between geomagnetic
latitude 45° N and geomagnetic latitude 0°. At least this fraction of 17%, Compton
concludes, must be charged, leaving the remaining 83% fraction to be interpreted.
But the experiments carried out with the first counters operated in coincidence
as telescopes, therefore sensitive to the direction of arrival, also showed a further
effect, namely that the intensity near the equator was greater from the West direction
than that measured from the East direction. These are the first evidence of what will
be called “East–West Asymmetry”, and which we will discuss in more detail in a
later chapter. According to Rossi’s results, for example, near the equator and at an
inclination angle of 45°, the intensity from the East is 76% of that from the West.
This means that at least 24% of the particles coming from the West (which corre-
spond to about 12% of those incidents vertically) are charged. This further increases
the minimum percentage of charged particles that must be present in the primary
radiation. Further arguments, based on the energy distribution of the particles, are
then used to establish an important feature, namely that the relevance of this latitude
effect is limited simply by the strength of the Earth’s magnetic field: if this were
much higher, the fraction of particles able to reach the Earth near the equator would
be practically null, demonstrating that almost all of the primary radiation must be
charged.
The second argument concerns the results that had been obtained in those years
with the use of counters operating in coincidence, also with the aid of cameras for
viewing the tracks. The development of cosmic radiation detection techniques had
made considerable progress in the previous decade compared to the use of the first
electroscopes and will be the subject of a more detailed discussion in Chap. 5. We
only recall here that by the date of 1936, in which Compton tries to defend his
considerations about the corpuscular nature of the cosmic radiation, the first Geiger
counters and the first cloud chambers had already been used, that the technique of
coincidence between individual counters had been introduced, also to determine the
direction of motion of the particles, and that in the meantime new particles had
been observed, such as the positive electron (positron) and the neutron, which we
will discuss later. The results of experiments conducted by several authors using
counters operating in coincidence are considered by Compton perhaps the most
striking evidence of the corpuscular nature of radiation, as they provide evidence of
the passage of single high-energy particles.
With reference to Fig. 3.13, for example, which shows a typical experimental
setup used in those years, the passage of a particle capable of giving a signal in
coincidence in the detectors C1 , C2 and C3 , allows to “trigger” the display (and
the relative photographic recording) of any tracks in the “Chamber”. In most cases
(90%) these photographed tracks are single tracks, left by highly penetrating particles,
capable of crossing 40–50 cm of lead along their path. Evidence of this kind had
been obtained from Rossi [Rossi1932a, Rossi1933a], from Auger and Ehrenfest in
1934 [Auger1934] and from Street and collaborators in 1935 [Street1935].
The experiments of Rossi [Rossi1933a] and Hsiung [Hsiung1934], in particular,
showed that these penetrating particles were not secondary particles generated near
the counters, at sea level, but came from much greater heights. Compton admits that in
principle they could be generated by gamma rays at high altitudes in the atmosphere;
however, he excludes this possibility, as they are subject to the same latitude effect
as the overall ionization due to cosmic radiation, and also show effects of directional
asymmetry near the equator. They, therefore, in Compton’s opinion, are generated
by primary particles of high energy, or are even themselves the primary particles that
make up the cosmic radiation.
Fig. 3.13 A typical experimental setup, used in the mid-1930s, for the display of tracks in a cloud
chamber, triggered by the coincidence between several counters arranged vertically [Street1935]
These considerations show how long and tiring the path leading to the current
visions of primary and secondary cosmic radiation has been. We are in a period
in which many of the elementary particles, in terms of which we describe today
the interactions that occur in the atmosphere due to the arrival of a primary, are
still to be discovered, and therefore the statements about the very nature of cosmic
radiation try to concatenate various experimentally observed elements, without yet
knowing in detail many of the aspects involved in the different processes. To these
arguments, Compton adds the consideration that in cloud chambers, in addition to
the single tracks due to the passage of highly penetrating particles, a small “shower”
is sometimes observed, produced by photons, but that these showers are also subject
to the latitude effect, demonstrating that these photons in turn must also be produced
by primary particles.
But there is another class of arguments, the third one, which Compton argues
in favour of the corpuscular nature of the cosmic radiation, linked to the effect on
the intensity of this radiation due to the motion of the Earth with the rotation of the
Milky Way. According to the results provided by the astronomers, it had already
been estimated that the rotation of our galaxy, the Milky Way, drags the Earth in
the direction of the Constellation of Cygnus, with a speed that is about 0.001 c,
that is, one thousandth of the speed of light. If the primary radiation consisted of
photons, they would be subject to the Doppler effect, and a difference should be
observed between those arriving from one side and those from the opposite side
with respect to the motion of the Earth, an effect that had been actually studied and
quantified by Compton and Getting [Compton1935]. The experimentally observed
effect indeed exists but is much less than what one would expect from a primary
radiation made up of photons and it is instead compatible with the hypothesis of a
radiation made up of charged particles, in which case the effect is very small, since
they would be strongly deflected by the magnetic field. Two interesting aspects of
these considerations should be noted: the first is that the presence of primary photons
is not entirely excluded, admitting that they could be part of the primary radiation,
but in an extremely small fraction of a few percent. The second aspect is that in this
discussion the primary charged particles are once again hypothesized essentially as
electrons or positrons.
As we said at the beginning of this section, Compton also intends to discuss the
arguments raised so far in favour of the hypothesis that the primary radiation is
constituted by gamma, arguments that were mainly supported by Millikan’s group.
According to these hypotheses, which Millikan had maintained up to that date, in
1936, it was possible to admit the presence of a small fraction, about 15%, of a
charged component, while the remaining 85% must have been made up of gammas. In
support of this hypothesis, Millikan cited some events observed in the cloud chamber
by Anderson and collaborators, events that showed the production of electron and
positron showers induced by gammas. However, these events were interpreted by
Compton as due to gammas, also present in the secondary radiation, that is, produced
in turn by the arrival of a primary in the upper part of the atmosphere. Further
considerations about the different absorption of electrons and high-energy photons
are also addressed and discussed, showing that on the basis of the knowledge already
established at that time, the processes induced by both photons and high-energy
electrons are substantially the same, as the high-energy electrons interact mainly
through radiative processes, that is, with the emission of radiation. This means that
the assertion that photons were 100 times more penetrating than electrons with the
same energy was to be discarded on the basis of now commonly accepted knowledge.
Summarizing all the pros and cons of the hypotheses about the nature of primary
radiation, Compton concludes that at the top of the atmosphere all observations
seem to indicate that at least 75%, and probably 95% near the poles, of the primary
radiation is made up of charged particles. At sea level, the evidence derived from
the latitude effect, from the coincidence measurements and from the small daily
variations observed, due to the motion of the Earth, indicate that no more than a few
percent of the primary radiation can be made up of gamma.
Regarding the typology of these primary charged particles, Compton can only
assume that they are electrons and positrons, although “There is also some sugges-
tion of the existence of protons. We cannot now discuss these interesting details”
[Compton1936a]. The crucial question, however, which according to Compton repre-
sents the most important feature of the last decade of research on cosmic radiation,
is precisely its charged corpuscular nature.
Despite these considerations by Compton, which seem to establish with certainty
the characteristic at least of charged particles as a source of the primary cosmic
radiation, still in 1937–1938 Millikan moves in a totally independent way from these
statements, practically never mentioning them, focusing attention on other aspects
but, on the other hand, then assuming at a certain point the “corpuscular” ideas as if
they were his own.
In one of the works from this period [Millikan1938b], Millikan, using results
relating to the intensity of ionization measured up to the top of the atmosphere by the
launch of balloons, argues that the incident radiation consists of electrons or a combi-
nation of electrons and photons, since their ability to interact is not very different, and
that these considerations arise from the analysis of their own results. Millikan even
comes to extract an energy distribution of the incident electrons, which he believes
may have energies between 1 and 17 GeV, distributed in different “bands”, with a
maximum around 6 GeV. In this work, the assertion that he himself demonstrated, in
a previous work of a few months before [Millikan1938a], is almost taken for granted
that the primary radiation is largely made up of electrons, and that they could not
interact with matter before entering the Earth’s atmosphere. This also indicates that
they do not have their origin within the stars or in places of the universe where matter
is present in abundance.
Regarding the origin of this radiation, Millikan still insists on a mechanism
according to which it is the transformation into energy of the entire mass of atoms
that is responsible for the production of this radiation, a sort of annihilation of the
most abundant atoms in the universe. Since the mass of a hydrogen atom corresponds
to approximately 1 GeV, the abundant elements 12 C, 14 N, … up to 56 Fe, would corre-
spond to energies of 12, 14, … 56 GeV. This annihilation should therefore produce,
in most cases, an electron–positron pair, and, to a lesser extent, a gamma pair.
Thus, even if on the basis of arguments deriving from his own results and without
making use of the arguments put forward in support of the corpuscular hypothesis by
Compton, Millikan seems to have “converted” his mind to this hypothesis, or, better,
seems to declare that he himself obtained the proof of these claims, an attitude not
unlike from that taken over ten years earlier, when recognizing the very existence of
radiation of extraterrestrial origin.
As for his previous work [Millikan1938a], the assertion that they are electrons
relies on the absorption of these radiations at various altitudes, based on which
Millikan excludes that they are protons. The real novelty lies in starting to consider
that the tracks of the penetrating particles observed, both visually, in the cloud
chamber, and through their passage in the particle counters operated in coincidence,
are not electrons but other charged particles (positive and negative) with a different
ionizing power, referring to the recent observation, just in 1937, by Neddermayer
and Anderson [Neddermayer1937], of what will subsequently be called the µ meson.
In the final part of the article, Millikan comes to discuss the “secondary” character
of this penetrating component, that is, the fact that these particles cannot be present
in the primary radiation but are rather the result of processes that take place in the
Earth’s atmosphere.
3.4 Further Contributions in Europe and Other Countries for Understanding … 61
3.4 Further Contributions in Europe and Other Countries

for Understanding the Nature of the Cosmic Radiation
Although the core of the activities related to the study of the cosmic radiation progres-
sively shifted, starting from the end of the 1920s, towards the other side of the
Atlantic, the contributions coming in that period from the environment of European
research—and, to some extent, from other countries in the world—were certainly not
negligible. Indeed, we have many examples of researchers who continued to work in
European institutions, particularly in Germany, France, Italy, Great Britain, as many
articles have pointed out [Sekido1982, Walter2012]. Significant contributions in the
physics of cosmic rays gradually came also from other, non-European, countries,
already during the 1930s. All these activities, which we will briefly mention in this
section, certainly contributed to clarifying further aspects of the cosmic radiation and
its nature, but above all they established, also through cultural exchanges and stages
of many researchers among the various Institutions existing at the time, a common
substrate in the study of this area of physics.
In Germany, the measurements aboard high-altitude balloons with a human crew
continued in the following years, even by Kolhörster, who organized an ascent to an
altitude of 12,000 m, using a balloon that was 26 m in diameter, filled with about
10,000 cubic meters of hydrogen. To reach these high altitudes, the crew needed
an artificial breathing system with an oxygen cylinder, with an autonomy of several
hours [Horandel2013]. Kolhörster himself in those years also actively contributed in
the study of the intensity of the cosmic radiation in the subsoil, installing particle
counters, also operating in coincidence, inside salt mines, up to a depth of about 1000
equivalent meters of water [Kolhörster1933, Kolhörster1934].
Measurements of the absorption of cosmic rays under a certain depth of water
were also carried out by the German Erich Regener (1881–1955) (Fig. 3.14), in Lake
Constance, up to a depth of 230 m below the surface of the water [Regener1932a],
with an automatic apparatus which recorded the result of the ionization measurement
on a photographic plate. Automatic recording devices were used by Regener in those
years also aboard unmanned balloons [Regener1932b], up to altitudes of 30 km, thus
extending the measurements previously carried out by Hess and Kolhörster. Regener
also made significant experimental contributions for the improvement of automatic
recording equipment, for the development of scintillation counters and in the use of
the first rockets as a tool for high-altitude scientific investigations. Regener’s pupil
and collaborator was the physicist Georg Pfotzer (1909–1981), who contributed to
many of the balloon ascents to evaluate the intensity of the radiation with altitude.
Still in those years, we may recall the flights of the Swiss physicist Auguste Piccard
(1884–1962), together with his brother Jean Piccard (1884–1963), with a pressur-
ized capsule, up to an altitude of over 15 km, for the study of various atmospheric
phenomena and of the cosmic radiation. Auguste Piccard had already made balloon
ascents since 1913. Subsequently, to reach even higher altitudes, in the stratosphere,
he had designed balloons with an airtight cabin, reaching an altitude of 16,000 m in
1932. Two years later his brother Jean reached 17,500 m. Several members of the
Fig. 3.14 Erich Regener (1881–1955)
Piccard family, in addition to the two brothers Auguste and Jean, also contributed
over the years to high-altitude explorations. In the following years Auguste Piccard
dedicated himself to submarine exploration, reaching a depth of 11 km below the
sea level, with the bathyscaphe Trieste, in the Marianas Trench. Figure 3.15 shows
some of the members of the Piccard family with the pressurized capsule employed
during one of the flights into the stratosphere.
First of all, it should be remembered that the cosmic radiation detection tech-
nique, initially based only on the use of electroscopes, had already undergone several
improvements over the years. In particular, an enormous impulse to the study of
cosmic rays came from the use of Geiger counters, widespread starting since 1928
[Geiger1928a, Geiger1928b] and which we will discuss in greater detail in a subse-
quent chapter, and from the possibility of evaluating—by means of coincidence
circuits—whether two or more counters produced coincident signals over time, a
Fig. 3.15 The photo shows William Swann (left), Jean Piccard (right), and Jeannette Ridlon-
Piccard, wife of Jean, close to the capsule employed for stratospheric flights. Source AIP Emilio
Segrè Visual Archives, Physics Today Collection
possibility that allowed the use of a telescope of counters, capable of signalling

the passage of a single particle through the two counters and therefore defining its
direction.
This possibility had been already exploited in 1929 by Bothe and Kolhörster in
Germany, carrying out a series of measurements with two counters placed vertically
on top of each other, laterally protected by a considerable thickness of lead absorbers
and with an absorbent layer also placed between the two counters [Bothe1929a,
Bothe1929b]. In this configuration (Fig. 3.16), it was possible to observe a relatively
large number of coincidence events, which can be interpreted as the passage of a
single penetrating particle through the two counters. The possibility that these events
were due to electrons, emitted through the Compton effect by gamma rays interacting
with the surrounding material, could be excluded, as these electrons have low energies
and should have been stopped by the absorber interposed between the two counters,
an effect not observed. Only considerable thicknesses, several cm, of heavy materials,
such as gold, produced a small reduction in the number of observed coincidences.
Gold has a density of 19.3 g/cm3 , higher than that of lead (11.3 g/cm3 ) but of course
it is not that easy or cheap to find, and Bothe and Kolhörster seem to have resorted
to the National Bank of Germany to obtain loan of blocks of gold to be used during
their measurements. Based on these results, the two German scientists had already
concluded, perhaps overestimating the interpretation of their experiment, that the
penetrating radiation observed was not made up of gamma rays but was rather of a
corpuscular nature [Bothe1929b].
Fig. 3.16 The arrangement used by Bothe and Kolhörster [Bothe1929a] for coincidence measure-
ments between two counters (Z1 , Z2 ) arranged one on top of the other vertically and laterally
protected by absorbers, with an additional thickness of absorber material placed between the two
counters
We will discuss later the contributions of Kolhörster related to the first evidence of
the existence of correlated particles in the secondary cosmic radiation. Again in the
1930s we should remember the studies performed by means of nuclear emulsions by
Marietta Blau, which we will recall later in connection with the phenomenon of the
interaction of primary particles with the atmosphere. Moreover, a large impulse to
cosmic ray investigations also came from the first permanent installations of cosmic
detection equipment in the high mountains, for example on Mount Hafelekar, to
which Hess himself contributed [Hess1932].
Other activities partly carried out in Germany concerned the first observations
of the influence of atmospheric pressure on the intensity of cosmic radiation, by
Myssowsky and Tuwim [Myssowsky1926], two Russian physicists who collaborated
with Kolhörster. Tuwim, in particular, had also developed a numerical algorithm for
the interpretation of coincidence data between two Geiger counters arranged in a
telescopic configuration, based on their geometric properties. Additional studies on
the influence of atmospheric pressure were done by Steinke [Steinke1930].
It should also be remembered that in the mid-1930s Heisenberg also dealt with
the theory concerning the cosmic radiation, in particular by studying the processes
that led to the formation of showers [Heisenberg1936] and the penetrating power of
this radiation [Heisenberg1938].
The first Italian contributions to cosmic ray physics have been discussed
in several texts mentioned in the bibliographic notes [Bonolis, Bertolotti, De
Angelis, Marafatto] and in many articles [Scarsi2005, Peruzzi2007, DeAngelis2010,
Bonolis2011a, Bonolis2011b, DeAngelis2012, Spillantini2013, Bonolis2014]. As
far as Italy is concerned, one of the important contributions to the physics of cosmic
rays, as it is universally known, is that of Bruno Rossi (1905–1993), whose work and
results we will be able to use extensively in this review, and that we want to delimit
in this section to the more specific question of the nature of cosmic radiation. Bruno
Rossi had started his scientific activity after graduating in Bologna in 1927. At the
end of the 1920s he was already involved in activities concerning the study of the
cosmic radiation, also intrigued by the availability of new working tools for the study
of this radiation, such as the new Geiger counters and the associated coincidence
techniques. The first results of coincidence measurements, carried out by Bothe and
Kolhörster, had already been made available, as we have seen, in 1929, showing that
the particles capable of crossing two counters vertically overlapping and separated
by a thickness of a few cm of material with a high atomic number, like gold, could
not be electrons produced by the Compton effect. Bothe and Kolhörster believed
that these penetrating particles were already a component of the primary radiation,
a hypothesis that later on turned out to be incorrect, but which allowed considerable
progress to be made in understanding the overall components of radiation.
Starting from the knowledge of these first experiments by Bothe and Kolhörster,
Rossi began in a few months his own research program dedicated to the study of the
cosmic radiation in the laboratory, building and setting up new kinds of counters,
improving the coincidence circuits, developing a series of geometric configurations
with different counters and with different amount of shielding between the counters,
using magnetic fields for the deflection of the particles, thus obtaining, in the course of
a few years starting as early as 1930, a very remarkable series of results [Rossi1930a,
Rossi1930b, Rossi1931a, Rossi1931b, Rossi1931c, Rossi1932a, Rossi1932b,
Rossi1932c, Rossi1932d, Rossi1933a, Rossi1933b, Rossi1934a, Rossi1934b]. We
are lucky to be able to learn his working method and the results obtained in this
progressing activity not only from numerous comments on his work, already cited, but
from the author’s own words, through several textbooks and autobiographical papers
that the Italian physicist published later during his career [Rossi1964, Rossi1981,
Rossi1990] (Fig. 3.17).
It is interesting, for example, to read from Rossi’s own story, that the ideas
expressed by Millikan about the origin of cosmic radiation, as due to the fusion of
hydrogen atoms to form heavier atoms, had not affected him positively [Rossi1981],
but that at first, he too had shared the prevailing idea of a primary radiation consisting
of high-energy gammas. The first results of Bothe and Kolhörster, therefore, which
introduced an alternative experimental method for understanding the type of radi-
ation observed, had particularly intrigued him. The introduction of an electronic
coincidence circuit, which he designed [Rossi1930a], and which we will discuss
in detail in Chap. 5, allowed to perform coincidence measurements even between
more than two counters, making it possible to imagine and set up different kinds
of geometric configurations, with a time resolution better than that of Bothe and
Kolhörster’s original apparatus, which was 0.01 s.
Among the first results of these experiments, performed by means of counters in
coincidence, Rossi observed the first evidence of the effect resulting from placing
a layer of material above the two counters arranged vertically on top of each other,
rather than in the middle between the two: in the first case there was an increase in
the coincidence rate between the two counters. In 1930 Rossi had spent a short time
in Bothe’s laboratory in Berlin, discussing both the experimental techniques used
Fig. 3.17 Bruno Rossi (1905–1993). Source AIP Emilio Segrè Visual Archives
and the first attempts to observe a dependence of the intensity of cosmic radiation on
latitude, during an expedition to Northern Europe. Although the results of this first
expedition had not been very significant, Rossi was aware of the importance of this
type of measurements for the understanding of the corpuscular nature of the cosmic
radiation, familiarizing himself with Størmer’s results on the deflection of particles
in the terrestrial magnetic field, and theoretically studying the effect of a possible
East–West asymmetry [Rossi1930b], it too linked to the corpuscular nature of the
radiation.
A first measurement to study the East–West asymmetry, carried out in Arcetri,
did not initially give the desired results [Rossi1931c], but, carried out subsequently
at latitudes close to the equator and at a higher altitude, in Eritrea, it clearly showed
an excess in the intensity of the radiation coming from the West [Rossi1933b,
Rossi1934a, Rossi1934b]. However, similar results had previously been published
independently by Johnson [Johnson1933a, Johnson1933b] and by Alvarez and
Compton [Alvarez1933]. These results once again proved the corpuscular hypothesis
to be right, also showing—and this went against generally accepted ideas—that the
primary particles were positively charged. We will be able to return to the East–West
asymmetry later, as well as to the first evidence of coincidences between counters
placed on the same level at a certain relative distance, an effect which, although
observed, was not investigated in detail by the Italian physicist.
In 1931, during the Conference on Nuclear Physics held in Rome, Rossi, at the
invitation of Fermi, presented his results concerning cosmic rays, which denied
Millikan’s hypotheses about the nature of the cosmic radiation. We know from
Rossi’s own words that Millikan, who was also present at the Conference, along
with most of the well-known physicists of the time, did not like his contribution very
much, which he ignored even in the years to come, as he used to do. Rossi’s contri-
bution to this Conference was instead appreciated by Compton, who seems to have
decided to deal more with the problem after hearing his presentation. According to
Bruno Rossi, indeed, it was the Rome Conference that marked the beginning of the
debate on the corpuscular or photonic nature of cosmic rays. As we saw, this debate
took the form in the United States of a real conflict, which went beyond the strictly
scientific aspect, and involved the personalities of two Nobel laureates, Millikan and
Compton. In Europe, however, according to Rossi’s opinion, the debate between the
two positions had much more quiet tones and did not undermine respect, friendship
and acceptance of the ideas of other colleagues.
Rossi’s subsequent experiments, performed with a set of three counters in coinci-
dence (to reduce the amount of spurious coincidences), separated by a considerable
thickness of lead, highlighted the existence of particles so penetrating as to emerge
even after a meter of lead. Moreover, in those years Kolhörster had also done experi-
ments with counters placed in a salt mine, thus revealing coincidences originated by
extremely penetrating particles.
Experiments of the same kind, with different configurations of three counters and
lead absorber screens, arranged above and between the counters, were performed
by Hsiung at the University of Chicago [Hsiung1934], in order to distinguish the
coincidences due to showers and secondary particles produced in absorbers them-
selves from those due to primary ionizing particles passing through the counters.
However, the current interpretation of these experiments still associated these highly
penetrating particles with the primary radiation.
Among the other physicists who at that time carried out their activity, linked to
the physics of cosmic rays, in Italy, even if they later moved to other countries, the
name of Giuseppe (Beppo) Occhialini (1907–1993) is certainly worth mentioning.
Occhialini, after graduating in Florence in 1929, began his career with Bruno Rossi,
to move almost immediately to Great Britain, where he collaborated with Blackett
on the experiments done with the use of a cloud chamber, coupled to a coincidence
circuit, which will lead in 1933 to the observation of positrons in the cosmic radiation.
Subsequently, in the second half of the 1930s, Occhialini moved to Brazil, at the
University of Sao Paulo, an Institution that some European scientists had helped to
found, moving and carrying out their activities there, to form a school of Brazilian
physicists. Finally, after a further stay in Great Britain during the war years, Occhialini
returned to Italy in the early 1950s.
We may also remember the expedition with the airship Italy, organized by Umberto
Nobile in 1928, which brought on board a variety of scientific equipment, including
electroscopes for measuring air ionization. Scientists of various nationalities also
participated in this expedition, which unfortunately ended tragically on its way back,
including the Czechoslovakian František Běhounek (1898–1973), one of the few
survivors of the expedition and a cosmic ray physicist, which he later described also
in a book [Běhounek1929]. It is worth remembering that precision measurements of
the intensity of cosmic radiation at latitudes close to the North Pole (66°–82° N) have
been recently carried out by the EEE Collaboration [Abbrescia2020], using a scin-
tillation detector telescope installed aboard of a sailboat, as part of the PolarquEEEst
scientific expedition, as we will resume in a later chapter.
In Great Britain, in the second half of the 1930s, the Manchester school, with the
contribution of Blackett and the numerous physicists, including foreign ones, who
spent periods of study and research in that laboratory, made many significant contri-
butions to the study of cosmic radiation [Sekido1982]. Contributing to this was the
increasingly critical situation in Europe in the years leading up to the Second World
War, which saw the transfer, escape or exile of many Central European scientists, in
some cases to Great Britain, to stay there, or as an intermediate destination to the
USA. Although Blackett was able to return personally to a full research activity only
at the end of the conflict, his laboratory continued to be operative even in the previous
years and during the same conflict, often by foreign physicists, one of which was the
Hungarian physicist Janossy, whom we will mention later.
In France, the first studies of the cosmic radiation are associated with the names
of Pierre Auger and Louis Leprince-Ringuet, pupils of Jean Perrin and Maurice de
Broglie [Ravel2013, Degrange2013]. Among the first studies performed by the two
French physicists, in the early 1930s, are the measurements of the intensity of the
cosmic radiation during a two month voyage by ship, from Le Havre to Buenos Aires,
results that had confirmed those obtained by Clay about the presence of a charged
component in cosmic radiation, influenced by the Earth’s magnetic field.
We will discuss later the contribution of Auger and collaborators, in the following
years, which led to the observation of coincidence events between counters placed
some distance apart and, therefore, to the evidence of extensive air showers, a result
that will open a new path in the physics of cosmic rays. Subsequent studies were
carried out with the creation by Leprince-Ringuet and Chanson of a laboratory located
in the high mountains, at 3600 m of altitude (Aguille du Midi, Mont Blanc), equipped
with a large cloud chamber and an intense magnetic field.
Lajos Janossy (1912–1978), Hungarian physicist, was one of Kolhörster’s students
in Potsdam, where he carried out many measurements with him in the early 1930s,
using coincident Geiger counters in various geometric configurations [Kiraly2013].
Janossy later moved to Manchester, where he worked with Blackett’s group and
many other foreign physicists who spent time in that Institute, continuing to study
different aspects of the cosmic radiation (directional asymmetries, magnetic field
effect, properties of penetrating showers …). Janossy returned to Hungary a few
years after the end of the Second World War, contributing to create a group dedicated
to cosmic ray physics at the Physics Institute of the Academy of Sciences, of which
he became director.
Experimental activities in Sweden were carried out as early as the mid-1920s
by the astronomer Axel Corlin [Corlin1927, Corlin1928], who also participated in
the campaign to measure the intensity of cosmic radiation at latitudes close to the
North Pole [Bartelt2018]. In the context of the activities related to the physics of
cosmic rays in Northern Europe certainly it should be remembered, even if we will
have the opportunity to further discuss this aspect, the figure of Hannes Alfvén
(1908–1995), Swedish physicist, Nobel Prize in 1970, whose name is linked to
magnetohydrodynamics, which has numerous connections with the problem of the
transport of cosmic rays in interplanetary space.
Cultural exchanges between researchers from different countries, despite the diffi-
culties both intrinsic (due to the scarcity of means of communication and transport),
and linked to the particular post-war period, were of primary importance in deter-
mining the development of physical activities in cosmic rays also in non-European
countries, as well as in the USA.
In Japan, the first studies of the cosmic radiation date back to the mid-1930s
[Sekido1982, Nishimura2013], with the contributions of Y. Watase (1907–1978) and
S. Kikuchi (1902–1974) at the University of Osaka, concerning the development of
showers [Watase1937], and with the creation in Riken—by Yoshio Nishina (1890–
1951)—of a Laboratory dedicated to various aspects of modern physics (Nuclear
Physics, Cosmic Rays, …). The first group then directed its interests towards nuclear
physics with accelerators while Nishina’s group continued activities in cosmic ray
physics. The first contributions in this area were obtained through the construction of
a cloud chamber with a diameter of 40 cm—very large for that time—and a magnetic
field of 1.7 T, to study the energy spectrum of cosmic rays and measure the mass of
muons [Nishina1937]. Subsequent activities were also addressed to measurements
of the cosmic radiation in the subsoil, for the study of the decay of the pion, and to
the monitoring of the intensity of cosmic rays in different geographical locations,
activities that suffered an abrupt halt due to the Second World War and were able to
resume only several years after the war ended (Fig. 3.18).
An interest and some contributions related to cosmic ray physics are also present
in India [Tonwar2013], with the figure of Debendra Mohan Bose (1885–1975), a
physicist who had studied in Europe, in Great Britain, at the Cavendish Laboratory,
and later, as a researcher, from 1914 to 1919 in Berlin, with Erich Regener. Bose
organized several measurements in the late 1930s, with nuclear emulsions at large
altitudes, in the Himalayas. Subsequent contributions were made by Bhaba (1909–
1966), who had obtained his doctorate in Cambridge in 1933 on the absorption of
cosmic rays and who conducted numerous measurements of the intensity of the
radiation at high altitudes, with instrumentation mounted on board of balloons or
by means of the use of an aircraft. Measurements of the intensity of cosmic rays
at various latitudes were carried out in the years 1937–1938 by Piara Singh Gill
(1911–2002), on the occasion of numerous ocean crossings between Vancouver and
Tasmania, as part of the doctorate obtained with Compton in 1940. Returned to India,
he continued his activity in cosmic ray physics, especially at high altitudes, also
founding an observation laboratory in Kashmir at 2700 m above sea level. Finally,
in the years immediately following the Second World War, Vikram Sarabhai (1919–
1971) also made contributions, with activities concerning the effect of the Sun on
cosmic radiation.
Fig. 3.18 Japanese physicists from Yoshio Nishina’s group in Tokyo in 1943, on the occasion of
the awarding of a prize to Yukawa. Source AIP Emilio Segrè Visual Archives, Yukawa Collection
If the contributions of the German group of Bothe and Kolhörster, as well as those
of Rossi, had made use above all of Geiger-Müller counters for the observation of
penetrating particles, another instrument, the Wilson chamber, was at the origin of
other experiments and gave rise to further experimental evidence regarding the nature
of the highly penetrating radiation. We will describe Wilson’s chamber in a later
chapter. Here we may recall how the first and most important results regarding the
observation of these penetrating particles, through the Wilson chamber, were obtained
by the Soviet physicist Dmitry Skolbetsyn (1892–1990), who can be considered the
founder of cosmic ray physics in Russia [Bazilevskaya2014], even if the first studies
of cosmic radiation had already been carried out by Lev Myssovsky and Lev Tuwim,
whom we have already mentioned for their collaboration with the German group
of Kolhöster. Myssowsky and Tuwim had made numerous studies of the absorption
of radiation in water and lead, they had studied the angular distribution of radi-
ation and the effect of atmospheric pressure [Myssowsky1925, Myssowsky1926,
Myssowsky1927, Myssowsky1928]. After the untimely death of both, however, it
was Skolbetsyn who continued this activity for a long time.
As early as the mid-1920s Skobeltsyn had made observations with a Wilson
chamber immersed in a magnetic field, initially for the study of electrons generated
by the Compton effect. Among the tracks of these electrons, mainly of low energy,
Skobeltsyn had however also observed tracks of highly energetic particles, over
20 meV, which could not be due to the radioactive source of RaC that he was using
at that time [Skobeltsyn1927].
Skobeltsyn obtained hundreds of photos, even stereoscopic, of tracks observed
through a Wilson’s camera immersed in a magnetic field [Skobeltsyn1929]. Among
these, some dozen showed evidence of tracks due to highly energetic particles, and
some even evidence of the interaction of these particles with atoms, signalled by
double or triple tracks that departed from the point of interaction. Skobeltsyn could
perhaps have recognized in some of these tracks also the first evidence of the positron,
but the magnetic field used was too low to allow this observation, which was then
obtained by Anderson a few years later, using a higher magnetic field. It is true,
however, that Anderson himself recognized Skobeltsyn’s merit in having paved the
way for the observation of these events linked to the cosmic radiation (Fig. 3.19).
Skobeltsyn also spent a period of research in France, where, together with Auger,
he observed in 1929 some first evidence of correlated tracks. When the first systematic
observations of coincidences between different counters were made, which we will
Fig. 3.19 Dmitry Skolbetsyn (1892–1990). Source Alchetron.com

discuss later, Skobeltsyn had returned to the Soviet Union, continuing to keep alive
relations with foreign colleagues, so much so that in 1933 a conference was organized
in Leningrad with the presence of many Western physicists.
In the second half of the 1930s, with positions of responsibility at the Academy
of Sciences, he organized a series of activities in cosmic ray physics, including
observations at high altitudes, in the mountains (Mount Elbrus, 3700 m, Pamir at
3760 m and then up to over 5000 m), as well as by means of balloons with radio
data transmission. Skobeltsyn was also the author of the first textbook on cosmic
rays in Russian, which served as the basis for the researchers of the following era
and continued his research and organization until late in his life.
3.5 Protons as an Essential Component of Primary

Radiation?
In the second half of the 1930s, the belief that the primary radiation could not be made
by gammas became increasingly accepted, based on an accumulation of evidence
related to geomagnetic effects, experiments with coincident counters, and under-
standing of the properties of the showers produced by the interaction of a gamma or
a charged particle, as we have seen in the previous pages, also due to the diversified
contribution of many researchers in different countries of the world.
Several works by William Francis Gray Swann (1884–1962), an English physicist
who later moved to the United States and became director of the Bartol Foundation at
the Franklin Institute, published around the mid-1930s, discuss in detail the aspects
related to the nature of the primary radiation, progressively moving towards the
increasingly convincing hypothesis that a good percentage of the primary radiation
was made up of charged particles, and more specifically, of protons rather than
electrons or positrons. In an educational article of 1961, Swann, at that time retired,
summarizes some historical aspects of the physics of cosmic rays of that period
[Swann1961], recalling how the hypothesis of the cosmic radiation as constituted
by photons of high energy, however simple and capable to qualitatively reproduce
the trend of absorption with the thickness of the atmosphere crossed, had not proved
capable of interpreting in detail the various aspects observed. The existence of a
more penetrating (hard) and a less penetrating (soft) component, observed from
the beginning, with which to be able to interpret the absorption in terms of two
exponential functions, with different slopes, had led Millikan to admit the existence
of a multiplicity of components, those he had discussed in terms of the various nuclear
processes capable of producing different ranges of energies.
The corpuscular hypothesis, however, had always been considered as a possible
alternative. The difficulty in assuming the corpuscular hypothesis arose above all
from the fact that charged particles do not follow an exponential or quasi-exponential
law of absorption, as they are mostly associated with the concept of range (distance
crossed by the particles before depositing all their energy). But if we hypothesize
3.5 Protons as an Essential Component of Primary Radiation? 73
these particles as electrons, the range of electrons emitted by radioactive substances

is typically a few millimetres in solid materials, while it was evident that the hard
component could also cross a lead thickness of several meters. Even assuming that
the energy of the electrons present in the cosmic radiation was 1000 times greater
than that present in beta decays, the Earth’s magnetic field would have had to deflect
these electrons in a very consistent way, producing large variations in intensity in
different places on the Earth’s surface.
The variations due to geomagnetic latitude had indeed been observed, but the
amount of these variations was a few percent, at worst about 10% passing from the
equator to the poles. While on the one hand this evidence confirmed the corpus-
cular nature of the primary radiation, on the other hand this was not compatible
with the fact that a large part of the primary radiation was made up of electrons.
In a review work of 1935, Swann summarizes in detail the status of the corpus-
cular theory of the primary cosmic radiation [Swann1935], without yet explicitly
mentioning the hypothesis that primary particles are protons. A few months later,
however, in an article of February 1936, Swann himself briefly discusses the “pro-
tonic” hypothesis [Swann1936] starting from recent considerations by Compton and
Bethe [Compton1934, Compton1936b], and hinting at an experiment, just performed
on his proposal, to evaluate the presence of protons at sea level.
In 1934, in fact, Compton and Bethe, in a short article in Nature [Compton1934],
had discussed the existence of two different components (A and B) in the cosmic
radiation capable of reaching the surface of the Earth, and of a third component
(C), present in the upper part of the atmosphere, as evidenced by the results obtained
aboard balloons launched into the stratosphere. Component A, of greatest importance
at the top of the atmosphere, is associated with photons, while component B with
electrons and positrons. Based on considerations about the loss of energy due to
radiative processes of the different known particles, Compton and Bethe suggested
that assuming that the component C consisted of protons was the most natural choice,
able to provide a coherent explanation of most of the known phenomena.
The results of the experiment were indeed published in the same year, 1936
[Montgomery1936], but essentially concerned the possibility that the protons could
also be found among the components present at sea level. The result is that the
percentage of protons at sea level was established to be a small percentage, between
5 and 12%, depending on the assumptions made. These experiments, based on the
conclusions drawn, however, do not exclude that protons are the primary particles,
capable of giving rise to the various components then observed at sea level.
The hypothesis of a primary radiation as constituted by protons, able to correctly
interpret most of the observed phenomena, is increasingly evident in a series of
communications and works published around 1939–1941, for example those of
Swann [Swann1939a, Swann1939b, Swann1940, Swann1941a, Swann1941b] and
by Carlson and Schein [Carlson1941]. Swann summarizes the status of recent
knowledge by stating that there is only one type of primary radiation, charged and
predictably made up of protons; that this primary radiation has an energy distribution
with a lower limit given by the possibility of penetrating the Earth’s magnetic field;
that protons—by means of mechanisms still unknown at the moment—are capable
of producing, upon their entry into the atmosphere, a certain number of mesotrons (µ
mesons); that the mesons produced by low energy protons, being emitted almost at
rest, given their very short average life, about 2 µs, as we will see later, will not have
the possibility to penetrate into the atmosphere, but will travel only a few hundred
meters and will be distributed on average in all directions; that the mesons produced
by protons of higher energy will be able to penetrate deeper into the atmosphere,
producing electrons and neutrinos, which will be emitted mainly in the direction of
motion of the original particles. Similarly, Carlson and Schein [Carlson1941], recog-
nizing that a model was still missing to describe the mechanisms of interaction of
the primary protons, assumed that an “explosion” occurred in which the proton lost
all its energy creating a certain number of mesons, multiplicity that increased with
the energy of protons.
The hypothesis of protons as the main constituents of primary cosmic rays increas-
ingly took shape as a natural choice. This despite the fact that the overall produc-
tion mechanisms of the particles produced in the atmosphere were only clarified
years later. Again in 1941, for example, Giuseppe Cocconi (1914–2008), an Italian
physicist who worked in his country of origin before moving to the USA in 1947,
in a short note in Physical Review [Cocconi1941], in confirming that the hypoth-
esis of primary protons is the most reasonable to explain the generation of the
different particles, including the “mesotrons” (µ mesons), as well as the gammas
and electrons/positrons, admits that the mechanism of production is still unknown.
This was therefore the situation in the early 1940s: the acceptance of a proton
nature of the primary radiation, more and more consistently accepted, even in the
absence of a detailed description of the mechanisms by which the primary protons
were able to ultimately produce all particles present in the atmosphere, down to the
sea level.
Chapter 4
New Particles and Their Links
with the Cosmic Radiation
Abstract Since its beginning, the study of the cosmic radiation has been undoubt-
edly linked to that of the elementary particles. The discovery of new particles in this
context and the understanding of their properties received a strong contribution from
cosmic ray physics. This chapter briefly recalls the link between these two important
areas, especially the discovery of the positive electrons in the cosmic radiation, made
through the observation of their tracks in the cloud chambers, and the first evidence
for the existence of particles with a mass intermediate between that of electrons and
that of protons. Further observations of cosmic ray tracks by visual methods led in
those years to the study and identification of the various mesons, and especially to the
understanding of the properties and role of pions and muons in the cosmic radiation.
A discussion concerned with first estimates of the muon mass by these experiments
is further discussed in Appendix E.
4.1 The Discovery of New Particles and the Links

with the Understanding of Cosmic Radiation
As Bruno Rossi pointed out [Rossi1964], the main problems related to the physics
of cosmic rays, particularly between the 1930s and 1940s, were twofold: on the one
hand, the attempt to understand the nature of the cosmic radiation, both primary and
secondary; on the other hand, the classification of the particles existing in nature.
The two issues were obviously closely linked to each other, and any progress made
in one of the two areas would have greatly facilitated the understanding of the other.
It was already established that the radiation capable of propagating in the Earth’s
atmosphere had at least two components, one "soft", being absorbed in a few cm of
lead, the other more penetrating, or "hard", crossing even meters of lead. However,
the nature of these two components was not established with certainty, in particular
that of the more penetrating component, the other being predictably associated with
electrons. The very existence of atmospheric showers, which we will discuss later,
in a special chapter, had been highlighted in those years, but neither the mechanism
by which the showers began nor their detailed composition, in terms of particles and
energy distribution, had been clarified.

https://doi.org/10.1007/978-3-031-24762-0_4
76 4 New Particles and Their Links with the Cosmic Radiation
A partial answer to these questions came, as well known, from the discovery of new
particles, which took place in the context of the study of the cosmic radiation, which
at that time was the only high-energy physics laboratory able to provide information
on the processes involved at energies significantly higher than those of radioactivity
phenomena.
We mentioned the first observations of the tracks left by energetic parti-
cles, obtained by Skobeltsyn with the aid of cloud chambers, as early as 1929
[Skobeltsyn1929]. These observations were extended, by means of apparatus using
more intense magnetic fields, in a series of experiments carried out since the 1930s
by Anderson and Millikan. The results of these experiments were manifold.
In one of the first series of observations it was established that the particles present
in the observed cosmic radiation had—almost all of them—the same elementary
electric charge, as they produced the same ionization density as that produced by fast
electrons [Anderson1932a]. In this first work Anderson, using a Wilson expansion
chamber with a magnetic field up to a maximum value of 21,000 gauss, clearly
observed the presence of tracks that curved in the opposite direction, and taking
into account the fact that they came from above, it turned out to be possible to
assign a negative or positive charge to each track. The evidence of these positively
charged high-energy particles was interpreted in that work as being due to protons,
while those negatively charged were identified as electrons. The analysis of these
tracks also allowed to extract an angular distribution of the tracks themselves, which
showed a prevalence for those emitted at small angles with respect to the vertical,
with a dependence similar to cos2 θ (Fig. 4.1).
Fig. 4.1 Angular distribution of the tracks measured by Anderson in one of his first works with the
Wilson chamber. The data obtained, represented by the blue symbols, are compared with a cos2 θ
dependence (red points). Data extracted from [Anderson1932a]
4.1 The Discovery of New Particles and the Links with the Understanding … 77
Figure 4.2 shows one of the images, obtained with a magnetic field of 12,000
gauss, which Anderson interpreted as evidence of the passage of a 27 MeV negative
electron and (probably) of a 450 MeV proton. The energies were estimated using the
value of the magnetic field used and extracting the radius of curvature of the track
from the photo. With a maximum magnetic field of 21,000 gauss, Anderson was
able to measure even tracks that had a radius of curvature of 700 cm, and therefore a
magnetic rigidity of 1.5 × 107 gauss cm, corresponding to kinetic energies of several
GeV for both electrons and protons (See Appendix D).
These particles, therefore, had a wide distribution in energy, with values that far
exceeded 1 GeV. In these first measurements it was possible to observe maximum
energies of about 5 GeV, a limit which later, by other authors, was brought to 20 GeV.
However, the analysis of the ionization density of these positive tracks soon led
to the exclusion of the fact that they were protons: the energy loss per unit length
of these particles was compatible with a mass much smaller than that of the proton,
for example with that, enormously smaller, of the electron. Anderson reached this
conclusion in a very short time, so much so that both in the preliminary article
published in Science at the end of 1932 [Anderson1932b] and in the more complete
article published at the beginning of 1933 [Anderson1933a] the presence of these
tracks of positive charge was associated with the first evidence of the existence of
positive electrons, or positrons.
Fig. 4.2 One of the images obtained by Anderson in his early works [Anderson1932a], showing
evidence of two tracks of opposite curvature, the first interpreted as a negative electron, the second
as a high energy proton. Figure reproduced from the work of Carl D. Anderson, Energies of cosmic
ray particles, Physical Review 41(1932)405. Copyright (1932) by the American Physical Society,
License No. RNP / 22 / MAY / 053614
This evidence was obtained by placing a lead plate in the middle of the observation
chamber, in order to observe and measure the radius of curvature of the track before
and after crossing the lead plate, which would have produced a loss of energy.
In some way, this arrangement also made it possible to establish whether the
particles came from above or from below, in which case the sign of the curvature of
the track had to be interpreted in the opposite way. While the results also highlighted
the presence of tracks coming from below, it was possible to establish with certainty
that the positive tracks could not be due to protons, but rather to positively charged
electrons. A proton with the same magnetic rigidity, in fact, would have given rise
to a much greater ionization density than that observed and would have had a much
lower range (maximum distance travelled before completely losing its energy).
Anderson’s observations were soon confirmed also by Blackett and Occhialini
in England [Blackett1933a, Blackett1933b], who used a chamber controlled by two
coincident Geiger counters to signal the passage of tracks [Blackett1932]. In a very
short time, the idea of the existence of positive electrons was considered as completely
"natural". Already in these works of 1933 Blackett argues almost for granted that low
energy positrons are easily produced by gamma radioactive sources, when gammas
interact with matter, or even emitted themselves as a result of nuclear transmutation
processes. Blackett refers to works by Anderson and Neddermayer [Anderson1933b],
by Curie and Joliot [Curie1933] and by Meitner and Philipp [Meitner1933], all
published in 1933, which show that in a few months the observation of positrons in
natural phenomena had become very common (Figs. 4.3 and 4.4).
The observation of the positive electron was related to Dirac’s theoretical consider-
ations, initially considered as difficult to understand, given the assumption of particles
with states of negative mass and energy [Dirac1928]. Dirac’s theory contemplated the
existence of two states of charge of the electron, which the observations of Anderson
and Blackett and Occhialini confirmed.
However, assuming the existence of these positive electrons allowed to explain in
a more direct and natural way a variety of otherwise unclear phenomena; for example,
the fact that an equal number of negative and positive electrons were produced in the
showers, by the more energetic photons, or the fact that positrons had a very short life
in matter, because once produced, they annihilated rapidly with electrons of negative
sign, producing photons, based on the equivalence between mass and energy.
The process of producing electron–positron pairs, in which the energy of the
incident photon was converted into the mass and kinetic energy of the pair, still needed
to be understood in detail. However, already in Blackett’s work [Blackett1933b], he
discusses the number of pairs produced, in terms of the absorber material and of the
gamma energy, observing that it grows as both quantities increase, and he refers to a
work, by Oppenheimer and Plesset [Oppenheimer1933], in which the first estimates
of what we can call the pair production cross section were reported.
A more complete theory of the interaction of gammas with matter, which also
included the processes of creating e+ e− pairs was then formulated by Bethe and
Heitler in 1934.
Before the formulation of this theory, it was already known that fast electrons,
slowed down in matter, produced photons (Bremsstrahlung radiation), but that this
Fig. 4.3 The historical image obtained by Anderson in 1933, showing evidence of a positive track,
interpreted as a 63 MeV positron that loses energy as it passes through a 6 mm thick lead plate,
positioned in the middle of the chamber, emerging as a 23 MeV positron [Anderson1933a]. The
high range shown by this track is incompatible with that of the proton. Figure reproduced from the
work of Carl D. Anderson, The positive electron, Physical Review 43(1933)491. Copyright (1933)
by the American Physical Society, License No. RNP/22/MAY/053615
radiative process represented only a small part of the overall process of energy loss
for such electrons, considering that the collisional mechanism, the energy loss by
ionization, was predominant. The new calculations by Bethe and Heitler, based on
quantum and relativistic theory, instead predicted different results, in particular that
these radiation losses increase with the energy of the particle, while those due to
ionization processes decrease rapidly with the increase of energy. From a certain
energy onwards, the loss of energy by radiation equals and finally exceeds that by
ionization. This critical value is different for the different materials, amounting to
about 10 MeV in lead, and to about 100 MeV in air. Furthermore, the theory of
Bethe and Heitler predicted that the energy losses due to radiation were enormously
greater for light particles, such as electrons, than those to which heavy particles, such
as protons, were subject, due to a factor including the squared mass of the particle.
These losses were also expected to be greater in materials containing high atomic
number elements.
For the processes concerning the production of electron–positron pairs, the theory
of Bethe and Heitler finally predicted that the cross section of this process increases
with the atomic number of the element, and that starting from the threshold value
(1.02 MeV) grew rapidly with the photon energy, until it reached an approximately
constant value, while the cross sections of the other processes through which photons
Fig. 4.4 Carl D. Anderson (1905–1991). Source AIP Emilio Segrè Visual Archives, Weber
Collection, W. F. Meggers Gallery of Nobel Laureates Collection
interact with matter—the photoelectric process and the Compton process—were

characterized by a cross section that decreased with the energy of the photons.
Following this theory, it appears that low-energy photons are mainly absorbed by
the photoelectric and Compton effect and, subsequently, at higher energies, through
pair-production processes. The energy at which the pair-production process begins
to become more likely than the Compton process is about 5 MeV in lead, and about
20 MeV in air. Figure 4.5 shows a quantitative example of the cross sections for
these processes, evaluated for air (assumed for simplicity as a mixture consisting of
80% Nitrogen and 20% Oxygen), as a function of the photon energy, from 10 keV
to 1 GeV.
Now, the fact that high-energy electrons have to interact with high probability by
means of radiative processes was in contradiction with the fact that the particles of
cosmic radiation, imagined up to that moment as electrons or positrons, were capable
of crossing even a lead thickness of one meter or more, so that Bethe and Heitler
Fig. 4.5 Cross section of photoelectric effect, Compton scattering and creation of electron–positron
pairs in the interaction of photons with air. Data extracted from the National Institute of Standards
and Technology [NIST] database
themselves initially suspected that this theory and the same quantum hypotheses on
which it was based, were not valid for very high-energy electrons.
However, it soon became clear, by analysing the structure of electromagnetic
showers, created by high-energy photons by means of the different processes
(creation of pairs, Compton effect, emission of bremsstrahlung radiation), that the
theory was able to explain satisfactorily this phenomenon and that could be consid-
ered substantially correct. A major part of the secondary cosmic radiation present
in the atmosphere could therefore be interpreted on the basis of this theory: in the
cosmic radiation there were photons and electrons, even very energetic, which gave
rise to these showers (cascade processes initiated by an electron or by a high-energy
photon). This component represented the so-called “soft”, or less penetrating, compo-
nent of the cosmic radiation. However, it remained to explain the presence of a much
more penetrating component, which could not be constituted by electrons.
The very discovery of the positron, as one of the constituents of the cosmic
radiation observed at sea level, could not alone explain the set of observations and lead
to a picture as complete as possible of the phenomena that occurred in the atmosphere
due to cosmic rays. The other experimental evidence that made a great contribution
to understanding the physics of cosmic rays was, as we know, that related to the
existence of other charged particles, of intermediate mass between that of electrons
and that of protons, the mesons.
Assuming, as it was initially done, that Bethe and Heitler’s theory was not valid
for large electron energies turned out to be an unreasonable choice. In those years
already several physicists, such as Williams and von Weizsacker had confirmed that
this theory gave correct results even at energies well above some GeV, and moreover
the strongly penetrating particles observed had energies both above and below the
GeV. For a particle not to radiate it had to be a particle heavier than the electron,
and at that time only protons were known as particles having a mass greater than the
electron.
But the hypothesis that they were protons (both positive and negative, which
although foreseen by Dirac’s hypotheses, did not correspond to any observation
made up to that moment) encountered difficulties, emerging from the observations of
Anderson and Neddermayer [Neddermayer1937, Neddermayer1938]. The particles
in question, in fact, as they ionized less than protons, should have a lower mass,
but, on the other hand, not radiating as much as the electrons, their mass must have
been larger than that of the electrons. From these results Anderson and Neddermayer
concluded in 1937 that they had observed with certainty particles of intermediate
mass between that of the electron and that of the proton.
Initially the measurements carried out by Anderson and Neddermayer were unable
to accurately establish the mass of these particles, because the measurable ionization
density in relation to the observed magnetic rigidity did not allow to distinguish
masses of intermediate value between electrons and protons. However, in a short
time, further observations by Street and Stevenson, in 1937 [Street1937], aimed at
the possibility of detecting low-energy particles (therefore more distinguishable),
made it possible to extract an estimate of the mass, around 200 times the mass of
the electron (See Appendix E). These particles, as it is known, were initially given
different names, including yukon, mesotron and meson. Later, with the discovery of
other mesons, this specific class of particles was given the name of μ mesons. More
precise measurements of the mass of these mesons were made by several groups in
the following years, but still in 1950, the mass estimated by Brode’s group was (206
± 2) electron masses.
The chapter concerning the nature of these particles, constituting the penetrating
part of the cosmic radiation, was destined, however, to remain complex for several
years, for various reasons. On the one hand, an initial erroneous interpretation of
the meaning of this particle, which was first identified with the particle theoretically
hypothesized by the Japanese physicist Yukawa in 1935, on the other hand the need
to understand the behaviour of these mesons and their decay.
Yukawa, in arguing that nuclear forces must be associated with some kind of
particle, just as photons are associated with electromagnetic forces, foresaw the
existence of particles associated with the nuclear field; because of the range of action
of these forces they were supposed to have a mass equal to some hundreds of elec-
tronic masses. The discovery of μ mesons was therefore associated, as it seemed
natural, with the theoretical prediction by Yukawa, who also hypothesized a decay
for these particles, hence one of the assigned names, yukons.
4.2 Properties of μ-Mesons 83
4.2 Properties of µ-Mesons
The evidence of the existence of these particles was soon accompanied by numerous
studies aimed at studying their properties, in particular their mass and their decay.
The number of particles detected at different heights can be used to get information
on the possible decay of the mesons. Measurements of the number of particles present
at various altitudes in the atmosphere had been carried out with accuracy by various
groups in England, France, Germany and Italy around 1936–1937. Already from the
first results it was understood that in comparing the paths crossed, corresponding to
the same mass but to different materials, the absorption of cosmic rays was appar-
ently different. This happens for example for air and water, in which case the same
surface density correspond to a very different linear thickness. Still imagining that
the particles were to be associated with the Yukawa particles, it was expected that
these should decay spontaneously.
In 1938, the German physicist Helmut Kuhlenkampff (1895–1971) found that
this property could naturally explain the anomalous absorption of cosmic rays in the
air. If we consider for example a very small thickness of water and imagine that the
mesons have an average life of the order of microseconds, none of these will decay
in such a limited space. However, a large fraction of mesons will decay through an
equivalent thickness of air of about 80 m instead of 10 cm of water; absorption is
therefore also governed by the effect of spontaneous decay. Rossi also contributed to
these studies, in the autumn of 1938 and in the spring of 1939, studying the instability
of the meson through a direct comparison between the absorption of mesons in air
and in a dense material. These experiments were mainly done after Rossi’s move to
the United States in 1939 at Compton’s invitation.
The experimental apparatus used by Rossi consisted of a certain number of Geiger
counters in coincidence, separated by a layer of lead and graphite sufficient to elim-
inate the soft component. With this apparatus, measurements were carried out at
different altitudes, up to 4300 m, and Rossi understood that the absorption in the
material was much lower than that observed in the air. The evident greater absorption
was due to the decay of the mesons flying through the atmosphere.
This measurement also made it possible to extract a rough estimate of the average
life of the mesons, which for low energies was estimated to be about two microsec-
onds. The measured average life of these mesons due to relativistic effects is however
greater than the average life of the mesons at rest. Taking into account the relativistic
formulas for time dilation, it can be estimated, for example, that for mesons of kinetic
energy 500 MeV the average life measured is almost 5 times greater than their average
life at rest.
Measures of the average life for mesons of different momentum were reported
by Rossi himself in 1941 [Rossi1941]. Similar experiments were carried out in the
following years by various groups both in the United States and in Italy, despite the
difficulties, in the Italian territory, during the war, to maintain research activities.
Figure 4.6, taken from the original work of Nereson and Rossi [Rossi1943] shows
the disintegration curves of mesons in various materials (Pb, Al, brass) compared
with the curve obtained in the absence of absorber material.
As to the mode of meson decay, this was established by the first photographs of the
decay obtained by Williams and Roberts in England using condensation chambers
[Williams1940]. One of these photos showed the track of a meson stopping in the
gas of the chamber and a further track departing from the end of the previous track;
this additional track was identified as that of a positive electron, and this showed, as
expected, that the decay of the meson produced an electron.
Further experiments were carried out in 1941 by Franco Rasetti (1901–2001),
an Italian physicist from the “Via Panisperna” group, who also emigrated, first to
Canada and then to the USA, after 1939. Rasetti used a coincidence circuit with
Fig. 4.6 Disintegration curves of mesons in different materials [Rossi1943]. Figure reproduced
from the work of N. Nereson and B. Rossi, Further measurements on the disintegration curve of
the mesotron, Physical Review 64(1943)199. Copyright (1943) by the American Physical Society,
License No. RNP/22/MAY/053654
4.3 The Discovery of the Pion 85
different time windows, to detect both the meson and the electron within a certain
time interval [Rasetti1941]. With this apparatus, mean lifetimes of the mesons at rest
equal to about 1.5 microseconds were measured. Finally, in 1942–1943, Rossi and
Nereson [Rossi1942, Rossi1943] performed a very detailed experiment on the decay
of mesons, using a set of Geiger counters and absorbers, using an electronic circuit
to measure the delay between the arrival of mesons and the emission of electrons in
a more accurate way. The result is a decay that follows an exponential law over time,
exactly similar to that of the decay of nuclei. The trend of these data will also allow
to extract a much more precise value of the average life of the mesons, equal to 2.15
microseconds, with an estimated uncertainty of 0.07 microseconds. More precise
measurements were obtained only many years later, using particle accelerators.
4.3 The Discovery of the Pion
Although several years later, even the evidence of the existence of a further meson,
the pion, which turned out to be the one predicted by Yukawa, was closely related to
the issues discussed in the previous sections about the nature of the particles present
in cosmic radiation and the processes that take place in the Earth’s atmosphere.
One of the considerations that led to the evidence of the pions was the different
behaviour of positive and negative muons. That they should behave differently in
interacting with matter had already been discussed around 1940 by the Japanese
physicists Tomonaga and Araki. The different hypothesized behaviour is linked to
the fact that a positive meson, due to its electric charge, cannot get very close to
an atomic nucleus, which is also positively charged. Therefore, the positive meson
will be subject to spontaneous decay. In the case of negative mesons, however, the
possibility of being captured by a nucleus before decaying represents a further way
for the meson to "disappear", with an apparent decrease in their average life. The
probability of capture of negative mesons, being linked to the electric charge, is
expected to be greater for heavier nuclei, with a high atomic number.
A first demonstration of the correctness of this interpretation was obtained by
Rasetti [Rasetti1941], with the same apparatus that had served to determine the
average life of the muons. By evaluating the number of muons stopped in a block of
iron and the number of decay electrons emitted, it was possible to estimate that only
a fraction of the overall number of muons, about 50%, decayed into electrons, while
the others presumably underwent a process of nuclear capture.
In the following years, experiments performed by Conversi, Pancini and Piccioni
with an apparatus capable of separating the positive muons from the negative ones,
a sort of magnetic lens, established that in the case of iron almost all the negative
mesons underwent the capture process, while the decay electrons were emitted only
by positive mesons. In the years around 1947–1948 these experiments were extended
to various materials used as absorbers, showing that the percentage of negative muons
that disappeared by nuclear capture was very small in elements with low atomic
numbers, while it became predominant in elements with high Z, such as reasonably

expected, based on the charge of the mesons.
However, it was unclear why, even in light elements (for example carbon), the
negative mesons in a certain (small) percentage decayed "normally", that is, without
being captured by the nuclei, since the average lifetimes involved were of the order
of microseconds, a time enormously greater than that typical of nuclear capture
processes.
The conceptual difficulties arising from this experimental evidence were twofold.
On the one hand, if muons were the particles predicted by Yukawa’s theory, they
would have to interact strongly with the nucleons (protons and neutrons) of the
nuclei, while this interaction seemed to be weak, since muons could sometimes
survive even for microseconds and then decay. On the other hand, if muons—as
was still believed—were produced by the interaction of the primary particles of the
cosmic radiation, therefore in nuclear processes with an extremely short duration
(10–22 s), why the inverse process, that is their absorption by the nuclei, did take such
a long time?
The answer to these questions came from a series of experiments carried out with
another experimental technique, which played an important role in the physics of
cosmic rays, that of nuclear emulsions, photographic films capable of recording the
passage of an ionizing particle and measuring the ionization density along the track.
A more detailed discussion of this technique will be addressed in the next chapter,
dedicated to the developments of the different experimental techniques that have
contributed to this field of scientific knowledge.
These experiments were performed by the Brazilian physicist Cesare Lattes
(1924–2005), by Giuseppe Occhialini, already mentioned, and Cecil Powell (1903–
1969), in England, exposing some nuclear emulsions to the flux of cosmic rays in
the high mountains. A series of introductory works, published in Nature in 1947,
concerned the analysis of the different types of tracks observed, which had already
shown evidence of the emission of mesons (understood as particles of intermediate
mass between that of the electron and that of the proton). Soon after, a series of
data collected on the Bolivian Andes, at 5500 m altitude, made it possible to estab-
lish in detail the existence of two different types of mesons, having different mass
[Lattes1947a], and the fact that some of these mesons could be produced locally,
through a sort of explosive disintegration of the nuclei [Lattes1947b]. Powell received
the Nobel Prize in 1950 for the use of photographic techniques in the study of mesons
(Fig. 4.7).
The observed tracks were compatible with the hypothesis that a meson of greater
mass decayed into a meson of lesser mass, corresponding to the already known muon.
The Yukawa particle, the π meson, or pion, was associated with this meson of greater
mass. Such π mesons could be produced in abundance in the nuclear interactions of
primary high-energy particles with the nuclei of matter. While in the original Yukawa
framework these mesons had to decay into electrons, it was seen instead that the π
mesons decay into μ mesons, and these in turn into electrons.
Because of their charge, either positive or negative, the same considerations also
apply to pions, about the possible capture by the nuclei of negative pions. And in
4.3 The Discovery of the Pion 87
Fig. 4.7 Giuseppe Occhialini and Cecil Powell in a conversation on Como Lake. Source American
Institute of Physics (AIP) Emilio Segrè Visual Archives
fact, for negative pions we do not observe the decay process in matter, but rather
an interaction that produces a star of particles, following the disintegration of the
nucleus.
If pions are not observed at sea level—or only a very small fraction of them -
this means that they decay much faster than muons. The lifetime of pions, as well as
their mass, were determined with great precision only later, with the first experiments
performed with particle accelerators. As for their mode of decay, since they produce
a muon, the conservation laws of energy and momentum allow to establish that
another particle, of practically zero mass, should also be emitted in the process,
which led to the hypothesis of a photon or a neutrino. The conclusion was that in the
decay of the pion, together with the muon, a neutrino was also emitted, as no photon
associated with this decay was observable. On the other hand, in the decay of muons
into electrons, at least two "invisible" particles had to be emitted, since the energy of
the decaying electrons was distributed over a wide range of values. These invisible
particles were also neutrinos.
In summary, therefore, the decay patterns of the two types of mesons found in the
cosmic radiation could be represented by the following processes:
π + → μ+ + υμ
π − → μ− + υμ
μ− → e− + υe + υμ
μ+ → e+ + υe + υμ
where the symbol υ denotes the antineutrino. For the decay of muons, in addition
to the electron, a neutrino and an antineutrino are produced. The average life of the
charged pions was estimated, from results obtained with particle accelerators, to be
about 26 ns.
From the first experiments carried out by means of particle accelerators, at the end
of the 1940s, the existence of a neutral meson was also confirmed, which according
to the mass could be classified as a π meson, and which was produced in abundance
in the same nuclear interactions that produced the charged pions. However, the latter
decayed mainly into two photons, with an enormously shorter average life, about
108 times shorter than that of charged pions.
Summarizing the situation, at the end of the 1940s, the essential components of
the secondary cosmic radiation and the processes that produced these components
were substantially defined. This period coincides with the beginning of the use of
particle accelerators for the study of nuclear processes, which until that moment
had only been investigate in nature through the processes generated by the cosmic
radiation.
4.4 The Discovery of the Neutron
Starting from 1928 Bothe and Becker had carried out experiments in Berlin, using
alpha particles emitted in the radioactive decay of Polonium to bombard samples of
various materials (Lithium, Beryllium, Boron, Fluorine) and to observe the conse-
quent production of a strongly penetrating radiation, emitted following this bombard-
ment. They interpreted this phenomenon as due to the emission of gamma rays
produced by the capture of alpha particles by these light nuclei, publishing the
results in 1930 [Bothe1930b], and also presenting them at the 1931 Congress of
Rome, already mentioned several times, where it raised further interest in Irène
Curie (1897–1956) and Frédéric Joliot (1900–1958), who did similar experiments,
finding that the penetrating power of this radiation was even greater than what Bothe
hypothesized [Curie1930, Joliot1931]. Even in this case the radiation was interpreted
as an extremely energetic gamma radiation. Subsequently, the same authors observed
the emission of protons by hydrogenated materials bombarded by this penetrating
radiation [Curie1932a]. However, further measurements showed that the character-
istics of this radiation were incompatible with the assumption of gamma radiation
[Curie1932b].
Shortly after, at the Cavendish Laboratory the English physicist James Chadwick
(1891–1974) (Fig. 4.8) repeated similar experiments, clearly demonstrating [Chad-
wick1932a, Chadwick1932b] that it could not be a gamma radiation, but rather
the emission of particles of unit mass (similar to that of protons) and zero charge,
4.4 The Discovery of the Neutron 89
Fig. 4.8 James Chadwick (1891–1974). Source: AIP Emilio Segre Visual Archives, W. F. Meggers
Gallery of Nobel Laureates Collection
what could be indicated with a neutron. The same Joliot and Curie, after an initial
skepticism, soon agreed with this hypothesis [Curie1932c].
Although Bothe, as well as Curie and Joliot, were not aware of the possible
observation of a new particle, of zero charge, on the contrary this observation was
within the expectations of other researchers. Chadwick in particular, following ideas
already expressed previously by Rutherford, was strongly convinced of the possible
existence of such a particle.
As it is well known, the discovery of the neutron contributed to rewriting much
of the nuclear physics known at that time, from the internal structure of nuclei to
beta decay phenomena. In this context we shall not deal with the consequences that
the introduction of the neutron brought to nuclear physics, which are adequately
dealt with in Nuclear Physics textbooks or in historical articles on the subject
[Nesvizhevsky2017]. Let’s just remember that while the proton appears to be stable,
with a lower limit on its mean life enormously greater than the age of the universe, the
neutron, on the other hand, is subject to decay, with an average life that was measured
starting from 1948, but which is still the object of precision measurements, with a
recent estimate of (879.4 ± 0.6) s [PDG]. This value implies that neutrons cannot
constitute a component of the primary radiation in the case of distant sources, since
they would decay quickly compared to the time needed to travel distances between
two close stellar systems. However, high-energy neutrons could in principle come
to Earth from distances comparable to the size of the Solar System. For example,
neutrons of the order of GeV would have free average paths of some astronomical
units (Sun-Earth distance) and correspondingly greater for neutrons of higher energy.
We will see later that neutrons also represent an important component of the cosmic
radiation present in the atmosphere.
Chapter 5
Developments of the Techniques
for the Detection of Cosmic Rays
Abstract This chapter is devoted to a review of the different techniques which were
employed for the observation of the cosmic radiation. After the improvement and
use of the electroscopes, even by means of automated systems of recording, strong
developments in the observation techniques came both from Wilson cloud cham-
bers, producing visual evidence of the passage of tracks, and from detectors based
on single-event counting, allowing to observe the passage of even a single ionizing
particle. Ionization chambers, proportional counters and Geiger-Muller counters
played a leading role in this respect, pushing the study of the cosmic radiation to an
unprecedented level of accuracy and allowing for new class of experiments, such as
those provided by the coincidence technique, widely employed after its introduction
at the end of 1920s. Nuclear emulsions and the early detectors based on scintillating
materials also contributed to the technical developments of particle detectors. All
these techniques are here discussed with ample references to the original papers
which marked the history of such detectors.
5.1 Introduction
From the first evidence of the existence of cosmic radiation, in the first decade of the
1900s, up to the 1940s, the detection of high-energy particles found in the secondary
cosmic radiation made use of different experimental techniques, mostly based on the
phenomenon of ionization that charged particles produce by collision mechanisms.
In this Chapter we will briefly summarize the techniques mainly used in those
decades, as they emerge from reading the original articles and results of the period.
An important aspect to keep in mind when discussing the working principles and
the overall performance of these detectors is their sensitivity, i.e., the capability to
signal even a small number of ions produced in their volume in a certain time interval
and therefore detect even the arrival of a few ionizing particles. From this point of
view, a basic feature is related to the possibility of detecting individual events, that
is, events due to the passage of a single particle, rather than the total number of
particles arriving on the detector in a given time interval. While the first measuring
devices were essentially detectors that integrated the flux of incoming particles over

https://doi.org/10.1007/978-3-031-24762-0_5
92 5 Developments of the Techniques for the Detection of Cosmic Rays
long intervals of time, the use of the first “counters” made it possible to identify
and count individual events, paving the way for the possibility of operating several
counters in coincidence and therefore of highlighting also qualitatively new physical
phenomena.
5.2 From Wulf’s Electroscopes to Automatic Recording

Equipment
As we discussed in Chap. 1, reporting the first observations of the cosmic radiation,

most of the measurements carried out at the beginning of cosmic ray physics mostly
involved the use of electroscopes, instruments capable of signalling the ionization of
air, which in turn produced the discharge of the device.
As already mentioned, the simplest version of an electroscope consists of two thin
metal sheets, electrically connected to an external conductor. The charge initially
transferred, which causes the plates to diverge, is not maintained, but after some
time it is neutralized, following the ionization phenomena in the air, caused inside
the instrument by the passage of the ionizing radiation.
We have already seen how many experimenters in that period contributed to
significant improvements to the basic idea of the electroscope, starting with Wulf,
who replaced the original thin leaves with two metallized quartz fibres, placed in
tension. The divergence of the fibres could be observed with greater precision using
a small microscope. These were the electroscopes that Wulf brought with him during
the well-known measurements at the top of the Eiffel Tower.
Further improvements were made by Kolhörster [Fick2014], to make the elec-
troscopes used by Wulf (or rather the electrometers, since they allowed quantitative
measurements) more suitable for balloon flights at high altitudes. In collaboration
with the company Günther & Tegetmeier, a new design of these instruments had
been introduced, with thicker walls, in order to withstand pressure differences of
over 650 mbar (500 torr), expected at altitudes close to 9000 m. The mechanical
assembly of the fibres and the microscope had been made less sensitive to changes
in temperature, but, above all, the sensitivity of the instrument had been increased
by reducing the electrostatic capacity. This last condition made it possible to take
readings of the positions of the wires every minute, so that the reading was associated
with a more precise altitude. On the contrary, Hess’s previous measurements only
allowed integrated measurements over an interval of 30 min, a time interval during
which the altitude reached by the balloon changed considerably.
Climbing to such high altitudes to make measurements required considerable
efforts and a dose of courage on the part of the experimenters. Already at that time the
opportunity had been considered to equip the measuring instruments with automatic
recording systems, which would allow the launch of unmanned balloons, to reach
altitudes prohibitive for human crews.
5.2 From Wulf’s Electroscopes to Automatic Recording Equipment 93
Apart from rare exceptions, the possibility of using some automatic equipment
made generally use of a setup including multiple balloons, which—once reached a
certain altitude—had to return to the ground, by exploding one of the balloons. This
solution proved to be competitive compared to the use of human crews. We have
already mentioned previously that at the beginning of the 1930s manned flights at
very high altitudes were carried out by Piccard [Piccard1932] up to 16 km above sea
level, using a hermetically sealed cabin. However, during the first of these flights the
crew ran into serious danger due to loss of cabin pressurization.
We know that Millikan and the group of US physicists were among the first,
as early as 1926, to install recording equipment on board during scientific expedi-
tions employing balloons [Millikan1926a], through photographs taken at intervals,
which took up the reading of the electrometer and the altimeter. Figure 5.1 shows
the structure of one of these devices used by Millikan in the first balloon flights
[Millikan1926b]. The electroscope employed two quartz fibres, similar to those used
in Wulf’s electroscopes, which were charged by a 300 V battery.
Figure 5.2 shows a photo of this early instrument [Millikan1926b]. The sky light
entering from aperture S1 (a thin vertical collimator) produced a shadow on photo-
graphic film, through the aperture S2 . A second photographic film could record the
reading of a thermometer and a barometer.
Fig. 5.1 Sketch of the

electroscope used by
Millikan on the occasion of
the first balloon flights with
automatic recording of the
measurements
[Millikan1926b]. Figure
reproduced from the work of
R. A. Millikan and R. M.
Otis, High frequency rays of
cosmic origin. II. Mountain
peak and airplane
observations, Physical
Review 27(1926)645.
Copyright (1926) by the
American Physical Society,
License No.
RNP/22/MAY/053658
Fig. 5.2 One of the electrometers with an automated reading system using photographic film,
used by Millikan during the first measurements aboard of balloons in the second half of the
1920s [Millikan1926a]. Figure reproduced from the work of R. A. Millikan and I. S. Bowen,
High frequency rays of cosmic origin. Sounding balloon observations at extreme altitudes, Phys-
RNP/22/MAY/053657
These are among the first data loggers in history, perhaps together with the early
chart recorders for environmental monitoring in use since the early 1900s: small
devices capable of periodically monitoring the values of physical quantities in an
automated way, for subsequent analysis. While the pioneering flights of Hess and
Kolhörster suffer from a certain artisan approach, the American spirit was probably
not foreign to this new approach, which enters the contest by bringing substantial
innovations to the measurement procedures and designing the conditions in an almost
industrial way. The electroscopes provided with recording devices, developed on the
recommendation of Millikan, were in fact small jewels, having a mass of about 300 g,
and therefore only required small balloons. Each electroscope was carried by two
balloons, filled with hydrogen and having a diameter of about one meter. The purpose
of the second balloon is to allow the instrumentation to return to the ground after
the destruction of the other balloon, and also to signal the presence of the object to
those who will find the instrumentation, with indications for its recovery. In fact, of
the four devices used during one of the first flights, three will be recovered and two
of them correctly recorded the data, reaching altitudes of 11.2 and 15.5 km, after
moving to a distance of about 130 km.
For the measurements carried out in the high mountains, or under a certain
depth of water, Millikan used different models of electrometers, whose construc-
tion details and method of operation are described in one of his works from 1926
[Millikan1926b].
5.2 From Wulf’s Electroscopes to Automatic Recording Equipment 95
Fig. 5.3 Electrical circuit for the use of the electrometers employed by Millikan [Millikan1926b].
Figure reproduced from the work of R. A. Millikan and R. M. Otis, High frequency rays of cosmic
origin. II. Mountain peak and airplane observations, Physical Review 27(1926)645. Copyright
(1926) by the American Physical Society, License No. RNP/22/MAY/053658
The way to operate an electrometer for quantitative measurements is to charge

the electrometer with an external voltage source of known value and measure the
change in the potential as a function of time as the electrometer discharges. With
reference to Fig. 5.3, which describes the circuit used for these devices, the switches
allow to apply a voltage to the fibres and, after a calibration phase, to read the values
of the voltage over time. From these data, knowing the electrostatic capacity C and
the volume V air of air contained in the instrument, the number N of ions produced
per unit of volume per second can be estimated by the relation
ΔV
N =C
eΔt Vair
where ΔV/Δt is the change in potential, expressed in Volts per second, and e
represents the electron charge.
In the case of measurements carried out under a certain depth of water, the elec-
trometers were immersed appropriately protected, left under water for a long time
(a few hours) and then recovered to be read, obtaining in this case only a few read-
ings as a function of time, which is not particularly critical, since in this case it was
necessary to measure what happened at a fixed depth, and not at a rapidly variable
altitude with time.
Electrometers with automated reading by means of photographic film recordings
were also used in more recent times by Millikan, for example for the geographic
survey carried out in various parts of the world in 1936 [Millikan1936].
High altitude measurements with automated electrometers, up to 28 km of alti-
tude (pressure of 22 mm of Mercury), were also reported in 1932–1933 by Regener
[Regener1932a, Regener1932b, Regener1932c, Regener1933], with instruments
whose reading was photographed at intervals of a few minutes.
5.3 Ionization Chambers
Devices that were widely used in Compton’s measurement campaigns in the early
1930s, for example in expeditions to various parts of the world to study the depen-
dence of ionization on geomagnetic latitude, were the ionization chambers filled with
high pressure Argon. They are essentially instruments that integrate the flux of the
detected ionizing particles, as they measure the total number of ion pairs created in
a certain time interval within the volume of gas.
A description of such apparatus is reported for example in [Compton1933]. The
following figure shows a sketch of the apparatus. It should be noted that in those years
the Geiger counters already existed (see next section), but Compton believed that for
this measurement campaign the ionization chambers were preferable, because of
their greater reproducibility and lower statistical error.
Filling them with high-pressure Ar produced a high ionization, which was
measured using an electrometer. The ions produced in the chamber were collected
by a steel electrode that conducted the current to the electrometer needle. Spherical
bronze and lead shells shielded the instrument from local gamma radiation, reducing
it to about 5% of the ionization due to cosmic radiation. In order to calibrate the
instrument, the readings were each time compared with ionization measurements
produced by a radioactive radium capsule (Fig. 5.4).
Ionization chambers filled with high pressure gas were also used by Montgomery
and collaborators in 1936 [Montgomery1936], using Nitrogen at a pressure of about
15 atmospheres. The chamber used in some experiments was segmented into two
independent parts, each connected to an electrometer with automatic recording and
operating in connection with Geiger counters whose reading was also recorded.
In general, in an ionization chamber there are two electrodes separated by a
dielectric (the filling gas) and held at a certain potential difference. The electric
field generated by the potential difference causes the ions to move along the lines
of force of the field, following a motion that overlaps the random thermal motion.
The average speed due to random motion is generally greater than the drift speed,
due to the presence of the electric field. The ions, once formed in the gas, can not
only diffuse, because of the thermal motion, or migrate with the electric field, but
also recombine with particles of opposite sign before they can be collected at the
electrodes. If the electric field reaches an adequately high value, most of the charges
produced by ionization can be collected on the electrodes, and the measured current
will be proportional to the total number of ion pairs produced.
How to read this current? If it were very high, it could be read by a galvanometer.
However, in most of the measurements related to the detection of cosmic radiation,
this very low current was read by electrometers. The variation over time of this
potential difference could give a measure of the current, and therefore of the total
number of ions formed, through the relation
dV
I =C
dt
5.3 Ionization Chambers 97
Fig. 5.4 Ar-filled high

pressure ionization chamber,
used by Compton during the
measurement campaigns at
the various geomagnetic
latitudes [Compton1933].
Figure reproduced from the
work of A. H. Compton, A
geographic study of cosmic
rays, Physical Review
43(1933)387. Copyright
(1933) by the American
Physical Society, License
No. RNP/22/MAY/053659
where C represents the capacity of the instrument. The integration time is given in
this case by the subsequent time interval between two readings.
Of course, an ionization chamber can also be used to detect the passage of a
single ionizing particle. However, it must be considered that in this case the voltage
pulse that can be obtained at the electrodes is very small and requires an appropriate
amplification system. Typically, for energies of some MeV, such as those deposited
by particles emitted by a radioactive source, or by the passage of secondary cosmic
radiation, the voltage pulses—for geometries typical of these chambers—are of the
order of mV, therefore they certainly need to be amplified, which became possible
in the following years with the progressive development of low-noise electronics.
Gas detectors based on ionization allow, by means of an appropriate choice of the
electric field (in turn dependent on the geometry of the electrodes and the voltage
applied to them), to choose an appropriate working point for the detector, also called
operational regime. At relatively low voltage, the number of ion pairs produced by
ionization may not be measured correctly, because the fraction of ions that will
undergo recombination processes in the gas is still high. As the applied voltage is
increased, the number of ions collected at the electrodes increases, until it saturates,
a condition in which ideally all the ions are collected, and this number represents
a measure of the total ionization produced in the integration interval. In case it is
possible to detect the passage of a single particle, the number of ions produced will
give an estimate of the energy deposited by the particle in the sensitive volume,
taking into account the average energy to produce a positive electron–ion pair in a
gas, a few tens eV.
It is known that ionization-based gas detectors can operate in different regimes,
depending on the electric field applied to the electrodes and the working pressure
existing inside the detector. For a detailed description of the operation of gas detectors,
reference can be made to one of the many modern textbooks existing in the literature
[Leo1987, Knoll2000, Leroy2004, Grupen2008].
The different regimes are generally illustrated with reference to a qualitative plot
which shows the number of ions produced as a function of the bias voltage, for a
constant pressure (Fig. 5.5). The vertical scale is logarithmic, which implies a huge
increase in the number of ions in passing from one regime to the next.
For low supply voltages (left part of the plot) the electric field is not sufficient to
collect all the charges produced by primary ionization, which are partially recom-
bined. The fraction of charges collected increases as the working voltage increases,
until a first “plateau” is reached, where the number of charges collected does not
increase further, even if the voltage is increased. This regime corresponds to the
complete collection of the charges produced by primary ionization. Ionization cham-
bers operate in this regime and produce, if used to detect single events, very low
voltage pulses.
Fig. 5.5 Operating regimes for a detector based on the ionization in a gas. The plot shows the
number of ions collected (in logarithmic scale) as a function of the power supply voltage applied
(at constant gas pressure). The two “plateau” (where the number of ions collected does not increase
further, even by increasing the voltage) correspond to the operating modes of an ionization chamber
and of a Geiger counter respectively
5.4 Proportional Counters 99
5.4 Proportional Counters
Since single particles, especially those with low ionizing power, produce a limited
quantity of ions and the corresponding amplitude of the voltage pulse would be too
small to be detected without adequate amplification, one of the possible strategies is
to exploit the phenomenon of secondary ionization, according to which the number
of ions initially produced by direct ionization can be multiplied, even by a very high
factor, hundreds of times. The strategy is to use a cylindrical geometry and a much
stronger electric field in a specific region of the sensitive volume. The cylindrical
geometry makes use of a cylinder-shaped electrode, of radius b (typically a few cm),
and of a central wire, of radius a, stretched along the axis of the cylinder and of very
small size. A potential difference V is applied between the two electrodes, with the
central wire at positive potential. Under these conditions the electric field has a radial
symmetry, i.e., its value depends on the modulus of the distance r with respect to the
cylinder axis, and takes on a value given by
V
E=
r log ab
The electric field is therefore very intense near the wire, and the electrons can
be accelerated in that region until they take on an energy that is capable of ionizing
in turn other atoms or molecules of the gas, multiplying the number of electrons
initially produced. In the ionization process, photons are also emitted, which in turn
can produce additional electrons. It can be shown that the total number of electrons
reaching the central electrode is given by the series
n + γ n2 + γ 2n3 + · · ·
where n is the number of electrons produced initially and γ is the proportionality

constant. If the term γ n < 1, the series converges to a value given by
n
M=
1 − γn
which leads to a number of electrons M approximately proportional to the number

of electrons initially produced. Such are the conditions for the detectors usually
called proportional counters. With reference to Fig. 5.5, the central area of the plot
shows a trend in which the collected charge increases with the applied voltage, as n
increases with the voltage. Furthermore, different particles, having different ionizing
power, would produce a different initial number of electrons and to some extent the
total number of electrons produced could be used to distinguish the different detected
particles. Proportional counters are capable of producing higher voltage pulses, which
can be recorded with an appropriate electronic system, generally preceded by a further
amplification of the signal. Proportional counters were used in subsequent years, for
example from the late 1940s, for various applications concerning radioactivity and
the detection of cosmic rays.
5.5 Wilson Cloud Chamber
The cloud chamber is inextricably linked to the name of the Scottish physicist Charles
Thomas Rees Wilson, already mentioned. Having first studied zoology and later
physics, Wilson had received a temporary position as a meteorological observer in
1894. Impressed by the observation of numerous atmospheric phenomena, including
cloud formation, Wilson realized that the ions existing within the gas could act as
centres for the formation of water droplets.
It had already been established that solid particles (e.g. dust) could act as centres of
condensation. Wilson, building his first cloud chambers at Cambridge, realized that
dust grains were not necessary for condensation, as long as there was rapid expansion
with a volume ratio, V1 /V2 , greater than a certain value. The first prototypes of these
devices, built in glass, were very fragile, but after numerous attempts, Wilson was
able, through a very rapid expansion, 1/100 of a second, to visualize some of the
classic atmospheric phenomena. All this took place in 1895, and initially had no direct
connection with the possibility of detecting the passage of particles contained in the
cosmic radiation. But already in 1896, with the arrival in Cambridge of Rutherford
and Townsend, and with the discovery of the phenomena of radioactivity and X-
rays, it became clear that these devices could visualize phenomena produced by an
ionizing radiation. At the beginning of 1896 X-rays sent on Wilson’s devices were
able to produce a dense mist, establishing the correlation between the condensation
nuclei in the gas and the ions created by the ionization source. The formation of these
droplets was due to the fact that the water vapor was supersaturated. The following
years saw a number of detailed measures to assess the conditions under which droplet
formation took place. It was established that for values of the expansion ratio below
a certain limit the condensation centres were the negative charges; increasing the
compression ratio the positive ions came into play, and finally the air molecules. It
should be noted that this technique led to a first estimate of the electron charge by
Thomson, evaluated as 2.2 10–19 C. Millikan, as it is known, subsequently used the
oil droplets to measure a more precise value of the electron charge.
Already in 1901, on the basis of measurements performed with his device also in
underground tunnels, Wilson understood that the production of ions could be due to
some external source of radiation [Wilson1901a, Wilson1901b]. It was only in 1910
that Wilson returned to dealing with these devices, after having investigated various
problems related to atmospheric electricity, with the idea that they could be a tool
for visualizing the passage of ionizing particles.
In 1910, Wilson’s device already allowed to synchronize the expansion of the gas
volume with the emission of the X-rays to be observed and with the lighting device
to obtain the photograph [Wilson1912]. Figure 5.6 shows Wilson’s original chamber
5.5 Wilson Cloud Chamber 101
at the Cavendish Laboratory Museum. Finally, in 1911 Wilson was able to obtain
the first detailed images of the passage of electrons and alpha particles (Fig. 5.7).
Subsequently, these chambers were used to visualize a remarkable variety of phys-
ical processes, including the observation of the positron, by Anderson, phenomena
Fig. 5.6 Improved version of Wilson’s chamber, 1911, exhibited at the Cavendish Museum.
According to what Wilson himself reported, no other specimens of this chamber were built. Source
Wikimedia Commons, Creative Commons Attribution-Share Alike 4.0 International License
Fig. 5.7 Tracks of alpha

particles, obtained in an
expansion cloud chamber.
Reproduced with permission
from the Royal Society from
C. T. R. Wilson, On an
expansion apparatus for
making visible the tracks of
ionising particles in gases
and some results obtained by
its use, Proceedings of the
Royal Society, London A
87(1912)277
related to cosmic radiation, such as the passage of muons and their decay, the produc-
tion of pairs, the production of showers, and, later, the observation of new particles,
such as V0 or hyperons. In 1929, as we discussed, Skobeltsyn observed for the first
time, by means of a Wilson chamber, tracks due to energetic particles in the cosmic
radiation.
As we know, Wilson received the Nobel Prize in 1927, together with Compton, for
having made it possible to observe the trajectories of charged particles through the
condensation of vapor. A detailed historical analysis of the introduction of the cloud
chamber in the 1930s has been reported, among others, by Leone [Leone2011].
Today, the use of small, manually operated, cloud chambers is an interesting
educational tool for observing the tracks of particles emitted by radioactive
preparations or particles of the cosmic radiation.
Demonstrative small cloud chambers, mainly for educational use, are available
on the market at affordable prices from the major suppliers of teaching equipment
for the study of physics. They can be of the expansion type, in which a compression
and subsequent expansion performed manually on a small volume leads to a super-
saturation of the mixture, making the passage of any tracks visible for a few seconds.
Other models, for longer observations, are of the cold plate type, to be cooled by dry
ice, or by Peltier cells, and make use of pure isopropyl alcohol. Once cooled, they
allow observations of the tracks for several hours. Today it is also possible to build
and operate this type of cloud chambers, according to the layout shown in Fig. 5.8.
Many articles or educational sites report similar examples [Barradas2010, ESPER].
Fig. 5.8 Operation diagram of a cloud chamber built for educational applications, based on the use
of dry ice and isopropyl alcohol
5.6 Geiger-Müller Counters 103
5.6 Geiger-Müller Counters
If in a proportional counter with cylindrical geometry, such as those described above,

the voltage difference between the central electrode and the outer cylinder is further
increased, the number of electrons produced becomes greater than the number of
initial electrons (γ n > 1). In this case, the mathematical series mentioned when
discussing proportional counters, which estimates the total number of electrons
produced for each primary pair, diverges. From a physical point of view there would
be then a discharge. However, if the product γ n is not much greater than 1, the
discharge fails to self-sustain, so it is extinguished after each single event. The counter
is therefore capable of detecting every single ionizing particle that passes through it.
It is in this regime, with reference to Fig. 5.5, that the Geiger-Müller counters work.
Geiger-Müller counters operate in correspondence to a second plateau, in which
the increase in the supply voltage to a first approximation does not produce a further
increase in the charges. In fact, this is not a strict plateau, as a few percent increase in
charge can still be observed in this region. Geiger-Müller counters are characterized
by a very high output voltage pulse, of the order of volts, and the amplitude of this
pulse is independent of the primary charge. These detectors are therefore not able to
distinguish between different ionizing particles or to evaluate the energy deposited
in the sensitive volume.
Although the first counter prototypes had been developed by Hans Geiger (1889–
1945) as early as 1908, under the supervision of Rutherford [Rutherford1908], to
detect alpha particles, it was only in 1928 that Geiger and Walter Müller (1905–1979)
made this type of counter usable also for the detection of electrons and gammas
[Geiger1928a, Geiger1928b, Geiger1929a, Geiger1929b], in the form of a compact,
robust, small size and low cost instrument.
The Geiger-Müller counter probably represents one of the best known tools, not
only in the scientific field but also in a broader context, so much so that it was defined
by Einstein as “The most sensitive organ of humanity” in addressing to Geiger in
one of his letters. The history of the introduction of this type of detector and the
significance it had in the field of experimental physics have been the subject of
numerous historical and educational articles [Korff2013]. It represented one of the
first examples of detectors capable of signalling the arrival of both alpha particles,
electrons and gammas, as well as the muons from the cosmic radiation.
A Geiger-Müller counter essentially consists of a conducting cylindrical tube,
which acts as a cathode and a thin wire, also conductor, arranged along the axis of
the cylinder, which acts as an anode. The tube contains a suitable gas (Argon, …)
or a gas mixture, in the most recent versions. Typical Geiger-Müller counters have
diameters of a few centimetres and lengths of 5–10 cm, but they could also reach
up to one meter in the larger models. Depending on the type of application, they
can be equipped at one end with a thin entrance window, made of mica, which also
allows the entry of alpha particles or low-energy electrons, which would otherwise
be stopped in the constituent metal layer of the cylinder.
Fig. 5.9 Basic circuit for

powering and taking a signal
from a Geiger counter
The basic power supply and signal pick-up circuit is sketched in Fig. 5.9. The
central wire is connected to a positive potential + V by means of a high value resistor.
The variation of the voltage produced by the collected charges is taken through a
capacitor. Because of the cylindrical geometry of the electrodes, the electric field
has a radial symmetry, as described for the proportional counter, and depends on the
diameter of the external cylinder and that of the central wire which acts as anode.
Typical potential differences of several hundred volts are applied between the two
electrodes, depending on the size of the counter.
In the case of cosmic ray detection, due to the very high energies involved, the
thickness of the outer cylinder does not have much influence on muons or high-
energy electrons, which therefore can be detected without any problem. As for the
gammas, they can be detected, albeit with a low probability (a few percent), since
when interacting with the material constituting the outer cylinder (usually a metal
with a high atomic number) they can produce electrons, which in turn interact in the
fill gas, producing ionization. The detection efficiency is close to 100% for all the
ionizing particles that are able to penetrate, directly or indirectly, into the internal gas
volume. Figure 5.10 shows an example of a typical Geiger counter for educational
activities.
In the first applications of Geiger-Müller counters, during the 1930s, various signal
reading circuits were used, using vacuum tubes as amplifier circuits and various
systems for counting the pulses [Curtiss1928, Libby1932].
After the first years of their discovery and their use in scientific research, Geiger
counters were also successful as a tool also available to the general public. It is
amazing to find how the diffusion of “kits” with numerous versions of Geiger coun-
ters, sold in particular in the US market from the 1940s to the 1960s [NRIC], was
really impressive.
Geiger-Müller counters are now easily available on the market, also in a version
accessible to the general public or for educational applications and constitute one of
the most interesting tools for carrying out a variety of experiments and activities
concerning the study of cosmic rays [Famoso2005, Blanco2006a, Blanco2006b,
Blanco2008a, Blanco2008b, Blanco2009, Goldader2010, Maghrabi2021].
5.7 Electronics and Coincidence Techniques 105
Fig. 5.10 An educational

Geiger counter (Photo by the
author, from the collection of
the Department of Physics
and Astronomy of Catania)
In today’s versions, the voltage difference is generally produced by a special

electronic elevator circuit, which, starting from a voltage of a few volts, supplied by
a battery, produces the required high voltage. These counters often contain also the
electronics capable of providing a standard logic pulse at the output, for example of
the TTL type (a transition from zero to a positive signal of 3–5 V), with a typical
duration of about a hundred microseconds, and to signal the arrival of each particle
by turning on an LED or producing a sound beep.
5.7 Electronics and Coincidence Techniques
The use of counters to detect the passage of ionizing radiations, especially the Geiger-
Müller counters, opened the way to the possibility of operating two or more of these
detectors “in coincidence”, that is to signal the “simultaneous” passage of the same
particle, or of two different particles, in the two counters. This possibility really
represented a qualitative, as well as quantitative, novelty in the detection of particles,
because it provided the possibility of highlighting new phenomena in cosmic ray
physics, not observable by the use of a single counter. The coincidence technique, as
it was named since its beginning, represents today a standard way for the selection of
physical events of interest, not only in cosmic ray physics, but in general whenever
particles or ionizing radiation are detected in modern physics experiments, from

astronomy to solid state physics.
The invention of the “coincidence” technique, that is the possibility of observing
“simultaneous” phenomena in different detectors, dates back to 1924, when Bothe, in
the context of the study of the Compton effect, developed a system to evaluate whether
the scattered photon and the recoil electron detected in two different detectors were
the result of a single physical process (with conservation of energy and impulse in
the single event) or of a simple statistical correlation (with conservation of energy
and impulse averaged over many events). As described by Bothe on the occasion
of the Nobel Prize [Bothe1954], the prototype for reading each of the two detectors
was done at first by recording with a writing needle on a moving paper roll, with
consequent photographic recording. The analysis of the two rollers made it possible to
establish, within a time interval of 10–4 s, that the two phenomena were simultaneous
and that therefore it was a single physical process. The technique, according to
Bothe himself, was rather rudimentary, and the consumption of film so high that the
laboratory where the film was placed to dry resembled an “industrial laundry”. The
same technique had also been used in 1926, on the occasion of another experiment on
the distribution of radiation, once again to establish whether the processes in which
photons interacted were single processes.
It was in 1929 that this technique was used, by Bothe and Kolhöster, with Geiger
counters, for the simultaneous detection of the passage of particles in different coun-
ters. A historical analysis of the valence and significance of the introduction of
coincidence circuits in cosmic ray physics has been discussed by several authors
[Pfotzer 1982, Bonolis2011b].
As an example, imagine having two cylindrical Geiger counters, represented as
in Fig. 5.11, and placed with their parallel axes, at a certain distance vertically on
each other. It is clear that each of these counters can be crossed by particles from
the cosmic radiation (tracks A, B), but there will be tracks capable of crossing both
counters if they come from certain directions (tracks C, D). If it were possible to
select only those tracks that cross the two counters, this would automatically entail a
selection of the incoming directions, within a certain angular interval defined by the
dimensions of the counters and by their relative distance. In the configuration drawn
in the figure, for example, the tracks that correspond to the passage of the particles
in both counters are tracks with an orientation close to the vertical direction. The
potential of this method is then clear, to select events in which the direction of the
track is defined, even though within certain geometrical limits.
In fact, the first uses of coincident counters refer to experiments, such as those
performed by Bothe and Kolhöster [Bothe1929a], in which it was useful to define
the average direction of the particles, in order to select only those that had crossed a
given thickness of interposed material between the two counters.
The technology for observing signals in coincidence was soon improved, and
as early as 1930, instead of photographic recording, the first electronic circuits,
based on vacuum tubes, coupled to mechanical recording systems, had been used.
Bothe himself used one of these circuits, which made use of a vacuum tube with
5.7 Electronics and Coincidence Techniques 107
Fig. 5.11 Geometric arrangement of two Geiger counters placed vertically, with parallel axes, seen
on two orthogonal planes. Tracks A and B cross only one of the two counters, while tracks C and
D cross both of them
multiple grids, to establish the coincidence between two detectors, improving the time
resolution and therefore reducing the number of spurious or “random” coincidences.
The evaluation of possible random coincidences in a coincidence experiment is
always an important aspect, in determining whether the number of events classified
as “coincident” is partially or totally influenced by the possible occurrence of inde-
pendent events within the same time window. This assessment can be made and is
indeed reported in the same work by Bothe and Kolhöster of 1929 [Bothe1929a].
According to this evaluation, the rate Rs of spurious coincidences (number of spurious
coincidences per unit of time) is given by
Rs = 2 R1 R2 Δt
where R1 , R2 are the counting rates of the two detectors individually and Δt is
the time coincidence window, i.e., the time interval within which the simultaneous
arrival of the two signals can be evaluated. With the same counting rate of the two
detectors (R1 , R2 ), the rate of spurious coincidences can be decreased, therefore,
by proportionally reducing the time coincidence window Δt, provided that possible
events of interest are not lost (Fig. 5.12).
Electronic circuits based on the use of vacuum tubes were further improved by
Bruno Rossi [Rossi1930a], and we can say that their principle is basically still used
today. In the vacuum tube version used by Rossi, two or more triodes are used, which
have the plate connected in common, while the grid of each triode receives, through
a capacitive coupling, the signal from the Geiger counter. Only when the grid of
each triode is simultaneously brought to a negative potential by the discharge of
the corresponding Geiger counter, an output signal will be obtained. The advantage
of Rossi’s coincidence circuit is that this technique can also be used in the case
Fig. 5.12 Walther Bothe

(1891–1957). Source AIP
Emilio Segre Visual
Archives
of multiple coincidences between more than two counters, while Bothe’s use of a
double-grid triode limited the possibility of coincidence to only two detectors.
In Rossi’s coincidence circuit (Fig. 5.13) the output signal was initially detected
by listening to the produced sound signal with a headset, which is impractical for
long measurements. Subsequently, a galvanometer was used, the deviation of which
was photographed, and later, a mechanical recorder capable of “counting” the coinci-
dence events. It was with this instrumentation that Rossi contributed to the study and
understanding of numerous aspects of the cosmic radiation, partly already described
in the previous chapters.
We will later see another example of the use of coincidence circuits, this time
not to signal the passage of the same particle through two distinct counters, but to
detect the simultaneous arrival of different particles in detectors placed at a certain
distance.
In the following years, with the development of nuclear physics and the first
experiments carried out with particle accelerators, the use of coincidence circuits
also became a common practice in nuclear and particle physics. As recalled by
Fig. 5.13 Principle of

operation of Rossi’s
coincidence circuit, based on
the use of two or more
triodes connected with the
common plate and with the
Geiger counters connected to
the grids of each triode
5.8 Nuclear Emulsions 109
Bothe during his Nobel Lecture, “Many applications of the coincidence method will
therefore be found in the large field of nuclear physics, and we can say without
exaggeration that the method is one of the essential tools of the modern nuclear
physicist” [Bothe1954].
The treatment of particle detector signals is today dominated by electronics and
microelectronics. Referring to the history of cosmic ray physics, up to the Second
World War, it is worth remembering that the first vacuum diodes were made by
Fleming as early as 1897, and that the first triodes, tubes equipped with a grid, in
addition to the cathode (filament) and to the anode, and used as the first amplifiers,
were made in 1906 by De Forest, and used about ten years later, as common elements
for radio sets.
5.8 Nuclear Emulsions
We also want to mention here another visual method for the detection of ionizing
particles, which played an important role in the development of knowledge concerned
with the physics of cosmic rays, especially in the period of the first decades of their
history, but which still finds some application today for specific experiments. This
is the use of nuclear emulsions, a special type of photographic plates coated with
a gelatin emulsion containing fine grains of silver halide. Nuclear emulsions, so
called because they can be used to record the passage of ionizing particles, must be
developed, after being exposed to the particles to be detected, and finally observed
under a microscope.
The track is geometrically defined with excellent precision (a few microns),
comparable to the precision in the tracking of particles obtainable with current silicon
detectors. Overlapping plates can be arranged to reconstruct the track of the particle
even in 3D. As a visual detector, it does not require any electronics, cables, or power
supplies for its operation, but of course the information is only available during
the analysis phase—often long and tedious—and not in real time. Because of their
characteristics, they have been used extensively for cosmic radiation measurements
especially at high altitudes in the mountains or aboard in balloons, contributing to
the discovery of some of the particles of modern physics.
There are extensive reviews, with historical notations, of the operating prin-
ciples and applications of nuclear emulsions, including textbooks [Powell1959,
Barkas1963] and review articles [Shapiro1941, Herz1966, Sime2013].
Since the early 1900s it was known that the silver halide grains in photographic
emulsions could signal the passage of ionizing particles under certain conditions.
Rutherford was one of the first to use photographic plates exposed to the radiation
emitted by radioactive materials. The next step, investigating whether it was possible
to detect the passage of a single particle, was taken by one of his collaborators, Sukeiti
Kinoshita, who placed a photographic plate near a low-intensity alpha source, in order
to calculate exactly the number of single tracks from which the plate would have been
hit in a certain time interval, and subsequently verify that the corresponding points
of impact of the track could be found in the plate, once developed. It was thus
demonstrated, in 1910, that photographic plates could detect single events, due to
the passage of single particles [Kinoshita1910].
The real reconstruction of a single track, as an ordered set of grains that define
its geometric characteristics, was made the following year, in 1911, by Maximilian
Reinganum (1876–1914), professor of physics in Freiburg, who observed tracks of
alpha particles arriving on the plate strongly inclined, therefore crossing a higher
effective thickness, with a more consistent energy deposit [Reinganum1911]. In
1912 the Reinganum results were confirmed and extended by Michl [Michl1912],
who was able to quantify some parameters of the observed tracks (length, number of
grains corresponding to alpha particles of different energy). This technique was soon
applied to a few experiments, such as the scattering of alpha particles from a thin
sheet, detecting the scattered particles within concentric circular rings on the plane
of the nuclear emulsion, and thus further verifying Rutherford’s law of diffusion
[Mayer1913], already experimentally studied by Geiger and Marsden [Geiger1909].
Kinoshita himself then observed “stars” of tracks due to numerous alpha particles that
radiated from a point of the emulsion contaminated with a radioactive preparation
[Kinoshita1915], as shown in Fig. 5.14 [Shapiro1941].
The detection of electrons, in addition to alpha particles, was another important
step in the use of these first photographic plates, by Sahni [Sahni1915]. In this case
the tracks, unlike the straight ones produced by alpha particles, were more irregular
and tortuous, as were those produced by the gamma radiation, which in turn produced
electrons.
It would therefore seem that already in the first decade of 1900 it was possible
to detect by such tool the passage of ionizing particles, and, in principle, also those
associated with the cosmic radiation. Indeed, it must be considered that particles of
unitary charge, such as protons, or those that would have been identified later as
mesons, have a much less specific energy loss than that of alpha particles, and would
Fig. 5.14 Tracks of alpha

particles that radiate from a
point of the emulsion,
contaminated by a small
amount of radioactive
material (Polonium)
[Shapiro1941]. Figure
reproduced from the work of
Maurice M. Shapiro, Tracks
of nuclear particles in
photographic emulsions,
Review of Modern Physics
13(1941)58. Copyright
(1941) by the American
Physical Society, License
No. RNP/22/MAY/053805
5.8 Nuclear Emulsions 111
have been very difficult to detect them with the sensitivity of the first photographic
plates, which became possible only a decade later.
The probability of observing the grains that make up the track depends in fact on
the energy dissipated in that volume and, therefore, on the specific energy loss of the
particle, in turn a function of the type of particle and its kinetic energy. The density
of the grains along the track, i.e., the number of silver grains per unit of length, is
given by the product
P×A×N
where P is the probability of producing a developable grain, A is the transverse area of

the grains and N the number of grains per unit of volume. In photographic emulsions,
however, following the development, a certain number of grains are always present,
randomly distributed throughout the volume, forming a background. The possibility
of distinguishing the track of a particle, that is an ordered set of grains that define
a trajectory, depends on the average separation between the grains of the track and
that between the background grains. A photographic emulsion, therefore, will not
be able to detect charged particles that have a specific energy loss below a certain
threshold.
One of the problems related to the use of nuclear emulsions is the time interval
within which development must be carried out, starting from the date on which the
emulsions were exposed to the passage of particles, before the image begins to “fade”.
This time was of the order of a month. The other problem was related to the need
to observe the plates under a microscope to reconstruct the tracks of the particles, a
work of patience and which took a long time; this procedure also generally produced
a mosaic of several enlarged images, even to explore a small area of the emulsion.
The type of tracks observed through nuclear emulsions varies, depending on the
type of particle and its energy. In some cases, the tracks are very thin and barely
perceptible with respect to the background, while in other cases the lines are thick.
The observed tracks can be straight lines or show large scattering effects. The detailed
analysis of the tracks observed in nuclear emulsions can therefore lead to quantitative
information about the density of the grains (extracting from this the specific energy
loss), the range of the particle, whether it is stopped inside the emulsion, the mean
scattering angle in the material, any deflection in the presence of a magnetic field.
The combined measurement of different quantities can also lead to an effective
identification of the particle.
Ordinary photographic films are not particularly suitable for use as nuclear emul-
sions, i.e., as track detectors, as they have a very small thickness (a few microns)
and the particles would have to cross them with a considerable inclination to give
rise to an appreciable loss of energy. Furthermore, the threshold value in dissipated
energy is much higher than the value corresponding to the minimum of ionization,
so much so that only strongly ionizing particles (for example heavy nuclei) could
exceed it. In the following decades, however, photographic plates with a much larger
thickness were produced and techniques were introduced to increase the sensitivity
Fig. 5.15 A few photographic emulsions used in the past (Photo by the author, from the collection
of the Department of Physics and Astronomy of Catania)
of the plates (by increasing the number of grains per unit of length) and to improve
the development of the plates by appropriate methods.
The possibility of using this technique for the detection of particles of cosmic
origin was therefore only addressed a decade later. In particular, it was the Austrian
physicist Marietta Blau (1894–1970), in Vienna, who investigated in the early 1920s
the possibility of using different types of photographic plates to also detect protons.
For this purpose, she used an alpha radioactive source, whose decay products, ener-
getic alpha particles, were capable of producing the emission of protons (called H-
rays in that context) from hydrogenated materials, such as paraffin. The first results
demonstrating this possibility were published in the mid-1920s [Blau1925]. In the
following years, Blau improved this method, using different types of support and
changing the conditions of plate development, also managing to interest the English
company that produced photographic material, ancestor of Ilford, to produce thicker
supports more suitable for the detection of particles [Blau1932] (Fig. 5.15).
We will further discuss Marietta Blau’s contribution to cosmic ray physics, with
her measurements done in the second half of the 1930s, using nuclear emulsions at
high altitudes, for the observation of nuclear interaction processes.
Because of this interest and the results achieved with this technique, from the
end of the 1930s some models of photographic emulsions specifically produced to
detect the passage of ionizing particles were made available by the major producers
of photographic films (Kodak and Ilford). The first emulsions of Ilford allowed to
trace protons of about 100 MeV of energy, while those produced by Kodak in the
following years also allowed the detection of particles at the minimum of ionization.
It should be noted that the development of these techniques and the realization of
these commercial devices saw a close collaboration between the world of industry
and that of research, represented in this context by Marietta Blau and subsequently
by physicists such as Powell and Occhialini at the University of Bristol, who had
heard of her first physics results at high altitude.
Although the Nobel Prize in Physics in 1950 was awarded to Powell “for his devel-
opment of the photographic method of studying nuclear processes and his discoveries
5.9 Detectors Based on Scintillators 113
Fig. 5.16 A plate installed aboard a stratospheric balloon shows the track of a cosmic radiation
particle, probably a high-energy alpha particle. Overall length of the track 550 μm [Shapiro1941].
Figure reproduced from the work of Maurice M. Shapiro, Tracks of nuclear particles in photographic
emulsions, Review of Modern Physics 13(1941)58. Copyright (1941) by the American Physical
Society, License No. RNP/22/MAY/053805
regarding mesons made with this method”, the systematic use of nuclear emulsions
for the quantitative observation of tracks produced by nuclear particles is undoubt-
edly linked to the name of Marietta Blau, who unfortunately had no recognition, and
had a very troubled life, while continuing to carry out scientific activity until 1970,
as we will see in the next chapter.
The use of nuclear emulsions, which in those years saw a period of great fortune
for the physics of cosmic rays, decreased in the following decades, in favour of
electronic detectors, even if they are still used today in particular experiments for the
investigation of rare processes.
Figure 5.16 shows a track due to cosmic rays (probably an alpha particle), observed
in one of the plates installed on board a stratospheric balloon [Shapiro1941].
5.9 Detectors Based on Scintillators
Although the period from the discovery of cosmic radiation up to the 1940s is char-
acterized above all by the use of the detectors that we have discussed so far, it is
worth mentioning in this context also the detectors based on the scintillation process,
which began to be used only in the following decades, to the point of replacing in
many cases most of the previous detectors. A historical review of the first uses of
scintillation-based detectors was first reported by Krebs in 1955 [Krebs1955].
The origin of modern scintillators actually dates back to the scintillation process,
studied and used in visual mode since the early 1900s. In 1903, for example, Crookes
used a Zinc Sulfide (ZnS) screen to detect the fluorescence created by the emitted
alpha radiation from a radioactive source. Observing the screen through a microscope,
Crookes could see individual flashes of light created by the individual interactions
of the alpha particles with the screen. The apparatus, called spintariscope, was used
for some time, for example to study the diffusion of alpha particles. In more recent
times, small spintariscopes were also used as scientific toys. Many historical models
Fig. 5.17 Diagram of one of

the first scintillation devices,
based on the use of
scintillating phosphors
coupled to a special Geiger
counter [Krebs1941]
of spintariscopes are visible in museum collections, for example that of the Oak
Ridge Associated Universities [ORAU].
The real beginning of the use of scintillators in research related to nuclear physics
and cosmic ray physics, however, can be dated to the early 1940s. One of these
devices, due to Krebs [Krebs1941], shown in Fig. 5.17, makes use of a Geiger counter
with a special design, which allows the entry of scintillation photons produced in the
scintillating material.
With these first devices, which were soon improved in the following
years [Mandeville1950a, Mandeville1950b, Mandeville1950c, Mandeville1950d,
Mandeville1950e], it was possible to detect alpha, beta and gamma radiation, up to
the creation of commercial equipment within a few years [Daggs1952]. However, this
configuration was soon replaced, in the mid-1940s, by devices based on photomulti-
pliers, whose first prototypes had already been developed several years before. The
first scintillation counters were thus made by Curran and Baker in 1944 [Curran1944],
who showed how alpha particles with energies of the order of MeV could be
detected. However, only a few years later, in 1947–1948, its potential and advantages
for the detection of high-energy particles were fully recognized [Marshall1947a,
Marshall1947b, Coltman1947].
In 1947 Marshall, Coltman and collaborators demonstrated the possibility of
detecting single events due to alpha particles, electrons or gammas. These detectors
also proved capable of detecting X-rays, protons and neutrons. Further improvements
of these first detectors were proposed by these researchers, concerning the possibility
of using different scintillating materials, a different shape or volume of the scintil-
lator, improvements in the collection of light and in general in the optical system
and associated electronics. Similar results on the detection of alpha particles were
also obtained in the same year by Broser and Kallmann [Broser1947a]. Soon after
the same authors published another work, referring to other possible scintillating
materials, such as CaWO4 , Zn2 SO4 [Broser1947b]. The following years, around the
end of the 1940s, saw further improvements and uses of these detectors in different
areas of research, from nuclear physics to cosmic ray physics, from medical applica-
tions to dating techniques [Bell1948, Hofstadter1948, Morton1949, Mayneord1950,
Pringle1950]. It was soon recognized that scintillator-based detectors actually consti-
tuted a new radiation observation tool, representing a breakthrough equal to that of
the introduction of the Geiger counter at the end of the 1920s (Fig. 5.18).
5.9 Detectors Based on Scintillators 115
Fig. 5.18 A scintillator coupled to a light guide to direct the scintillation light towards a photo-
multiplier (Photo by the author, from the collection of the Department of Physics and Astronomy
of Catania)
The use of liquid organic scintillators to build cosmic ray detectors with a large area
(of the order of a square meter) had some disadvantages, including high flammability,
volatility, and toxicity. Other types of scintillators, free from these defects, had a high
cost. For this reason, low-cost plastic scintillators were developed in the 1950s. In a
work of 1957 [Clark1957a], for example, the preparation technique of this type of
scintillators is discussed, in view of the realization of a large number of devices to
be used for the detection of cosmic rays for a surface array.
Detectors based on scintillators of different compositions, both organic and inor-
ganic, with reading of the scintillation light carried out by means of photomultipliers
or more compact photosensors (Avalanche Photo Diodes, Silicon Photomultipliers
…), sometimes in connection with the use of special optical fibres (WaveLenght
Shifter fibres, WLS) have become common in recent decades and are also widely
used for the detection of particles associated with cosmic radiation. We refer to one
of the specialized textbooks [Leo1987, Knoll2000, Leroy2004, Grupen2008] for a
detailed description of the use of scintillation counters.
Chapter 6
The Interaction of Primary Cosmic Rays
in the Atmosphere
Abstract The chapter focuses on the interaction of the primary particles with the
Earth atmosphere. This aspect played an important role up to 1950s since it was
the only mean of observing high energy nuclear processes, before the advent of
particle accelerators. The first evidence of nuclear disintegration processes producing
secondary tracks came from the early observations by Marietta Blau, who employed
nuclear emulsions at the end of 1930s. Nuclear interaction events occurring at high
altitude were studied in the following years by different methods, confirming the role
and the importance of high-altitude laboratories, where most of these investigations
were carried out. An extended list of these laboratories is also reported in Appendix F.
The study of high energy particle interactions in the atmosphere gave also the first
evidence for a complex primary radiation, as well as for the production of a variety
of secondary particles, whose properties were extensively studied from 1940s on.
6.1 The First Evidence of Nuclear Interactions of Cosmic

Rays
When the nature of the primary cosmic radiation, as essentially consisting of high-
energy protons, was more and more accepted, it became important to understand how
they interacted with the nuclei of the Earth’s atmosphere and by which mechanisms
the different particles present in the secondary component were created, some of
which are able to reach the ground level.
The first experimental evidence of nuclear interaction processes capable of
producing a multiplicity of particles (nuclear disintegrations) came from the observa-
tions of the so-called “cosmic ray stars” (or disintegration stars), obtained by Marietta
Blau and Hertha Wambacher, in 1932, by the use of nuclear emulsions exposed on the
mountains, at an altitude of 2300 m, and which we have mentioned when discussing
the use of this method for the observation of particle tracks.
Some results on the nuclear interactions produced by particles of the cosmic
radiation had been obtained in those years also through measurements performed
with cloud chambers. For example, Anderson and Neddermayer reported several
results [Anderson1936] from a series of measurements conducted on Pike’s Peak
https://doi.org/10.1007/978-3-031-24762-0_6
118 6 The Interaction of Primary Cosmic Rays in the Atmosphere
(4300 m), while Brode and collaborators [Brode1936] also obtained at sea level
some evidence of the existence of processes due to particles heavier than protons.
Marietta Blau, an Austrian physicist, had contributed since 1923 to the study of
the tracks of charged particles, firstly using commercial photographic emulsions and
managing to identify the tracks of protons and alpha particles. Subsequently, using
photographic emulsions produced specifically by Ilford, Blau significantly improved
the visibility of these tracks, also determining the energy of fast protons observed in
the emulsions, as well as that of neutrons, based on the recoil protons. By exposing
a certain number of these emulsions for several months at an altitude of 2300 m, in
addition to the observation of normal proton tracks, Blau and Hertha Wambacher,
her assistant, realized the existence of specific events, in which the observed tracks
(sometimes even a dozen) came from a single point. These events were interpreted as
the disintegration of a heavy nucleus present in the emulsion, due to the interaction
with the primary track [Blau1937].
Figure 6.1 shows the diagram of an event reconstructed by Blau and Wambacher
[Blau1937], with the presence of several tracks, some of which are drawn with a
more or less thick line, indicating a greater or lesser ionization along the path.
Blau’s contribution in the use of nuclear emulsions and in the observation of
nuclear interaction processes induced by primary cosmic rays was of great impor-
tance, even if her work has often been overlooked or not properly considered. Powell
himself had begun to use nuclear emulsions for the observation of nuclear interaction
processes only in 1938, after having recognized their usefulness following the work
of Blau (Fig. 6.2).
The events of the Second World War and the Nazi occupation of Austria also
played a part in Marietta Blau’s personal life. Blau, as a Jew, was forced to leave
Fig. 6.1 Schematic diagram of the first observations, by Blau and Wambacher [Blau1937] of the
existence of disintegration stars, produced by the interaction of a primary particle with a nucleus
present in the nuclear emulsion. The thickness of the line indicates a greater or lesser density of
grains in the emulsion, while the arrow indicates the direction of the track, from the emulsion
towards the glass. The bottom interrupted track shows a longer track (not to scale)
6.1 The First Evidence of Nuclear Interactions of Cosmic Rays 119
Fig. 6.2 Marietta Blau in 1937, at the time when she obtained the first images of “disintegration
stars” by means of nuclear emulsions taken to high mountain. Source AIP Emilio Segrè Visual
Archives, Gift of Eva Connors
Austria in 1938 and to live in Norway, Mexico and then in the United States,
without being able to fully continue her research activity, which in the meantime
her colleagues in Austria—loyal to the regime—had appropriate. Her contribution
was often omitted, for example from the list of possible Nobel candidates, which
initially also contained her name, proposed by Schrodinger; Powell himself, in his
speech on the occasion of the Nobel award, did not even mention her papers and her
contribution [Strohmayer2006].
The tracks observed were initially attributed by Blau to particles with a high ioniza-
tion capacity, such as low-energy protons or alpha particles. Once the possibility that
these stars could be produced by small quantities of radioactive substances present in
the emulsions was excluded, their origin was recognized as due to cosmic rays. This
hypothesis was later confirmed by the fact that the probability of observing these
events increased if the emulsions were exposed at high altitudes, where the intensity
of the cosmic radiation is greater [Setter1939]. The discovery of mesons in those
years also suggested that they were produced precisely because of these interactions,
as secondary products in the atmosphere, excluding the possibility of processes in
which pairs of mesons were produced directly by photons, in analogy to the creation
of electron–positron pairs.
Experiments aimed at detecting penetrating mesons created as a result of nuclear
interactions in materials were carried out by Wataghin and Janossy around 1940,
using configurations of Geiger counters arranged under a layer of lead, or with
additional layers of lead placed between the counters. Because of the thickness
of lead involved, the energy of the particles capable of producing these secondary
particles had to be of the order of some GeV. Such events were also observed by
means of condensation chambers.
The characterization of these nuclear interactions went through a systematic series

of measurements made at different altitudes and on different materials. It immediately
became clear that the particles capable of producing these interactions could not
be photons or electrons, because these processes took place even after crossing a
considerable thickness of lead, capable of absorbing photons and electrons.
To evaluate whether the interactions were produced by the same μ mesons
present in the cosmic radiation, the measurements carried out at various altitudes
were decisive. These showed that the number of particles which produced these
nuclear processes, with the creation of penetrating showers, increased with altitude
much more rapidly (about a factor of 10) than the increase with the altitude of the
μ mesons. It was therefore demonstrated that neither photons, nor electrons, nor
muons had an appreciable probability of interacting with atomic nuclei, producing
secondary particles through nuclear processes. The possible candidates as probes
capable of producing these interactions were therefore protons, neutrons and π
mesons, a hypothesis that was further confirmed by the activity carried out in the
following years using detectors brought to high altitudes, both in laboratories in the
high mountains and by equipment launched aboard balloons at very high altitudes.
In summary, these nuclear interactions induced by particles present in the cosmic
radiation are capable of producing a great variety of reaction products, from charged
particles to neutrons, which then follow a different fate, depending on their nature and
the materials with which they interact: first of all the air in the terrestrial atmosphere,
where the first interaction of the particles coming from outside the Earth takes place,
and subsequently also the interactions of the secondary products.
6.2 Interactions in the Atmosphere and First Evidence

of a Complex Primary Radiation
The phenomenology based on the observation of the tracks produced in high-altitude

nuclear emulsions confirmed the hypothesis about the nature of primary cosmic rays,
mostly identified as composed by high-energy protons. Although in principle even
neutrons and pions were capable of producing nuclear interactions, both are subject
to spontaneous decay—neutrons with a lifetime of about 14 min, pions with an
enormously shorter lifetime of tens of nanoseconds—and therefore they could not
travel long in space before reaching the Earth. It was therefore more reasonable to
assume that the primary particles were protons, stable particles. As for their energy,
it was already clear that it had to be higher than some GeV.
Although some evidence of primary tracks due to nuclei heavier than protons
had been obtained, as we have seen, as early as 1940, it was only in the late 1940s,
however, that further measurements with nuclear emulsions performed by Bradt,
Peters and others collaborators of the Rochester group, at great height in the atmo-
sphere, about 30 km, using sounding balloons, showed more systematically events in
which the primary interaction was produced by very dense tracks, typical of nuclei
6.2 Interactions in the Atmosphere and First Evidence of a Complex Primary … 121
heavier than protons [Freier1948a, Freier1948b, Bradt1948]. A systematic work on

the abundance of such nuclei in the primary radiation was published by Bradt and
Peters in 1950 [Bradt1950]. A spectacular image of one of these events, taken from
their work, is shown in Fig. 6.3. In this picture we can see the interaction of a heavy
nucleus, identified as a Calcium nucleus (estimated atomic number Z = 20 ± 1),
which interacts with one of the nuclei present in the nuclear emulsion employed
(Silver or Bromine), producing a star with 42 outgoing tracks of charged particles.
Some of them are low ionization tracks, barely visible in the figure, and identified
as probable tracks of mesons created as a result of local showers, while other tracks,
with a high ionization density, show evidence of secondary particles produced in the
process. One of them produces a further nuclear interaction in the same emulsion,
less than 100 μm away from the first interaction. The energy of the primary particle,
in this event, was estimated to be greater than 100 GeV.
Fig. 6.3 A nuclear interaction event, produced at high altitude by a heavy primary, identified as
a high energy (> 100 GeV) Calcium nucleus, with one of the nuclei present in the emulsion. The
interaction produces dozens of secondary particles, one of which is capable of giving rise to a
further nuclear interaction a few tens of microns away [Bradt1950]. Figure reproduced from the
work of H. L. Bradt and B. Peters, The heavy nuclei of the primary cosmic radiation, Physical Review
77(1950)54. Copyright (1950) by the American Physical Society, License No. RNP/22/JUN/054397
Postponing for the moment a detailed analysis of the composition of the primary
radiation, which will be resumed later, we are here faced with the first quantitative
evidence of the fact that the primary radiation is not constituted only by protons,
although these are the most abundant component: in the primary radiation it is also
possible to find the presence of nuclei much heavier than protons, with mass greater
than that of oxygen, which raises further questions about the origin and nature of the
primary radiation. In this same work of 1950, in fact, Bradt and Peters discussed the
significance of the presence of these nuclei in the primary radiation in connection
to their origin (whether deriving from the Sun or the Galaxy) and the acceleration
mechanisms that can justify the observed energies.
If not only protons but also other complex nuclei were present in the primary
radiation, the question arose whether other stable components, such as photons and
electrons, could also be present. The search for these additional components in the
primary radiation occupied an important role in the activity of researchers in the
following decade. Research groups from MIT and the University of Minnesota in
the USA initially obtained negative results, until, about ten years later, the first
evidence was also obtained of the presence of electrons [Earl1961, Meyer1961] and
high-energy primary photons [Kraushaar1962].
A more complete quantitative discussion of the composition of the primary cosmic
radiation will be dealt with later, also referring to the most recent results obtained in
this field. However, we can now mention the fact that the other components represent
a total of less than 10% of the most abundant component, consisting of protons, and
that among these the greatest abundance belongs to the nuclei of Helium. The most
probable interactions that occur in the upper part of the atmosphere, and which
produce secondary particles capable of reaching even sea level, therefore originate
to a large extent from these light nuclei.
6.3 Production of Other Particles in Nuclear Interactions
It should be remembered that in those years the study of these first high-energy nuclear
interactions were the only means of investigating the interaction process between two
nuclei, with the exception of the known processes at much lower energies, typical of
radioactivity. For many years, therefore, before the advent of accelerators, the only
interaction processes that could be investigated, even at energies of the order of a
hundred MeV or GeV, were those that gave rise to these “stars” with multiple tracks
leaving the point of interaction.
These experiments also gave rise to the observation, during the 1940s, of entire
classes of new particles, as secondary products of these interactions, and on which we
will not go into many details. It will be enough to mention, for example, the evidence
of the existence of the neutral pion, which decays in very short times, mainly into
two gammas, observed in cosmic rays [Powell1950] and also in the first experiments
carried out with accelerators [Bjorklund1950], of the lambda neutral particle (Λ0 ),
which can decay into a proton and a negative pion, of the positive Sigma particle
6.4 The Role of High-Altitude Laboratories 123
(Σ + ), which decays into a neutron and a positive pion, and of the charged mesons
K± , which decay into a considerable variety of different channels. Over the years, the
possible rules were also enunciated that allowed to establish allowed and forbidden
modes of the various decay channels, based on the conservation of certain variables.
Furthermore, following these observations, the class of antiparticles expanded
enormously, consistently with Dirac’s theory. The first observations of events compat-
ible with the existence of antiprotons and antineutrons, related to cosmic ray physics
experiments, were then confirmed by high statistics experiments performed with
particle accelerators in the 1950s.
6.4 The Role of High-Altitude Laboratories
We have already seen how, from the very beginning of the study of the cosmic radi-
ation, the possibility to carry out measurements at high altitudes was considered
an important aspect, initially to understand the basic question of the extraterrestrial
origin of this radiation, later to characterize the intensity of the radiation as a func-
tion of altitude, finally to study the nuclear interaction processes by which primary
particles were able to generate secondary particles.
Although many measurement campaigns were organized using airplanes or by
means of balloons (firstly handled by human crews, later on by automatic devices),
the role of the laboratories located over the top of high mountains has always been
central to the studies of the cosmic radiation, as it offered the possibility of installing
detection devices larger and heavier than those that can be carried on board of balloons
or airplanes, and above all for the possibility of operating these devices for long
periods, in order to monitor the flux of cosmic rays over time or measure rare events.
The history of many of these laboratories and the role they played in the research
activity in cosmic ray physics, in the training of new scientists, in cultural exchanges
and in international collaborations is itself a topic of considerable interest [Korff1982,
Zanini2009].
Already in the early years after the discovery of the cosmic radiation, measure-
ments had been carried out in the mountains, creating small permanent observa-
tion structures. Victor Hess himself had installed measuring equipment in 1913 on
Hochobir, in Carinthia, at an altitude of 2040 m. After returning from the United
States, Hess organized a series of observations on the Sonnblick, at 3100 m, with
ionization chambers, which had to be recharged every 4 h and therefore required
continuous work shifts. The transport of materials, in particular of the iron shielding
of the equipment, required considerable manpower, due to the lack of suitable means
of transport that reached that altitude. Later, taking advantage of the construction of
a cable car on Mount Hafelekar (2334 m), Hess then built an Observatory on this
mountain [Hess1932], which was put into operation in 1931, and which over the years
became a reference point also for many other foreign researchers, as evidenced by
the messages left in the GuestBook of the Laboratory [Steinmaurer1982] (Fig. 6.4).
Fig. 6.4 Top, fragment of the GuestBook of the Laboratory on Mount Hafelekar. On the bottom
Victor Hess at the Laboratory, in November 1936, the period of the awarding of the Nobel Prize.
Source V. F. Hess home page, University of Innsbruck (http://physik.uibk.ac.at/hephy/Hess/hom
epage/)
Starting from the 1920s, many observation stations were built in various parts
of the world, as evidenced by a Report prepared in the mid-1950s [Korff1952,
Korff1982], which in its original version included many other details on the
characteristics and the organization of each station.
One of the first stations in North America for the study of cosmic rays was the
station located on Mount Evans, at 4350 m. In 1928 Millikan and Compton began
high altitude experiments in Peru, on Mount Huancayo (3350 m), near the geomag-
netic equator. Also in South America, on Mount Chacaltaya in Bolivia, one of the
6.4 The Role of High-Altitude Laboratories 125
Fig. 6.5 View of the high-altitude laboratory of Chacaltaya, in the Bolivian Andes. Source
American Institute of Physics, Emilio Segrè Visual Archives
highest stations (5200 m) for observing cosmic radiation was built. In Europe, several
observation stations were installed both in the Alps and the Pyrénées. One of these
stations was on the Jungfraujoch, reachable by a railway, at 3570 m, and from the
very beginning it had an international character, with contributions from Austria,
Belgium, France, Germany, Great Britain and Switzerland. Other observing sites in
Europe were that of the Aiguille du Midi (3600 m), in France, the Italian site Testa
Grigia (3380 m), that of the Pic du Midi, in the Pyrénées (2860 m), originally an
astronomical observatory but with equipment also for the detection of cosmic rays.
Survey stations were also built on the Asian continent, for example in Gulmarg,
India, as well as in the Hawaii. Appendix F reports a list of the main high-altitude
stations existing in the mid 50s (Fig. 6.5).
Of course, one of the problems of high-altitude installations, especially in that
period, was the accessibility of the site, in particular whether it could be reached by
suitable means of transport through roads, in which case the shipment of equipment,
even heavy, was relatively smoother. Other means of reaching the site, such as the
cable car, or manual transport, at least for critical parts of the path, greatly limited
the equipment that could be transported and the maintenance of the site itself. In
principle, altitudes of 5000–6000 m represent the upper limit for a reasonable access
to an observation site. These altitudes correspond approximately to half the thickness
of the Earth’s atmosphere.
High-altitude installations have been, and still are, useful even for other disci-
plines, such as meteorology, geology, and also biology, in addition of course to
astronomy.
Chapter 7
Extensive Air Showers
Abstract A new and important step to understand the properties of the cosmic radi-
ation came with the observation of time- and space-correlated particles, originating
from the interaction of a high energy primary particle at the top of the atmosphere.
The existence of extensive air showers dates back to the observations made by Pierre
Auger in the Alps, with counters separated up to 300 m, although some evidence had
been observed by Rossi a few years before. The study of the properties of extensive
air showers opened an entire new area in the history of cosmic ray physics. Early
evidence of this phenomenon is discussed with reference to the original papers and
results, including the first measurements of the decoherence curve, i.e., the probability
to observe coincidences as a function of the separation distance between the counters.
The chapter then presents in a quantitative way some of the main features concerning
atmospheric showers: their intensity profile, the longitudinal and transverse develop-
ment, their time properties, and the density distribution of the associated secondary
particles. Results from modern simulations of extensive air showers based on the
Corsika package are also included in the chapter and further discussed in Appendix
G and I.
7.1 Secondary Processes and Local Showers
We have already mentioned the observations of individual tracks of particles by

means of the condensation chambers. Looking at the pictures of these tracks, it had
been sometimes noted the simultaneous presence of several tracks close to each
other. Such observations had been made for example by Skobeltsyn in 1928. From
the observation of these tracks Skobeltsyn had deduced that the tracks came from
regions close to the observation area, even if it was not clear whether they came
from a single point (vertex). In any case, this was evidence of processes in which
other secondary particles were produced. These secondary particles were most likely
electrons produced by the Compton effect in the material overlying the detectors.
A consequence of these processes is that the number of the secondary particles
depends on the thickness of the material, according to a “transition” effect: electrons
or photons of high energy, interacting with a material, especially materials with a high
https://doi.org/10.1007/978-3-031-24762-0_7
128 7 Extensive Air Showers
atomic number, produce in turn photons or secondary electrons. The number of these
by-products initially increases with the thickness of the material, as new electrons
are produced; however, by further increasing the thickness, the electrons produced
start to be absorbed into the material itself. Detailed measurements of this kind were
carried out by Rossi with different configurations of Geiger counters operating in
coincidence and displaced according to geometries that made it possible to highlight
these effects [Rossi1933a].
Figure 7.1 shows possible configurations of Geiger counters. In configuration A,
in which 3 counters are superimposed vertically, they can detect tracks of a single
particle which cross all the counters at the same time; configuration B is instead
made up of 3 counters arranged in such a way that a single particle cannot cross
them all, for example if the centres of their circular sections occupy the vertices of an
equilateral triangle. A possible coincidence between these 3 counters would therefore
give evidence of the passage of at least two independent particles, represented by the
dashed tracks. By analogy, a configuration of 4 detectors, if they are arranged with a
suitable geometry (C), would indicate the passage of at least 3 independent particles.
The appropriate configuration to be used can be optimized if the approximate point
at which these secondary processes can occur is known.
To study how the number of such secondary processes depends on the thickness
of the material placed above the detectors, Rossi studied the number of coincidences
between two or more detectors appropriately arranged, as the thickness of the material
placed above the detectors was varied, for example with lead plates of increasing
thickness, placing this amount of material at different vertical distances from the
counters. Figure 7.2 shows the data obtained by Rossi with a lead layer placed at a
distance of 14.6 cm from the counters.
As it can be seen, the insertion of a thin layer of lead above the detectors (1–
2 cm, corresponding approximately to 10–20 g/cm2 ) produces a notable increase
in the number of coincidences, while a further increase in the material thickness
leads to a slow decrease in the number of observed coincidences, due to the progres-
sive absorption in the lead itself of the particles produced, until reaching an almost
constant value. This is the so-called Rossi transition curve [Rossi1933a], an effect
Fig. 7.1 Geometric configurations of cylindrical Geiger counters (seen in section) operating in
coincidence, which allow to observe events due to the passage of a single particle (A), or of several
individual particles (B, C)
7.2 First Evidence of the Existence of Atmospheric Showers 129
Fig. 7.2 Rossi transition curve, represented by the number of hourly coincidences between several
detectors placed below a lead layer and operated in coincidence, as the interposed thickness of lead
varies. Data extracted from [Rossi1933a]
which has been also reproduced in many educational experiments in cosmic ray
physics [Blanco2008a].
The fact that the coincidence rate between the counters did not decrease to exactly
zero, even with a very large amount of lead, was for some time an incomprehen-
sible aspect of the phenomena related to the cosmic radiation. In fact, one would
expect that the increase in the thickness of material leads to a complete absorption
of the secondary particles produced by these “local” showers, due to the interaction
of a single photon or high-energy electron in the material, with the production of
secondary electrons. The coincidences, although rare, observed even with a large
thickness of lead, could not therefore be due to electrons, but to more penetrating
particles, the nature of which was later clarified. At the same time, the observation of
coincidence events even between counters placed at large relative distances helped
to clarify the nature of one of the most striking phenomena in the physics of cosmic
rays, namely the existence of atmospheric showers of enormous transverse size, able
of develop along the longitudinal direction in the various layers of the atmosphere.
7.2 First Evidence of the Existence of Atmospheric Showers
The evidence of correlated particles in the cosmic radiation present in the atmo-
sphere is generally associated with the measurements of Pierre Auger and collabora-
tors, published in 1938–1939 [Auger1938a, Auger1938b, Auger1939a, Auger1939b,
Auger1939c]. In summarizing the results of these measurements [Auger1939c] Pierre
Auger recalls how, in addition to the soft particles deriving from the decay of mesons,
there must be other spatially and temporally correlated components associated with
the showers created by the interaction of the primary particle. This coherence effect
between different particles can be somehow highlighted by coincidence measure-

ments between individual detectors, placed at a certain relative distance and able to
detect their arrival. Auger refers to a series of measurements, performed both at sea
level and at high altitude, on the Jungfraujoch (3500 m) and Pic du Midi (2900 m),
using Wilson chambers and counters, which had shown evidence of those that will
be immediately called Extensive Air Showers.
Some preliminary evidence of coincidence events between distant counters had
actually already been obtained by Bruno Rossi in 1933, while performing coincidence
measurements between two Geiger counters, in particular to study the dependence of
the number of observed coincidences on the orientation of the detector telescope, for
the study of the so-called East-West effect. Occasionally Rossi had noticed how some
coincidence events also occurred with counters arranged at a large distance between
them [Rossi1934b], in which case the number of observed coincidences had to simply
correspond to the expected number of spurious coincidences. Strangely, however, the
number of observed coincidences always exceeded the estimated number of spurious
coincidences, a phenomenon that Rossi describes as if from time to time showers of
particles arrived close to the detectors, producing coincidences. Rossi, as he himself
writes, did not have the time to investigate the phenomenon further, but certainly what
he observed was due to physical coincidences between particles of an extensive air
shower, which somehow represents first evidence of this phenomenology.
Bothe and Kolhöster at the same time that Auger’s results were published had
also observed similar phenomena [Bothe 1937; Schmeiser1938; Kolhöster1938],
placing Geiger counters at some distance. The data obtained by Schmeiser concerned
coincidences between relatively close counters, up to about one meter away, while
those obtained by Kolhöster reported the coincidence rate between two detectors
up to about 75 m apart [Kampert2012]. Both these datasets were obtained at sea
level. The trend of the number of coincidences observed as a function of the distance
represents what is usually called decoherence curve, and is linked, as we will see
later, to the characteristics of the shower, in particular to its transverse profile.
From the knowledge of electromagnetic showers (cascades) it was expected that
electrons and photons, components of these showers, should arrive at sea level sepa-
rated by small distances. For example, it was possible to calculate that the average
distance between two electrons belonging to a shower created at a height of 10 km
by an electron of energy equal to 10 GeV should be of the order of a few meters.
Such considerations had been made, among others, by Clay and Blackett. As just
mentioned, however, these observations by both Auger and the aforementioned
authors showed evidence of correlations between particles of the same shower up to
much greater distances than could be expected (Fig. 7.3).
By placing two or more counters in the air, at a certain distance, Auger had in
fact observed a certain number of coincidences, a number that decreased as the
relative distance between the counters increased. While the measurements reported
by the other authors had been carried out for relatively small distances, up to a few
tens of meters, Auger had carried out measurements up to much greater distances, of
hundreds of meters, still observing a small effect. It must be said that the possibility of
observing “true” coincidences is determined by the amount of “spurious”, or random,
Fig. 7.3 Pierre Auger

(1899–1993) in 1978. Source
Fonte: American Institute of
Physics Emilio Segrè Visual
Archives
coincidences, which always occur in a coincidence experiment between two or more

detectors, as discussed in detail in Chap. 5. The quantity of random coincidences
per unit of time, as we have seen, depends on the counting rate of the individual
detectors and on the coincidence “window”, that is, on the time interval within which
the signals produced by the two detectors are considered as simultaneous. In Auger
measurements, this coincidence interval was reduced, following the development
of the coincidence circuits by Roland Maze [Maze1938], to less than 5 μs. In the
detector configuration used by Auger, it could be estimated that with two counters of
area equal to 200 cm2 each, the expected random coincidence rate, for measurements
done at high altitude, was about 1 per hour.
With an apparatus consisting of two counters placed horizontally, the number
of coincidences was measured by Auger at various distances, up to about 300 m,
as shown in Fig. 7.4. As can be seen, the probability of observing such coinci-
dences rapidly decreases with increasing distance and can be related to the theo-
retical description of the interaction processes of cosmic rays in the atmosphere,
in particular with the resulting lateral profile of the shower. In the original figure
by Auger the experimental results were compared with the predictions of a model
describing the development of the “standard” cascades [Euler1940]; consistent devi-
ations were observed for distances of more than about ten meters, testifying to the
fact that different processes intervened with respect to the electromagnetic showers
produced by individual electrons and photons. According to these processes, the
probability of having coincidences should have been reduced to practically zero for
distances beyond a certain value—about ten meters—while true coincidences could
Fig. 7.4 Number of hourly coincidences observed between two horizontal counters, as a function
of their relative distance, obtained from the first Auger experiments
be still observed, albeit in a limited number, up to enormous distances, of hundreds

of meters.
The observation of coincidences even between very distant detectors, at distances
greater than those that can be expected from the separation between electrons in a
local shower, led to the conclusion that extensive showers of enormous size existed,
caused by processes other than the standard electromagnetic processes that produced
local showers. The comparison between measurements made at low altitudes and
those made at high altitudes, in which the number of coincidences at great distances
was significant, more than at low altitudes, naturally led to the idea of a shower that
developed throughout the atmosphere, originated from a single primary particle, with
a particle density of even several tens per square meter and distributed, at sea level,
over an area of enormous dimensions. Further evidence of the existence of correlated
particles associated with extensive air showers was also obtained in those years by
Lovell [Lovell1939] by means of cloud chambers triggered by Geiger counters.
In the years after the first evidence of these extensive air showers, a detailed work
to characterize their properties and behaviour was started, in a similar way to what had
been done for the individual particles of the cosmic radiation in the previous decade.
Thus, for example, Neher and Pickering [Neher1940] investigated, by means of a
certain number of Geiger counters operating in coincidence, the possible dependence
of the number of extensive air showers on varying latitude, finding no effect within
the limits of the statistical uncertainty associated with their measures. Auger himself,
with Jean Dadin, published in 1940 a short note [Auger1940] concerning the possible
diurnal variations in the number of showers observed. No striking effect was observed
in these measurements, with the exception of a slight increase in correspondence with
solar noon, which was presumably attributed to thermal causes.
A few years later, in 1942 [Auger1942], Auger also investigated the influence of
atmospheric pressure (therefore of particle absorption) on the number of showers,
identified through the coincidence between two detectors placed at a certain separa-
tion distance, 3 m and 13 m in the case of these observations. The measurements were
carried out both at sea level and at an altitude of 2900 m. At sea level, the dependence
of the number of hourly coincidences between two detectors placed at a distance of
3 m, on the atmospheric pressure, made it possible to obtain, albeit with high errors,
a barometric coefficient equal to 9% increase for a decrease in pressure equal to 1 cm
Hg. Another series of measurements, performed for two different relative distances
between the counters, led to a coefficient of (23 ± 5)% per cm Hg for a separation of
13 m and of (9 ± 2)% per cm Hg for a separation of 3 m. The different value of the
barometric coefficient reflects the different nature of the coincidences observed in the
two cases, being those measured at a short relative distance strongly influenced by
local showers, mainly composed of electrons and photons, while those measured at
greater distances are indicative of extensive air showers, with prevalence of muons.
Preliminary measurements of this kind had also been reported by Cosyns in 1940
[Cosyins1940], who had observed some evidence that the barometric coefficient was
higher for showers extending farther away. Additional considerations about the effect
of atmospheric pressure on the detection of secondary particles will be discussed in
Chap. 11.
Further measurements at different altitudes and on board aircraft had been already
carried out during the 1940s, also by Auger and collaborators. Within a few years, the
first models were also formulated to try to interpret the existence of these showers and
their properties, including the presence of high-energy mesons within the showers.
At the end of the 1940s, about a decade after their discovery, the knowledge of
extensive air showers nevertheless presented several problematic aspects. A reading
of the scientific papers of that period by the physicists most directly involved in this
activity, including Janossy and Auger himself, shows some of these problems.
In the work of Janossy [Janossy1949] it is recognized, for example, that exten-
sive air showers are mainly made up of electrons and photons but contain a certain
percentage of particles that are more penetrating than electrons. The nature of these
particles—as discussed in this work, also quoting the work of Auger and his group
[Auger1949]—seems to be unable to be traced back to that of ordinary mesons, but
rather to that of Λ mesons, or heavy electrons: particles with a mass equal to 3–10
times the mass of electrons. Experimental evidence had also been cited of the exis-
tence of these hypothetical particles, obtained by Cowan [Cowan1948] through the
observation of a track in a cloud chamber operated at high altitude. We recall that
only around the end of the 1940s was the role of mesons clarified, with the observa-
tion of the pion and its possible decay, so a unitary framework about the structure of
the showers was still being defined at the time.
7.3 An “Operational” Definition and the First Properties

of Extensive Air Showers
An excellent review of the situation concerning the results and problems concerned
with the extensive air showers at the end of the 1940s is that carried out by Giuseppe
Cocconi (1914–2008) in 1949 [Cocconi1949a]. Cocconi starts from the considera-
tion that extensive air showers are the only known phenomenon in nature in which
enormous energies are at play, between 1015 and 1018 eV (later, even larger energies
will be measured, to about 1020 eV), and believes that, after a decade since their
discovery, the results and interpretations are sufficiently developed to allow a clear
picture of the situation.
What is meant by an extensive air shower? To clear the picture of possible confu-
sions about the experimental description of what an extensive air shower is, Cocconi
provides a sort of operational definition: an event is to be interpreted as due to an
extensive air shower when we observe the simultaneous arrival in n (> 1) horizon-
tally placed detectors of at least one ionizing particle due to cosmic radiation, with
the condition that the horizontal distance between the detectors is at least 10/n–20/n
times greater than the average size of the detectors used, and in any case not less than
2–4 m. While this operational definition of an extensive air shower can eliminate
some showers of small size or made up by only a few particles, it certainly helps to
eliminate events that could be improperly considered to be extensive air showers, for
example single particles arriving to a detector and accompanied by some secondaries
generated in the air or locally produced. A set of Geiger counters (or ionization cham-
bers) placed at a certain relative distance can therefore be used to signal the arrival
of a shower by meeting the requirements described above. The coincidence events
between sets of detectors operating in these conditions will therefore be indicative of
the detection of an extensive shower, with the presence of a large number of particles
scattered over a very large area, hundreds or thousands of square meters.
Cocconi then recalls that the detection of these showers has highlighted up to that
moment the existence of three components: a “soft” component, a highly penetrating
component, and a neutron component, even if the understanding of these different
components, especially for the last two, is not yet complete.
With regard to the soft component, it can be said that coincidence events between
small unshielded detectors (Geiger counters or ionization chambers) placed outdoors,
basically are coincidences due to electrons, which make up the vast majority of the
particles present in the shower.
An experimental characterization of the number of coincidences observed
between detectors is provided by the measurement of the so-called “decoherence
curve”, i.e., the probability of observing coincidences between two detectors as a
function of their distance d. The first Auger measurements around 1938, already
discussed (see Fig. 7.4), had shown evidence of coincidence events even between
detectors located 300 m away. In subsequent years, coincidence measurements
between distant detectors were carried out by Skobeltsyn, Zatsepin and Miller
[Skobeltsyn1947], obtaining evidence of events up to a distance of 1000 m.
7.3 An “Operational” Definition and the First Properties of Extensive Air … 135
Fig. 7.5 Arrangement of detectors used by Skobeltsyn and collaborators to measure coincidences
between pairs of detectors placed at large distances [Skobeltsyn1947]
The technique used by Skobeltsyn and collaborators consisted in arranging the

pairs of counters (1–2) and (3–4) superimposed at a small relative distance (1.5 m),
making a “local” coincidence between the two counters of each pair, and then look
for coincidences between pairs of distant detectors (Fig. 7.5). In doing so, the number
of spurious coincidences could be greatly decreased, as it involved quadruple rather
than double coincidences.
Taking advantage of this arrangement, Skobeltsyn and collaborators were able
to measure the decoherence curve represented in Fig. 7.6, using 4 sets of Geiger
counters, each with a relatively large sensitive area, equal to 1840 cm2 . The coinci-
dence time between the two counter pairs was about 4 μs, which allowed a spurious
coincidence rate of 0.03 per hour, against a measured coincidence rate, at a distance
of 600 m, equal to 0.6 per hour. The measurements shown in Fig. 7.6 were carried
out at a rather high altitude, at 3860 m, on Mount Pamir.
Another important property of extensive air showers concerns the density Δ of
particles (number of particles in the shower per unit area). Measurements aimed
at studying the density of particles were carried out by Cocconi and collabora-
tors [Cocconi1944a, Cocconi1946a] both at altitudes close to sea level and at about
2000 m above sea level, obtaining results that could be parameterized with a power
law of the type
N(Δ) = kΔ−γ
where N(Δ) represents the rate of showers with density > Δ, and γ represents an expo-
nent whose value depends on the observation altitude. The technique for extracting
the particle density in the shower consists in comparing the number of coincidences
obtained between n counters and (n − 1) counters in the same area and is described in
Appendix G. In the measurements carried out by Cocconi and collaborators, particle
density were estimated from about 10 to over 200 particles/m2 .
Fig. 7.6 Decoherence curve measured by Skobeltsyn and collaborators with the arrangement of
counters shown in the previous figure, up to relative distances of about 1 km [Skobeltsyn1947]
An important feature to be evaluated for extensive air showers is their abundance

at different altitudes: the first studies of the dependence of the rate of extensive air
showers on altitude date back to the early 1940s. Hilberry, for example, investigated
the variation in the number of showers observed, from sea level up to 4300 m on
Mount Evans [Hilberry1941]. The detection apparatus consisted of a set of 4 Geiger
counters, two placed in a telescopic configuration on each other and the other two
positioned at a distance of 1.25 m symmetrically with respect to the telescope. The
detectors were simply placed in a car (a station wagon of the time), which made
several stops between sea level and the maximum altitude reached. The number of
quadruple coincidences between the different counters, which can be interpreted as
the measured number of showers, is shown in Fig. 7.7.
Fig. 7.7 Number of hourly coincidences measured between different Geiger counters, as a function
of the altitude of the observation site [Hilberry1941]
7.3 An “Operational” Definition and the First Properties of Extensive Air … 137
As it can be seen from these results, the intensity of the showers increases a lot
with altitude, such that at an altitude of 4000 m the number of showers observed
is about 15 times higher than at sea level. Already in this work the results were
compared with a trend obtained by the assumption of a power law for the primary
energy distribution, with an exponent of the order of 2.6–2.8, which starting from a
certain altitude gave a good representation of the results.
Measurements at higher altitudes were carried out a few years later, using detection
equipment installed on board airplanes, up to an altitude of 12,000 m [Kraybill1948,
Kraybill1949]. The detection apparatus, consisting of nine Geiger counters of 75 cm2
active area, had been mounted for these measurements on board a B29 aircraft,
flying between the altitude of 6000 m and 12,000 m. The counters arranged in the
same line along the longitudinal part of the aircraft, at various relative distances,
were operated selecting different combinations of triple coincidences; under these
conditions the number of spurious coincidences was in the order of 0.03/h, while
the rate of measured coincidences was in the order of 10 coincidences/hour. A first
set of results was obtained by flying at a constant altitude (or rather at a constant
atmospheric pressure), as a function of the geomagnetic latitude, observing negligible
variations in passing from latitudes 0°–30° N to the interval 30°–63° N.
The number of triple coincidences as a function of altitude is shown in Fig. 7.8.
Compared to the data shown in the previous figure, these data, also measured at higher
altitudes, suggest the achievement of a maximum, or plateau, around the altitude of
7000–8000 m, followed by a subsequent decrease.
These characteristics, observed in extensive air showers, suggested that the soft
component present in the showers was made up of electrons and photons, generated
through electromagnetic cascade processes induced by very high-energy primary
electrons (> 1014 eV) in the upper part of the Earth’s atmosphere. This assumption
about the existence of very high-energy primary electrons is not necessary, something
Fig. 7.8 Rate of atmospheric showers (triple coincidences between Geiger counters located on
board an aircraft) as a function of the altitude [Kraybill1949]. The two datasets, with different
symbols, refer to different distances between the counters
of which Cocconi himself was aware, admitting that other primary particles can
also generate the same electromagnetic cascade, with the presence of electrons and
photons. This means that looking at the soft component of the showers is not the best
strategy to understand their nature, even if this component is predominant. Rather,
it is necessary to investigate the other components, though they have very small
abundances: the penetrating component and the neutron component.
For the study of the penetrating component, the detection devices can be similar
to those used for the soft component, i.e., combinations of counters arranged into
appropriate geometric configurations and operating in coincidence, with the condi-
tion that they are effectively shielded with an adequate amount of lead, at least 20 cm,
above the detectors and along their sides. Under these conditions, several observable
aspects proved effective in elucidating the main properties of atmospheric showers.
A first aspect concerns the contemporary existence of the soft component and the
penetrating one. In fact, measurements of the penetrating component in the showers
showed that a soft component was always associated with it and vice versa; in other
words, the two components were always present, so that it was not possible to explain
the soft component in the showers with the existence of a single process, such as
a pure electromagnetic cascade. Conclusions of this kind were reached following
the measurements carried out by Cocconi and collaborators [Cocconi1946b], and
by Treat and Greisen [Treat1948], through the use of shielded (for the penetrating
component) and unshielded (for the soft component) counters. Similar conclusions
were also obtained by Fretter [Fretter1948] by observations of events in cloud
chambers.
A second aspect concerned the relative abundance of the two components, soft
and penetrating. At sea level the penetrating component was about 2% compared
to the soft one, while this proportion increased with altitude, resulting in 3% and
4% at altitudes of 3200 m and 4300 m respectively, as obtained from the measure-
ments reported by Treat and Greisen [Treat1948]. These values, however, are only
indicative, as the abundance of the measured soft component strongly depends on the
thickness of the absorber surrounding the detectors. Furthermore, as it soon became
evident, this proportion between the two components is different in the different
regions of the shower. In fact, electrons and penetrating particles are created with
different mechanisms, which may result in a different angular distribution for the two.
The study of this proportion between the components can therefore effectively help
to understand the nature of the processes involved in their production, in particular to
understand if the structure of the extensive air shower includes a single “core” or can
have a “multiple core” structure, as some theories predicted at the time [Lewis1948].
As regards the neutron component, evidence of the existence of neutrons
associated with extensive air showers had already been obtained in those years
[Tongiorgi1948a, Tongiorgi1948b], by means of BF3 proportional counters, esti-
mating an abundance of this component (at least for low energy) similar to that of
the penetrating component—about 1–2% at sea level. Although at that time the exper-
imental data on the production of neutrons were still scarce, it seemed evident that
they were produced not by the soft component, but by a more penetrating component,
7.4 Towards a More Complete Description of the Formation of Extensive … 139
and that therefore a non-negligible fraction of this component was also constituted
by nucleons.
These and other arguments supported the hypothesis that the two components,
soft and penetrating, were the result of a unique process, which starts from the
interaction of a primary, most likely a very high-energy proton. Unified descriptions
of such processes had already been proposed at that time by Greisen [Greisen1948].
7.4 Towards a More Complete Description

of the Formation of Extensive Air Showers
In the aforementioned article by Greisen [Greisen1948] we can find a concise

but clear description of the generation of an extensive atmospheric shower, which
revolves around the existence of the neutral meson. By reviewing experimental
evidence of various kinds, from the first evidence of the existence of a neutral meson,
to the absorption of mesons in the subsoil, to the dependence on the altitude of the
number of local showers created for a different thickness of material positioned above
the detectors, Greisen concludes that a unitary description of these processes can be
made by assuming that a primary proton—or a heavy meson—interacts, by nuclear
interaction, with a nucleus of the atmosphere, producing groups of mesons, which
contain both charged mesons (of both signs) and neutral mesons, as well as nucleons.
Neutral mesons almost instantly decay into two gammas, thus initiating an electro-
magnetic cascade. A further contribution to the production of electrons and positrons
comes from the decay of charged pions into muons and from the decay of the latter
into electrons or positrons. The electrons in turn produce gammas through the mech-
anism of the emission of bremsstrahlung radiation and the positrons can annihilate
themselves with the electrons of ordinary matter giving rise to gammas. Due to the
very high energies, the particles of the electromagnetic cascade are emitted along
directions little different from that of the primary particle. Even the charged pions
emitted in these nuclear collisions are characterized by high energies, and therefore
are emitted mainly within a cone with a small aperture with respect to the initial
direction of motion. Through successive inelastic collisions with the nuclei of the
atmosphere, the charged pions can in turn produce a cascade consisting of mesons
and nucleons.
A typical structure of an extensive air shower is shown in Fig. 7.9. The most
penetrating components of this shower, apart from the weakly interacting neutrinos,
are represented by muons, capable of reaching the sea level, despite their average
life at rest, due to the relativistic dilation of time. The more energetic muons are
also capable of crossing a considerable thickness of solid material, such as terrestrial
rock, and therefore of being detected even at great depths in the subsoil, while the
electromagnetic component is absorbed much more easily even by a very small
amount of material.
Fig. 7.9 A qualitative representation of a typical atmospheric shower, with its main components.
The interaction of a primary particle gives rise to a nuclear interaction in which nucleons and pions
(charged and neutral) can be produced. The decay of the neutral pions into two gammas gives rise to
the electromagnetic component, with processes of creation of electron–positron pairs. The charged
pions can decay, producing muons, or further nuclear interactions. Muons and neutrinos represent
the most penetrating component of the shower
Much progress has been made since the 1950s, about the experimental study and
understanding of extensive air showers, up to the present time, in which some of
the most relevant problems concerning modern physics are strictly related to the
study of showers of extreme energy. Without going into too much detail of all the
mechanisms of interaction that contribute to determining the detailed structure of a
shower, we will briefly review some of their main properties, as they emerged from
the experimental activity carried out since 1950s.
7.5 The Study of Atmospheric Showers Since the 1940s
We know today that most of the atmospheric showers are created as a result of the
interaction of extremely energetic hadrons, which arrive isotropically on the Earth’s
atmosphere, coming from the outer space and giving rise to subsequent nuclear
collisions with the nuclei of the atmosphere, as a result of which a huge number of
secondary products are produced, which in turn can decay or interact to produce new
particles.
7.6 The Longitudinal Development of an Extensive Air Shower 141
As the shower propagates longitudinally in the atmosphere, along the initial direc-
tion of motion of the primary particle, it also spreads transversally. The shower
includes a hadronic component, in which, however, a significant part of the energy is
converted into electrons and high-energy photons, through the decay of the neutral
pions, thus contributing to the electromagnetic component. The proportion between
the abundance of the various particles that make up the shower changes along the
direction of propagation and is therefore different at different altitudes. Muons and
neutrinos represent the most penetrating component of the shower.
The development of the extensive air shower along the atmosphere is accompanied
by further physical processes (Cherenkov emission, fluorescence radiation, radio
waves) that have been used in more recent years as an additional tool to understand
the properties of the showers.
The longitudinal and transverse extensions of a shower essentially depend on the
energy of the primary. For low-energy primaries, the development of the shower stops
in the upper atmosphere, and many of the secondary products, with the exception
of muons and neutrinos, are unable to penetrate to sea level. On the contrary, for
extremely high energies, the development occurs down to altitudes close to sea level,
with the presence of the other components as well. The electromagnetic component
of the shower is in any case absorbed even by a modest layer of material, while
energetic muons are able of penetrating even into the subsoil down to great depths.
Furthermore, neutrinos, due to their reduced interaction cross section, can cross
an enormous thickness of rock—even the entire Earth—without being appreciably
reduced in number. In fact, their detection is extremely difficult and requires huge
detectors to reach a reasonable probability of interacting within their volume.
7.6 The Longitudinal Development of an Extensive Air

Shower
After the discovery of the existence of extensive air showers, one of the basic char-
acteristics that was investigated was the study of these showers at different altitudes,
both in terms of their abundance and their composition. A large number of measure-
ments were made using detectors positioned at different altitudes, from the sea level
to accessible mountains. For higher altitudes, up to about 12,000 m in height, balloons
and, above all, airplanes were used, capable of carrying a certain number of small
detectors operating in coincidence [Kraybill1948, Kraybill1949]. However, due to
the limited transverse dimensions of these arrays of detectors, it was possible to
characterize only showers with a small size.
One of the main properties of an extensive air shower is the number of particles it
contains, or shower size. The shower size is often expressed in terms of the number
of charged particles present, since gammas and neutrinos require different detection
techniques, or even in terms of the number of electrons/positrons only, since they are
the most abundant component at sea level. The shower size essentially depends on
the energy of the primary particle, as well as on the angle of inclination with which
it arrives in the atmosphere, and on the height at which the first interaction occurs.
The shower size grows with the energy E of the primary almost in proportion to it,
following a law of the type
Ne = αEn
where the exponent n is of the order of 0.9.

The longitudinal development of the shower—therefore the shower size at various
atmospheric depths—can be experimentally evaluated by observing the fluorescence
light in the air, produced by the passage of charged particles, or by the light produced
by the Cherenkov effect.
It has a characteristic trend, which starts from low values near the point of first
interaction, reaches a maximum at a certain atmospheric depth and then decreases
again near the ground (atmospheric depth of about 1000 g/cm2 ). The relationship
between altitude and atmospheric depth is discussed in Appendix H.
The position of the maximum of the shower, Xmax (atmospheric depth at which the
number of particles is maximum), progressively moves towards greater atmospheric
depths (lower altitudes) for greater primary energies, as can be seen in Fig. 7.10,
which reports one of the first detailed simulation calculations made by Grieder in the
late 1970s [Grieder1979].
In the first measurements, the uncertainty in the longitudinal profile was so high
as to allow only the estimate of the average value of the distribution and not the
distribution itself. A comparison between the experimentally obtained distributions
and simulation calculations of the development of the showers allows today to eval-
uate the goodness of the hadronic interaction models and other aspects of the overall
development of the extensive air showers. Usually, the statistical fluctuations from
Fig. 7.10 Shower size of extensive air showers, induced by primary protons of energy 1013 eV
(lower curve) and 1016 eV (upper curve), at different atmospheric depths, highlighting the
longitudinal development of the showers. Data adapted from [Grieder1979]
shower to shower, with the same energy and orientation of the primary, are high, so
it is necessary to consider the average information extracted from a large number of
individual shower events.
The longitudinal profile ultimately depends on the physical processes that govern
the development of the shower, both of the hadronic and the electromagnetic type,
therefore on the cross sections (probability) of these processes, on the multiplicity
of particles created in each interaction and on the proportion between the different
particles, as well as by the kinematical constraints which govern the production of
each individual particle species.
Profiles of this kind can be fitted with semi-empirical functions to parameterize
their behaviour. One of the widely used functions was introduced by Gaisser and
Hillas [Gaisser1977]:
( ) X max
λ
X
N (X ) = Nmax e(X max −X )/λ
X max
where X represents the atmospheric depth (expressed in g/cm2 ), Nmax is the number
of particles corresponding to the maximum Xmax , while λ represents a character-
istic depth (usually assumed to be 70 g/cm2 ). Figure 7.11 shows as an example a
typical longitudinal profile (apart from the multiplicative factor Nmax ), which can
be described by the previous parameterization, assuming a value of Xmax equal to
300 g/cm2 and a value of λ equal to 70 g/cm2 . Appendix I reports additional examples
of longitudinal profiles evaluated from parameterizations of this type.
In this relation it is assumed that the first interaction takes place exactly at an
atmospheric depth X = 0. However, it is possible to generalize this relationship by
Fig. 7.11 Typical shape of a longitudinal profile of an extensive air shower, as a function of the
atmospheric depth, according to the parameterization of Gaisser and Hillas, with Xmax = 500 and
λ = 70 (in g/cm2 )
introducing a further parameter X0 , which denotes the atmospheric depth at which

the first interaction occurs [Pryke2001].
The dependence of the position of the shower maximum on the energy of the
primary is characterized by a quantity which is called the elongation rate. This quan-
tity, introduced by Linsley [Linsley1977] is defined by the variation of the quantity
Xmax per decade of energy:
d X max
ER =
d log10 E
It therefore represents a measure of the displacement of the maximum of the

shower when the energy of the primary varies by a factor of 10. As an example, in
the range of energies between about 1017 and 1020 eV, the elongation rate extracted
from the data is of the order of 70 g/cm2 per energy decade [Walker1981].
Simulation calculations carried out using codes that make use of the most common
hadronic interaction models show that the atmospheric depth at which the maximum
of the shower occurs depends not only on the energy but also on the nature of the
primary, i.e., whether it is a proton, or an electron, a gamma or a heavy nucleus, as
well as on the angle of inclination with respect to the vertical.
The attenuation of an extensive air shower in crossing the thickness of air that
constitutes the Earth’s atmosphere can be discussed both in terms of the attenuation
of the rate of these showers at different altitudes, and in terms of the absorption of
the number of particles which constitute it. Although a complete description of the
development of a shower occurred in more recent years, the concepts of attenuation
length for the rate (rate attenuation length), Λatt (expressed in g/cm2 ), and length of
absorption for the number of particles (particle number absorption length), λabs (also
in g/cm2 ), are still used today, together with their inverse quantities μatt (cm2 /g) and
μabs (cm2 /g), i.e. the attenuation and absorption coefficients.
The attenuation of the rate, that is the frequency of showers with a given shower
size, as found in many experiments, follows, to a first approximation, an exponential
law with the atmospheric depth X measured along the axis of the shower:
I (X ) = I (X 0 )e− Λatt = I (X 0 )e−μatt X

X
so, the variation with the atmospheric depth X will be
d ln I (X ) 1
− = μatt =
dX Λatt
The absorption of the particles of the shower can be evaluated by assuming that
the shower size has a distribution represented by a power law
I (N ) = k N −(γ +1)
where γ is the spectral index, approximately equal to 1.7. It can be shown that the
variation with atmospheric depth will be given by
d ln N 1 1
− = μabs = =
dX Λabs γ Λatt
For inclined showers, it is necessary to consider the actual atmospheric depth

crossed, based on the angle of inclination. At a given altitude, or atmospheric depth,
an inclined shower will need to be more energetic than a vertical shower to produce
the same shower size. In other words, at the same atmospheric depth, the rate of
showers with a given shower size will decrease with the increase in the polar angle
of inclination according to the relationship
I (X, ϑ) ∼ (cos ϑ) X/Λatt
Measures to determine the attenuation and absorption coefficients (or the corre-
sponding attenuation or absorption lengths) were carried out starting from the early
1940s using small arrays of detectors at various altitudes or mounted on board aircraft.
Ground measurements at various altitudes were carried out, among others, by Auger
[Auger1939a, Auger1939b, Auger1945], Cocconi and collaborators [Cocconi1946a,
Cocconi1949a, Cocconi1949b, Cocconi1949c, Cocconi1949d], Broadbent [Broad-
bent1950], Hilberry [Hilberry1941], Treat and Greisen [Treat1948], and Antonov
and collaborators [Antonov1957], some of which have already been cited. These
measurements, as well as the first obtained on board airplanes, used very small
detector arrays (a few meters in size). Among the first measurements carried
out on board airplanes, we should mention those made by Maze [Maze1948],
Biehl [Biehl1949], Kraybill [Kraybill1948, Kraybill1949, Kraybill1954a, Kray-
bill1954b], Hodson [Hodson1952, Hodson1953a, Hodson1953b], and above all, an
extensive series of investigations organized by Antonov and collaborators in the
years from 1960 to the mid-1980s [Antonov1960, Antonov1964a, Antonov1964b,
Antonov1971, Antonov1975, Antonov1983], by means of detectors distributed over
larger areas (about 30 × 30 m2 ), making use of space located not only in the fuselage
but also along the wings of the aircraft.
The dependence on the zenith angle ϑ corresponding to the axis of the shower is a
further information that allows us to better understand the properties of the shower,
since the amount of atmosphere that the particles must cross to reach a given altitude
h is equal to X(h, ϑ = 0)sec(ϑ). For large zenith angles (shower axis close to the
horizontal direction) it is necessary to take into account the curvature of the Earth to
evaluate the thickness of air crossed, as it was done for the individual particles (see
Appendix J). The numerous measurements carried out already in those years made it
possible to establish that the angular dependence of the rate of the showers roughly
follows a relationship of the type
I (ϑ) = I (ϑ = 0)(cos ϑ)n

where the exponent n depends on the shower size (for example it is of the order of
8 for a particle density of about 20/m2 and increases up to n = 10 for a density of
250/m2 ). The exponent also decreases with altitude.
Historical measurements of the dependence on the zenith angle were carried out
in the 1940s by Hilberry [Hilberry1941], Cocconi and Tongiorgi [Cocconi1946c,
Cocconi1949e], Deutschmann [Deutschmann1947], Williams [Williams1948] at
altitudes between sea level and about 3000 m, and by Biehl [Biehl1949] and Kraybill
[Kraybill1954b] at higher altitudes, up to about 9 km, by means of instrumentation
on board aircraft. Further measurements, in different locations around the world
and at different altitudes, were also made in the following decades. As an example,
Fig. 7.12 shows an angular distribution (dependence of the shower rate on the zenith
angle) obtained at sea level by Bassi and collaborators [Bassi1953], for shower sizes
of the order of 105 –106 , by means of the time correlation between 3 detectors (liquid
scintillators, with a sensitive area of 600 cm2 ) placed at a relative distance of a few
tens of meters, in a right-angled triangle configuration. The distribution obtained
shows a strong dependence on the cosine of the zenith angle, compatible with a very
high exponent.
The measurement of the direction of arrival of the shower (in polar coordinates θ,
φ) allows not only to determine the dependence of the rate on the amount of matter
crossed, but also to carry out studies on the influence of the geomagnetic field and
look for possible effects of anisotropy, as we will see later. The direction of arrival of
an extensive air shower can be estimated by reconstruction of the individual shower
particle with tracking detectors or by the use of an array, observing the difference in
the time of arrival of the particles in the individual detectors of the array (Appendix
K).
We can define an “age” parameter s of a shower, which characterizes the stage
of development of the shower in the atmosphere. This concept derives from the
treatment of electromagnetic showers, but it can also be applied to hadronic showers
Fig. 7.12 Angular distribution (shower rates as a function of the zenith angle), obtained by Bassi
and collaborators [Bassi1953] at sea level. Data adapted from their work
7.7 The Transverse Development of an Extensive Air Shower 147
[Kamata1958]. In the case of an electromagnetic shower, the “age” parameter is

linked to the atmospheric depth X by the relation
3t
s=
(t + 2β)
where t indicates the ratio between X and X0 (= 1030 g/cm2 ), and β = ln (E/Ec ), being
E the energy of the primary, Ec the critical energy for the production of secondary
particles. Referring to the longitudinal development, as represented for example
in Fig. 7.11 or in the plots shown in Appendix I, the first part of the shower is
characterized by a parameter s < 1, while the second part, after the maximum, is
characterized by s > 1.
Showers generated in the upper part of the atmosphere are called “old” and have a
high value of the s parameter. They are generally produced by lower energy primaries
or heavier nuclei. On the contrary, “young” showers have a limited value of s, being
produced in the lower part of the atmosphere, and are generally associated with more
energetic showers or created by light nuclei.
7.7 The Transverse Development of an Extensive Air

Shower
An important property of extensive air showers is given by the spatial distribu-

tion of the associated particles. In the case of electron-generated showers, the parti-
cles produced are expected to be more concentrated around the axis of the shower,
with a lateral (transverse) profile governed by the Molière distribution, while in
the case of hadron-generated showers, we may expect a larger spread of the parti-
cles around its axis. Quantitative measurements of the transverse development of
extensive air showers were first obtained in the late 1940s, for example by Williams
[Williams1948] and Cocconi [Cocconi1949f].
In the experiment carried out by Cocconi and collaborators, for example, a few
detectors were used (each of them based on a multiple coincidence between different
counters, partly shielded by lead), to determine the “core” of the shower, while addi-
tional, segmented, detectors, located in various positions, were employed to measure
the density of particles (both of the electrons and of the penetrating component) at
various distances from the core itself. In this work we may find also the first exam-
ples of “pictures” of the shower distribution, obtained from the lighting of neon tubes
arranged geometrically in the array, which give an idea, albeit approximate, of the
presence of particles at various distances from the core.
The experiment, for that time, had a certain complexity, both in terms of the
number and type of detectors to be used (including a cloud chamber), and for the
location and management of the experiment itself. The detectors, in fact, installed
at an altitude of 3260 m, were positioned inside tents, and interconnected by long
cables, up to a length of about 200 m, which represented the maximum separation

between the detectors. The electronics required for the various detectors, as well
as multiple coincidence circuits, with resolution times of 25 μs, included a sort
of “trigger” capable of controlling data acquisition by the other detectors and also
taking photographs from the cloud chamber. The heavy materials that acted as screens
had masses of the order of tons and had to be moved together with the detectors.
Although the article is signed by only three authors, they explicitly acknowledge that
the management of this experiment, which lasted over two months, had required the
contribution of many other collaborators, both in the construction and management
phases.
The results of this experiment made it possible to quantitatively evaluate the
density of electrons at various distances from the core of the shower and to compare
the observed shape with the standard theory of electromagnetic cascades. This result,
deduced from the cited article [Cocconi1949f], is shown in Fig. 7.13.
In the original figure [Cocconi1949f], the data are well reproduced by the curve,
based on Molière’s theory, which at that point could be used to evaluate the overall
number of particles (shower size). As an example, Cocconi and collaborators cite the
characteristics of the largest observed shower, with a density of up to 500 particles/m2
Fig. 7.13 Particle density (soft component) estimated at various distances from the core, normal-
ized to that measured at a distance of 48 m, obtained by Cocconi and collaborators [Cocconi1949f].
The data are adapted from their work
at a distance of 100 m from the core, corresponding to a total number of particles in the
shower equal to 108 . Although with various limitations regarding the knowledge of
the primary interaction point in the atmosphere, from these data it could be estimated
a minimum energy of the primary equal to 1017 eV. By contrast, events originating
from showers with very small size, with a density of about 4 particles/m2 at a distance
of about 3 m from the core, corresponded to a total number of particles of the order
of 4000.
As for the penetrating component, this was estimated to be only 1–2% compared
to the soft component, therefore not easily measurable in single showers. It was
possible, however, to statistically estimate that the fraction of the penetrating compo-
nent compared to the soft one increased with distance from the core (e.g., 1.5% at
48 m and 2.5% at 96 m). The distribution of the penetrating component is therefore
“flatter”, extending up to large distances from the core, where it is more abundant
than the soft one, while the latter is more concentrated around the core.
The lateral widening of a shower at altitudes close to sea level can give rise to
very large dimensions for high-energy showers, which can cover areas of several
km2 , depending on the shower size. For example, it is illustrative to see maps of the
coordinates (X, Y) of the impact points of the individual particles of an extensive air
shower, to realize the number and distribution of these particles at a given observation
level. Experimentally reconstructing such a map would involve using a detector with
continuous coverage, sensitive to position and of enormous area, equal to the typical
size of a shower with that energy, which of course is not possible for large showers.
However, we can have a fairly precise idea of these distributions looking at the shape
of simulated showers, employing one of the simulation codes existing today, which
take into account the possible physical processes not only of the primary cosmic, but
also of the secondary particles generated during the various steps.
Figure 7.14 gives an example of these maps, showing the position (X, Y) of the
impact point in the horizontal plane, at the sea level, of the most abundant charged
particles in the shower (electrons and muons), assuming a shower produced in the
atmosphere by a vertical primary proton of energy equal to 1013 eV (10 TeV), whose
axis intersects the origin (0, 0). The simulations were carried out with the CORSIKA
code, one of the most used in the field of cosmic ray physics [Heck1998].
As can be seen, the particles—in this case several hundred—are distributed over
a very large area, although the majority is concentrated in a smaller area around the
origin. The number of secondary particles and the extension of the shower of course
increase with the energy of the primary particle.
The lateral profile of a shower can be quantitatively represented through the radial
distribution of the density of particles, which expresses the number of particles per m2
existing at a certain radial distance r from the axis of the shower. These distributions
are generally different for the different particle species contained in the shower.
Figure 7.15 shows a typical radial distribution for electrons and muons in showers
originating from primary protons of energy equal to 1012 eV, calculated on the basis
of air shower simulation codes.
Fig. 7.14 Map (X, Y) of the impact points at sea level of charged secondary particles (electrons and
muons) generated during the development of an extensive air shower created by the interaction of
a primary proton, of energy 1013 eV, incident vertically. Simulations produced with the CORSIKA
code [Heck1998]
Fig. 7.15 Average radial distribution of charged particles (electrons and muons) in showers gener-
ated by protons of energy equal to 1012 eV. A lower threshold in energy, equal to 10 MeV and 1 GeV
respectively, was assumed for electrons and muons. Simulations produced with the CORSIKA code
[Heck1998]
These distributions are in principle symmetrical with respect to the axis of the
shower; however, the effect of the Earth’s magnetic field can deflect charged low-
energy particles in different ways, creating asymmetry effects, which are generally
negligible for higher-energy particles.
The experimental measurement of the density of particles in different positions,
using detectors located on an appropriate area, can be used to estimate the position
of the core and the direction of the axis of the shower (initial direction of flight of
the primary particle). As an example, Figure 7.16 shows a two-dimensional map of
density values (number of particles per m2 ), calculated according to the CORSIKA
simulation code for extensive air showers [Heck1998].
The density values refer in this case to showers generated by protons of energy
equal to 1015 eV, with the axis at the origin (X = 0, Y = 0). The particles considered
are muons and electrons, with a minimum threshold in energy. The numerical values
show the average density in each cell (10 × 10 m2 in size), evaluated in the range
from X = −50 m to X = +50 m, and from Y = −50 m to Y = +50 m. As it appears
from the map, in the vicinity of the central cells, corresponding to the position of the
core, a density of about 70 particles/m2 may be estimated, which gradually decreases
moving progressively away from the core.
Fig. 7.16 Average density of particles (muons and electrons) evaluated in a 10 × 10 m2 grid,
around the axis of the shower, located at the origin. Showers generated by vertical primary protons
of 1015 eV energy. Simulations produced with the CORSIKA code [Heck1998]
Particle densities near the core can be very high, even thousands of particles
per m2 , in the case of very energetic showers, but they rapidly decrease with radial
distance. Their trend, in the case of purely electromagnetic cascades, was calculated
by Nishimura and Kamata [Kamata1958]. In the range of values of the age parameter
s between 0.5 and 1.5 this distribution can be written in an approximate form as a
function of s, using the dimensionless variable x = r/rM (ratio between the radial
distance perpendicular to the axis of the shower and the radius of Molière rM ) as
⎡(4.5 − s)
f (s, x) = x s−2 (1 + x)s−4.5
2π ⎡(s) ⎡(4.5 − 2s)
where ⎡ represents the Gamma function. More complex functions to describe these
distributions have been reported recently [Morales2019; Fadhel2021].
7.8 The Time Profile of an Extensive Air Shower
During the longitudinal and transverse development of an extensive air shower, due to
the differences in speed of the various particles, and especially because of the different
paths they follow because of the interaction processes and the various decays, not all
particles arrive at the ground simultaneously. In particular, those far from the shower
axis have on average a certain delay compared to the more central ones. We can then
imagine the front of the shower as having the shape of a disc with a slight curvature,
as shown in Fig. 7.17, which reports the results of a simulation for a shower induced
by a proton of 1015 eV energy. Each point in the plot represents the arrival time of
each single particle in the shower, associated with the distance X from the core. As it
is seen, on average the particles farthest from the core, at a distance of a few hundred
meters, have slightly longer times than those arriving in the vicinity of the core.
Due to the very high speed of the particles in the shower (close to the speed
of light), all the particles of the shower reach the ground within a very short time
interval, of a few nanoseconds near the axis of the shower, and slightly greater at the
ends of the disc. We can imagine this curved disc as having a certain thickness, which
increases with the distance from the axis, due to greater paths by the particles and
corresponding larger statistical fluctuations. The time profile of the shower represents
the distribution of the arrival times of the particles at sea level, with respect to the
arrival of the first particle. Although most of the particles are concentrated in times
of a few tens of ns, there may be a very long tail (hundreds of ns, up to a few
microseconds), produced by the slowest particles in the shower (low-energy muons
and nucleons, for example neutrons), while energetic muons, pions and kaons are
concentrated in the region with very short arrival times, also due to their short average
life. Figure 7.18 shows a typical temporal profile of a shower, obtained from Corsika
simulations, in the case of vertical protons with energy 1013 eV.
Relative timing measurements between detectors were first introduced by Bassi,
Clark and Rossi in the early 1950s [Bassi1953], using detectors based on liquid
7.8 The Time Profile of an Extensive Air Shower 153
Fig. 7.17 Characteristic shape of the wave front of a shower, generated by a proton of energy
equal to 1015 eV. Results obtained from CORSIKA simulations [Heck1998]. Each point in the
figure reports the arrival time (expressed in ns) of the individual particles in the shower, and the
corresponding distance X from the position of the core
Fig. 7.18 Distribution of the arrival times of charged particles (muons and electrons) in a shower
generated by a vertical proton of energy equal to 1013 eV. Times are measured with respect to the
first arrived particle. Results obtained from CORSIKA simulations [Heck1998]
scintillators, placed at distances between 3 and 30 m. They also made it possible to

investigate the spatial (or temporal) thickness of the disk representing the wave front
of the shower, and to extract, as already mentioned above, an angular distribution of
the shower with a certain shower size.
The temporal information on the arrival of the particles from a shower in several
detectors (at least three) placed at a certain relative distance, in fact, allows to evaluate,
on the basis of the sequence of arrival times, the direction of the shower axis by
triangulation methods (See Appendix K). Subsequently, in the 1960s, the technique
of relative timing between distant detectors was improved to the point of being able
to reconstruct the direction of the shower axis, therefore the direction of arrival of
the primary, with excellent precision. Arrays of detectors with good time resolutions
allows to determine this direction even with resolutions of the order of 0.1° if the
detectors are properly arranged.
Chapter 8
The Detection of Extensive Air Showers
Abstract The main experimental methods employed in the detection and recon-
struction of extensive air showers are here discussed and commented, with wide
examples of the techniques and analysis procedures originally employed for this task.
The structure and performance of the ground particle arrays, collection of detectors
to identify and reconstruct the arrival of a shower, are here discussed and a descrip-
tion is given of the main detection facilities which were used from mid-1940s. The
chapter also discusses the detectors arrays based on the Cherenkov effect, the fluo-
rescence detectors and the radio wave detection technique. A brief description of
modern arrays, even those built with educational purposes is also presented. Details
on how to reconstruct the shower direction by the relative timing associated to several
detectors is discussed in Appendix K.
8.1 Direct and Indirect Methods
From what has been said so far about the nature and main properties of an extensive air
shower, created by the interaction of a high-energy primary particle in the upper part
of the atmosphere, it is clear that the primary particle could in principle be detected
directly, by means of suitable detectors placed at very high altitudes, or, even better, on
sounding balloons or satellites, even outside the Earth’s atmosphere. Such detectors
would have the advantage of directly detecting the primary particle, to try to determine
its mass and energy. Such methods, called direct methods, however, have several
limitations. First of all, the dimensions and the maximum weight of the detectors are
strongly influenced by the fact that neither aboard balloons launched at stratospheric
heights, nor aboard satellites (or as happens today, aboard the International Space
Station) it is possible to install detectors of huge size. Since the flux of primary
particles is greatly reduced as their energy increases, direct detection is realistically
possible only up to certain energies of the primaries, of the order of 1014 –1015 eV.
Beyond these energies, indirect methods must be used, in which the information on
the primary is obtained from the properties of the shower created in the atmosphere,
by means of sets of detectors placed on the ground. In most cases, detector arrays,
based on solid or liquid scintillators, are used for the detection of the secondary
https://doi.org/10.1007/978-3-031-24762-0_8
156 8 The Detection of Extensive Air Showers
particles contained in the shower. However, showers also produce other effects in
the atmosphere that can be used for indirect detection, such as Cherenkov radiation,
emission of radio waves and fluorescence light. Each of these components has given
rise to specific techniques to detect the arrival of an extensive air shower and to
measure some of its properties. More recently, the possibility of using acoustic effects,
created by energetic and compact showers in dense media, such as water, has also
been explored. In this chapter we will briefly mention each of these techniques,
with particular reference to the use of particle detector arrays, due to its historical
importance and the wide diffusion of this method. A recent review, very extensive and
complete, of the characteristics of extensive air showers and their detection methods
is that of Grieder [Grieder2009].
8.2 Arrays of Particle Detectors
The most common method for the detection of extensive air showers and for the
reconstruction of their properties is to make use of suitable groups of detectors
installed on the ground—either at sea level or at higher altitudes—which can operate
in coincidence, indicating the simultaneous passage of individual particles belonging
to the same shower. As we have seen, the number of particles present in a shower,
even of not too high energy, is considerable and, for very high-energy showers, it
can reach values higher than a million particles. It is therefore clear, especially for
high-energy showers, that the detectors will only be able to detect a small fraction
of these particles, based on the number of devices used, their size and geometric
configuration. It is therefore a sampling of the shower, which is done in the positions
occupied by the detectors, trying to deduce from this sampling some of the general
properties of the shower itself. If the individual detectors are able to provide an
estimate of the particle density detected in different positions, these values can be
used to evaluate the position of the core (axis of the shower), the overall size of the
shower and the shape of the particle density at various distances from the core, i.e.,
its lateral profile. This requires that the detectors not only signal the passage of the
particles but also provide a response proportional to the energy deposited in their
volume.
In terms of the energy released by a single particle at the minimum of ionization
(for example a vertical muon), and assuming that this quantity is known for the
detector, the deposited energy may be used to give a signal proportional to the number
of equivalent vertical muons passed through its sensitive surface, and therefore to
the density of particles in that position. This (analog) information can be extracted
from scintillators, either liquid or solid, after an appropriate calibration phase. The
temporal information associated with the arrival time of the particles in independent
detectors can instead give information on the incoming direction of the particles and
therefore on the axis of the shower, as discussed in Appendix K.
The reconstruction procedures for extensive air showers however are not trivial, as
it is necessary to take into account different aspects which may distort the information.
8.2 Arrays of Particle Detectors 157
For showers induced by vertical or almost vertical primary particles, the pattern of
the distribution of the particles in the plane in which the detectors are arranged is
symmetrical with respect to the axis, within the inevitable statistical fluctuations,
and assumes a circular shape. However, for low-energy showers, the influence of the
Earth’s magnetic field can induce significant deviations from the original direction
of motion, and different effects can even be observed for positively charged particles
compared to negatively charged ones (separation of charges). Finally, for inclined
showers the pattern of the distribution of the particles on the ground no longer has
a circular shape, but rather an elliptical shape (Fig. 8.1), for obvious geometrical
considerations. Additional aspects can further distort the shape of the distribution by
producing asymmetries.
An array is therefore a set of N detectors arranged on a plane (although this is
not a completely necessary condition) in certain geometrical positions (xi , yi ), i =
1,… N. The geometrical arrangement and the number N of detectors determine the
topology of the array. For example, imagine a geometrical configuration in which the
detectors are located in a “hexagonal” geometry (Fig. 8.2). This has the advantage
of arranging the detectors with equal relative distances between each detector and
the six detectors surrounding it, with a repetitive arrangement of this pattern, which
can be replicated to include a very large number of similar detectors.
Fig. 8.1 Pattern of the

particle distribution of a
shower, in the case of
vertical (left) and inclined
(right) primaries. In the latter
case, the pattern takes on an
elliptical shape
Fig. 8.2 Typical hexagonal

arrangement of detectors in
an array for the detection of
extensive air showers.
Topologies of this kind have
been widely used to build
detector arrays, even with a
very large number of similar
detectors
This configuration has been, and still is, widely used even in large experiments for
the detection of extensive high-energy air showers. In other cases, it is preferred to use
a more compact structure (with shorter distances between the individual detectors)
in the central part of the array, and a more “looser” structure in the periphery of
the array, possibly with detectors with a larger sensitive area, to compensate for the
lower particle density.
The scale of the relative distances between the individual detectors and the overall
dimension of the array determine the ability of the array to be sensitive to showers
of lower or higher energy. More rigorously, generally on the basis of simulations, a
quantity called the acceptance of the array is evaluated, commonly expressed as a
function of the energy, or as a function of other parameters of the shower, such as
the shower size, zenith angle of arrival, location of the core, ... [Grieder2009].
Another important parameter to establish the operating conditions of a detector
array is represented by the choice of the event acquisition condition, what in the
language of experiments is called “trigger”. This is an arithmetic-logical condition
established on the basis of the information provided by the detectors, to determine
which are the “good” events to be acquired and which are to be discarded. The
choice of a trigger is essential, as it is rarely possible to acquire and store data from
all possible events, especially if there are many detectors in the array, and the rate
of events by each detector is high. As an example, if we had 50 detectors and each
had a single event rate of 100 per second, we would have 5000 events per second
to consider. However, these events would be of little significance in this context,
if the purpose is to detect the arrival of showers of a certain minimum energy. In
this case, a more appropriate trigger condition could be the coincidence—within a
certain time window—between at least k distinct detectors (for example k = 3 or 4,
...), a condition that greatly reduces the possible events to be acquired, preferentially
selecting showers of energy higher than a certain value. A further condition to be
defined in the trigger can be that of imposing a certain minimum energy deposited in
each detector, which corresponds to a density of particles above a given threshold. In
complex experiments even more elaborate triggers are used, based on the geometrical
arrangement and the type of detectors employed.
The time coincidence window within which to search for signals from multiple
detectors depends on the relative distance between the different detectors. If we
assume that most of the particles in a shower travel at the speed of light, and we
also want to detect showers with a direction of arrival close to the horizontal, we
must consider the time for the shower front to arrive from one detector to another.
Thus, with detectors 100 m apart, the time required for a particle moving at speed
c, to propagate from one detector to another, is of the order of 300 ns. In this case,
adequate time windows (at least 2 × 300 ns) must be chosen, in order to allow the
detection of coincident particles from any arrival direction.
8.3 Detector Arrays Based on the Cherenkov Effect 159
8.3 Detector Arrays Based on the Cherenkov Effect
For the detection of the Cherenkov light produced during the development of an
extensive air shower in the atmosphere, photon detectors can be used, with a wide
angular aperture, therefore with a considerable portion of the sky in their field of
view.
Cherenkov light is produced by the passage of particles that have a speed higher
than the speed of light in that medium. The Cherenkov effect was observed in the mid-
1930s by Pavel Cherenkov and theoretically studied by Franck and Tamm. These,
together with Cherenkov, obtained the Nobel Prize in 1958. A presentation of this
phenomenon in connection with the technique of observation of extensive air showers
has been reported, among others, by Watson [Watson2010]. Reminiscences of the
period in which this technique was first used are reported by Jelley [Jelley1981] and
by Thompson [Thompson2012]. Cherenkov himself was aware of the fact that the
emission of this radiation could be used for the detection of particles. The production
of Cherenkov light in the air was instead proposed by Blackett in 1948 as a technique
for the detection of extensive air showers [Blackett1948]. Blackett had estimated that
the intensity of this radiation should be very low, of the order of 0.01% of the total,
and therefore difficult to observe.
Galbraith and Jelley, following Blackett’s considerations, however, believed that
a large shower, containing a large number of particles, could produce enough
Cherenkov light to be detected. In 1953, making use of simple equipment (a metal
cylinder, a surplus WWII parabolic reflector and a 5 cm-diameter photomultiplier,
positioned in the focus of the reflector), they were able to observe signals associated
with cosmic rays for the first time [Galbraith1953], as they were coincident with
those produced in an array of large area Geiger counters by which the arrival of a
shower was detected. Measurements carried out in the high mountains during the
following years, in more specific conditions (for example with polarizing filters, to
verify if the light was really polarized, as expected for the Cherenkov radiation) gave
more significant and definitive results [Jelley1955].
In the case of air under standard conditions, the refractive index n is equal to
1.0003, only slightly different from that of vacuum; the minimum velocity that the
particles must have to produce the Cherenkov effect is therefore given by c/n =
0.9997 c. Cherenkov light is produced largely in the ultraviolet, within a small cone
around the direction of motion of the particle [Cerenkov1937], of angular aperture
θ governed by the relation cos θ = 1/βn, where β = v/c. A typical particle of a
shower can also produce 10–20 photons per meter of path, so the total number of
photons produced, taking into account the high number of particles present in a
high-energy shower, is enormous. One of the problems related to the use of this
technique is that the observational conditions of the sky must be more than good: a
low level of background brightness (light pollution), absence of clouds, atmosphere
free from dust and aerosols, low rainfall. Also, observations should generally be
limited to those nights when the moon is not visible. Overall, therefore, the fraction
of observing time (duty cycle) is quite limited. Each detector can determine the
number of photons and the time of arrival. The reconstruction of each event can be
done with the same method used for the charged particle detector arrays, once the
positions (xi , yi ) of each detector in the array are known. The substantial difference
with respect to charged particle detectors is that Cherenkov light detectors measure
the photons produced along the entire path of the shower, while the former detect
particles that have reached the ground.
It should be remembered that the same charged particles that reach the ground
can produce Cherenkov light inside the detectors, especially if they have an adequate
volume. A typical example are the water Cherenkov detectors, containers (tanks)
of suitable volume, filled with pure water, through which the charged particles can
produce photons, subsequently collected by photosensors, for example large area
photomultipliers. In the case of water, the refractive index is such that the threshold
for the Cherenkov effect is much lower (about 0.77 c). This technique was intro-
duced by Porter and collaborators in 1958 [Porter1958], using square section vessels,
measuring 1.20 m × 1.20 m, about 0.9 m deep, filled with demineralized water.
The internal surface of the containers was coated with a reflective material, and the
photons produced in the water were collected by a single large photomultiplier (5,, ),
mounted vertically and in contact with the water. With this technique, the energy
density was measured as a function of the distance from the axis of the shower, up
to distances of about 500 m. This technique is practically the basis for the large
detector arrays that were built in the following years, such as that of Haverah Park
[Tennent1967] and the Pierre Auger Observatory in Argentina [AUGER_OBS].
8.4 Fluorescence Detectors
The development of an extensive air shower also produces the emission of fluores-
cence radiation into the atmosphere. This is due to the de-excitation of the nitrogen
ions and molecules and is mainly concentrated in the band between 300 and 400 nm,
with an isotropic emission. Because of the isotropic angular distribution of this radi-
ation, it can be observed in principle from any direction (as opposed to the highly
directional Cherenkov radiation), using suitable optical photosensors. The first proto-
types of fluorescence detectors for observing atmospheric showers were developed
by Bunner [Bunner1967], as part of the activities carried out by the Greisen’s group
at Cornell University around the end of the 1960s. In these prototypes, a number of
photomultipliers were used that looked at the sky through a Fresnel lens (Fig. 8.3).
Despite the high number of photomultipliers (500, divided into 10 groups), with a
Fresnel lens of 0.1 m2 area in each module, this apparatus, operated for a long time,
was not able to adequately detect the associated fluorescence light upon the arrival
of a shower, also because of the high light pollution of the area where the apparatus
was installed.
A different design, based on large diameter mirrors (1.5 m), with a limited number
of photomultipliers placed in the focal plane of the mirror, was used with better results
in the mid-1970s [Bergeson1975], both because this design allowed a collection of
8.4 Fluorescence Detectors 161
Fig. 8.3 One of the modules

of the apparatus used by
Bunner [Bunner1967] for the
detection of fluorescence
light, using photomultipliers
that looked at the sky
through a Fresnel lens
light approximately a factor of 20 higher than in the previous design, and because the
apparatus was installed at Volcano Ranch, New Mexico, where a surface array also
operated, in an area with much clearer air. This is the apparatus that was later called
Fly’s Eye, made up of 79 modules, each including the mirror and a set of 12 photo-
multipliers. Detectors based on this design have proved suitable for the detection of
very high-energy showers, in particular for studies related to the determination of
Xmax , mass composition and estimation of the energy of the shower. Their use is in
any case affected by the particular atmospheric conditions (light pollution, air purity
...). In more recent times, the combined use of several such detectors has made it
possible to obtain stereo images, or, in coincidence with surface arrays, to use hybrid
shower reconstruction techniques.
8.5 Detection of Radio Waves Associated with Extensive

Air Showers
The possibility of associating the formation of an extensive air shower with electro-
magnetic waves emitted in the radio wave frequency band has been investigated for
at least 50 years, with mixed results.
On the one hand, the mechanism itself that should contribute to generating elec-
tromagnetic signals in appropriate radio frequency bands was not, especially at the
beginning, fully understood. On the other hand, it must be realized that any detec-
tion technique based on radio waves must take into account the enormous “noise”
existing in this part of the spectrum of electromagnetic waves, due to the presence
of natural and artificial sources (radio, TV, communication systems, …). This made
very difficult to consider the “inclusive” observation of showers, using the infor-
mation extracted from radio signals alone, while the possibility of observing these
signals in coincidence with those produced by conventional detectors, sensitive to
the passage of particles, appeared more promising.
A review of the experiments performed in the early period of this activity was
published by Allan [Allan1971], while a recent review of the status of this technique
was made available by Schröder [Schröder2017]. The first predictions of the emission
of radio signals from particles of an extensive air shower date back to the 1960s
[Askaryan1962], and the first experimental evidence came shortly after [Jelley1965;
Allan1966; Vernov1967].
In the work of Jelley and collaborators, for example [Jelley1965], the arrival of
the showers was signalled by a set of Geiger counters operating in coincidence, while
for the reception of radio signals, an antenna system consisting of 72 oriented dipoles
was used along the East–West direction, operating at a frequency of 44 MHz, with
a bandwidth of 4 MHz. The simultaneous observation of radio signals and showers
was observed with an oscilloscope, triggered by the arrival of a shower. Similar
systems, even if operating at different frequencies, were also used by Allan and
collaborators [Allan1966], in the context of the Haverah Park array, which already
allowed a “traditional” study of the showers, as the reconstruction of the shower
size and of the direction, by the use of particle detectors. The antennas used in this
study consisted of eight dipoles, arranged a hundred meters away from the particle
detectors. Over a period of about two months of observation, an analysis of radio
signals synchronized with showers showed evidence of 28 events compared to 625
showers detected. Measurements of this type were then carried out in the following
years by many experiments capable of detecting the arrival of extensive air showers.
In this first investigative phase, the frequencies used were typically tens of MHz
(frequency range still used today), but in some cases both higher (hundreds of MHz)
and lower (of the order of MHz or lower) frequencies were also exploited.
As a result of this first exploratory phase, qualitative evidence of the emission
of radio signals synchronized with the propagation of a shower in the atmosphere
was clearly established. From a quantitative point of view the information was not
very significant and above all it was not able to provide reliable information on the
8.6 An Example of Reconstruction of Extensive Air Showers in the 1950s 163
properties of the showers observed, so much so that interest in the 1970s towards the
use of this technique as a quantitative means of investigation slightly decayed.
However, a greater interest in radio technology has resumed in recent decades,
with a better understanding of the mechanisms of production of radio signals, with the
advent of digital technologies and with the use of large observational installations,
which now make the technique mature also for quantitative characterizations of
extensive air showers. It can be said that today almost all the large detection facilities
for the physics of cosmic rays have created an appropriate system for detecting
radio signals in coincidence with the other detectors of the array. From the point of
view of the performance of this technique, in a few decades we have gone from a
situation in which the observation of radio signals associated with cosmic rays was
little more than qualitative evidence of the existence of the phenomenon to a situation
in which the possibility exists that in some respects this technique could supplant
traditional detection techniques. Observations in the radio frequency band can also
be done in principle with a duty cycle of 100%, as they are not subject to limitations
related to light or the day/night cycle. Although traditional techniques will likely
not be supplanted, the combined use of radio techniques together with those based
on particle detectors, on the detection of Cherenkov radiation or fluorescence light,
seems the most promising way for the near future.
8.6 An Example of Reconstruction of Extensive Air

Showers in the 1950s
We have already mentioned the procedures for reconstructing the essential parameters
of a shower, based on the information produced by a set of detectors located over a
certain area. Many of these procedures are independent of the particular detection
technology used and were formulated as early as the late 1940s. They are basically
based on the measurement of the differences in the arrival times of the particles in the
different detectors [Bassi1953] and on the density sampling measured in different
positions [Williams1948]. In the works of the MIT group [Clark1957b, Clark1961],
for example, where Bruno Rossi also worked, reference is made to the procedures
used for the analysis of the data obtained from a small array of detectors, including
11 large area scintillators (diameter 1.05 m), positioned along two concentric circles
(5 detectors for each ring and a central detector), the largest of which had a diameter
of 460 m. In some periods a further group of 4 detectors was positioned along an
innermost ring, to detect even lower energy showers. We can refer to this detection
apparatus to understand some of the issues related to the reconstruction of showers
(Fig. 8.4).
The trigger condition in this experiment was given by the coincidence between at
least three detectors, with a signal of such amplitude as to indicate the passage of at
least 10 particles of the shower in each detector.
Fig. 8.4 Geometrical arrangement of the detectors in the array used by Clark and collaborators in
the 1950s [Clark1961]. The innermost ring (C) was used in some periods to detect showers of small
energy, while for most of the measurements the outermost rings (M, D) were used, which included
5 detectors each, in addition to the central detector. Figure reproduced from the work of G. W. Clark
et al., Cosmic-ray air showers at sea level, Physical Review 122(1961)637. Copyright (1961) by
the American Physical Society, License No. RNP/22/JUN/054686
The system used to simultaneously measure the amplitude of the signals produced
in each detector is of historical interest. We must remember that there were no
analog-to-digital converters at that time (the first prototypes of which were devel-
oped precisely in the period of the Second World War and initially covered by military
secret). The signals produced by each photomultiplier, after an appropriate amplifi-
cation, were sent to a corresponding number of oscilloscopes, whose screens, placed
side by side, could be photographed by a single camera at the arrival of an event,
in order to evaluate the amplitude (and therefore the particle density) measured in
each detector. Figure 8.5 shows the schematic photographic recording of a very high-
energy event measured with this system [Clark1961].
The direction of the shower axis was obtained by comparing the arrival times of
the particles in the individual detectors by means of a minimization procedure (see
Appendix K), with an uncertainty of the order of 5 degrees.
The lateral profile could also be determined for each shower, using different
possible parametrizations of the radial density of particles f(r). Initially the following
shape was used:
N e−r/r0
f (r ) =
2πr0 r
8.6 An Example of Reconstruction of Extensive Air Showers in the 1950s 165
Fig. 8.5 Photographic recording of the coincident signals observed in the oscilloscopes associated
to each detector, in the measurements performed by Clark et al. in the 1950s [Clark1961] for the
measurement of high energy showers with an array of scintillators. Figure reproduced from the
work of G. W. Clark et al., Cosmic-ray air showers at sea level, Physical Review 122(1961)637.
Copyright (1961) by the American Physical Society, License No. RNP/22/JUN/054686
where r0 is a free parameter, and r the radial distance from the centre, with coor-
dinates (X, Y). The parameters N (total number of particles in the shower), r0 and
the position (X, Y) of the shower centre were determined by a procedure which
minimized the deviations between the measured densities and those predicted by the
previous relationship. Later, other parameterizations were also used [Clark1961].
Based on these procedures, one of the first maps in equatorial coordinates (right
ascension, declination) of the directions of arrival of the reconstructed showers
was reported, for different classes of the total number of particles in the shower
[Clark1961]. This work summarizes the results obtained over a few years of measure-
ment with this apparatus, with the dual purpose of studying the energy spectrum and
the distribution of the arrival directions of the high-energy showers and at the same
time of investigating the detailed structure of these showers. Figure 8.6 shows one
of these maps, for shower sizes larger than 108 .
Fig. 8.6 Map in equatorial coordinates (right ascension, declination) of the arrival direction of
extensive air showers reconstructed with the procedure described by Clark et al. [Clark1961].
Figure reproduced from the work of G. W. Clark et al., Cosmic-ray air showers at sea level,
Physical Review 122(1961)637. Copyright (1961) by the American Physical Society, License No.
RNP/22/JUN/054686
8.7 Arrays for the Reconstruction of Extensive Air Showers
The construction of effective arrays for the detection of extensive air showers began,
as we have seen, in the period following the Second World War, in the early 1950s.
Measurements with small detector arrays had in fact been carried out already in the
mid-1940s, for example by Cocconi [Cocconi1944a, Cocconi1944b, Cocconi1946b]
(Fig. 8.7), with configurations (see Appendix G) that allowed to estimate the average
density of particles in an extensive air shower. Slightly more complex configurations
were used by Cocconi himself in the following years [Cocconi1949f], at an altitude
of 3260 m (Echo Lake), with several detectors deployed up to relative distances of
about 200 m, and by Bassi [Bassi1953], which we already have quoted in connection
with the reconstruction of some properties of the showers.
That atmospheric showers could have an enormous transverse dimension, even of
the order of km, had already been ascertained from various measurements carried out
at the end of the 1940s [Williams1948], in which it was possible to detect showers
with an estimated energy around 1015 eV, and in some cases, even greater. This gave
rise to the creation of arrays with a gradually increasing size, with a greater number
of independent detectors positioned within a larger area. In the mid-1950s the MIT
group, with Bruno Rossi and collaborators, used the array already described in the
previous section, with about fifteen detectors, each with a sensitive surface of about
one square meter, based on plastic scintillators, and displaced within a diameter of
460 m, replacing previous detectors that had been made with liquid scintillators.
8.7 Arrays for the Reconstruction of Extensive Air Showers 167
Fig. 8.7 Giuseppe Cocconi

(1914–2008). Source alchet
ron.com
The same group, with Clark, developed and operated in those years several medium-
sized arrays also in other locations in the world, partly moving the detectors already
made at MIT also in India and Bolivia, above all to look for evidence of possible
anisotropies or cutoff in the energy spectrum.
The era of large arrays began in the 1960s, with the proposals of John Lisley,
Livio Scarsi and Bruno Rossi, and the construction of the array of Volcano Ranch
[Lisley1961], near Albuquerque, in the New Desert Mexico: a set of 19 detectors
each with an area of 3.3 m2 , located 442 m away from each other, in a “hexagonal”
configuration with a maximum diameter of 1.8 km (Fig. 8.8).
This array was further magnified to an outer ring radius of about 3 km
[Linsley1963], observing events with an energy around 1020 eV for the first time.
In the same period, the BASJE array (Bolivian Air Shower Joint Experiment) was
built in Bolivia, at 5200 m altitude, in connection with a large area muon detector
(60 m2 ), particularly intended to observe possible showers poor in muons, which
could be indicative of high-energy gamma-induced showers [Suga1962]. The accep-
tance of the array, as it later proved, was however too small to reveal extremely rare
events such as those searched for.
The following years, from the second half of the 1960s onwards, saw the creation
of increasingly gigantic arrays, some of which are still in operation today. A complete
list of existing devices or those that have worked in the past can be found in the
excellent review by Grieder [Grieder2009] or by using resources available on the
Web, which summarize the status of existing facilities, for example [MPI]. In addition
to those mentioned, we cite below only some other examples of arrays that historically
represented a reference point in the first years dedicated to the measurement of
extensive air showers.
The Haverah Park facility, in Great Britain, was developed starting in 1967, on
the initiative of Alan Watson and collaborators, to measure high-energy showers,
from 4 × 1017 eV to 1020 eV. A complete summary of the activities conducted with
Fig. 8.8 Topological configuration of the Volcano Ranch array [Linsley1961], a set of 19 large
area detectors spread over an area of approximately 2 km2 . The figure shows, for the particular
event measured, with an estimated energy around 1019 eV, the reconstructed position of the shower
core and the density of particles/m2 measured in each detector. Figure reproduced from the work
of J. Linsley et al., Extremely energetic cosmic ray event, Physical Review Letters 6(1961)485.
Copyright (1961) by the American Physical Society, License No. RNP/22/JUN/054761
the Haverah Park array from 1968 until 1987, when data taking was interrupted, is
reported by Lawrence, Reid and Watson [Lawrence1991]. The Haverah Park array,
located at low altitude (220 m above sea level), consisted of a number of tank-based
detection units filled with water, with Cherenkov light-sensitive photomultipliers.
Four detectors, each with an area of 34 m2 , placed at the vertices of a triangle,
with the fourth detector in the centre, acted as triggers for the detection of events.
Other detectors, grouped into 6 sets of 4 detectors each, with an area of 13.5 m2 ,
were arranged approximately along a ring about 1.5 km away, around the trigger
detectors. Further detectors, singly or in groups, were arranged in the intermediate
space. The entire array occupied an area of about 12 km2 .
Given the complexity, type and number of detectors that made up these large
arrays, it should be remembered that these installations generally undergo an evolu-
tion over time, taking into account that the period of operation of these detection
facilities is measured on the scale of decades, during which new detectors are added
or replaced, the technology of detection, signal transmission and data acquisition
evolves, and various upgrades and improvements of the equipment take place.
In the years between 1968 and 1979 an array (Sydney University Giant Air shower
Recorder, SUGAR) remained in operation, located in Narrabri in Australia, at an
altitude of about 250 m above sea level. The array covered a total area of about 70 km2 ,
and was equipped with 54 underground detection stations, each of which included
two liquid scintillator tanks 50 m apart and with an area of 6 m2 . The detectors
were placed at a depth of about 1.5 m underground, corresponding to a threshold of

0.75 GeV for vertical muons [Brownlee1968]. Energy region measurements above
1017 eV were reported in later years [Winn1986].
Another array was built in the 1970s in Yakutsk (Siberia), with 58 surface detec-
tors and 6 underground detectors as well as 48 detectors for the Cherenkov radia-
tion produced in the atmosphere, reconfigured several times over the decades, with
stations spaced between 100 and 500 m, spread over a total area which over the years
was increased up to 17 km2 [Ivanov2009].
The AGASA (Akeno Giant Air Shower Array) was built in Akeno, 120 km from
Tokyo, on an area of about 100 km2 , between 1987 and 1991, with 111 scintillators
of 2.2 m2 area and 27 muon detectors of various surface [Chiba1992]. A previous
1 km2 —array, totalling 156 detectors with 1 m2 area, and 11 muon detectors, had
already been in operation since 1979.
In Australia, the Buckland Park Air Shower Array was operated in the early
1970s and included about 40 scintillators to detect showers with energy of the order
of 1014 eV [Clay1984]. A second array in the same area was built in the 90s, also
employed as a teaching tool for cosmic rays.
From the end of the ‘80s the GRAND array was operated at the University of
Notre Dame, made with a grid of 8 × 8 detectors, each of which equipped with
several floors of proportional wire chambers of about 1.3 m2 , which contributed up
to recent years to numerous aspects of physics [GRAND].
The Chicago Air Shower Array (CASA) was a large array, operated in the 1990s
in Utah (USA) [Gibbs1988]. It consisted of 1089 detectors, located on a square grid
of 33 × 33 detectors spaced 15 m apart. Each detector was based on 4 scintillators of
60 × 60 cm2 sensitive area. It was operated together with a second group of detectors
(MIA, Michigan Muon Array), consisting of 1024 scintillators with an area of 1.9 ×
1.3 m2 , installed underground, at a depth of 3 m, for the measurement of muons. The
set of these detectors (CASA-MIA) was able to detect showers above 1014 eV with
good efficiency and operated continuously until the end of the ‘90s.
The ARGO-YBJ array was built starting from 1996 in Tibet, at an altitude of
4300 m, with a layer of RPC (Resistive Plate Chambers) type detectors with a total
area of 6700 m2 , sensitive to position, and is particularly dedicated to the study of
high-energy gamma radiation.
A large array created in the following years was the KArlsruhe Shower Core and
Array DEtector (KASCADE) [Antoni2003], developed in Karlsruhe, Germany in the
1990s, and later extended as KASCADE-Grande [Apel2010]. These are large facil-
ities made with different types of detectors. The original KASCADE setup included
an array of 252 scintillator-based detection stations, located along a rectangular
grid with 13 m spacing, over a total area of 200 × 200 m2 , coupled to a central
detector (hadron calorimeter) and tracking detectors for muons. The KASCADE-
Grande extension, built later, included an additional 37 scintillator detector stations
(10 m2 ), located at a distance of about 130 m over an area of 0.5 km2 . This array was
operationally decommissioned in 2009.
Facilities of enormous size dedicated to the detection and study of very high-
energy cosmic rays are currently in operation, including the Telescope Array in
the USA, the GRAPES-3 observatory in India and the Pierre Auger Observatory in
Argentina [AUGER_OBS].
The Telescope Array is a hybrid detection system based on a surface array with
over 500 scintillators, spread over an area of over 700 km2 , with a distance of 1.2 km
between them, located in Utah at about 1400 m above the sea level [TA].
The GRAPES-3 experiment is an extension of previous facilities built in India.
Located in Ooty, it currently includes about 400 scintillation detectors and a huge
(560 m2 ) muon tracking apparatus [GRAPES-3].
The Pierre Auger Observatory is a hybrid detection system, which has undergone
a continuous evolution over time, reaching a total area of about 3000 km2 . Currently
it includes 1660 surface detectors, each consisting of a tank of 12 m3 volume, located
at about 1500 m relative distance, to read the Cherenkov light produced in the tank,
27 fluorescence detectors grouped in 4 stations, to detect the light produced in the
atmosphere, and further detectors of various nature to complement the information
provided by the main detection systems [AUGER_OBS].
Because of the complexity of these detection facilities and of their continuous
evolution, we refer to their website or specific publications for a detailed description
of their structure and the physics problems to which they are dedicated.
We also want to mention in this context the fact that in the last decades, a variety of
detector networks have been created for the detection of extensive air showers, with
mainly educational purposes, but also able to contribute to scientific investigations.
These arrays differ from each other for the characteristics of the detectors used,
for their topology and location, and for the associated educational and scientific
activities around them. A review of a variety of educational experiments on cosmic
rays, updated to 2008, is reported in [Blanco2008a]. As far as the organization of
detector networks is concerned, the following should be mentioned in this context:
• The NALTA consortium (North American Large area Time coincidence Arrays)
which groups together several arrays located in the North American and Cana-
dian territory (ALTA, CHICOS, CROP, WALTA, SALTA), which have often
operated jointly, based on small scintillation detectors installed at educational
establishments [NALTA].
• The HISPARC project (High School Project on Astro-Physics Research with
Cosmics) [HISPARC] in the Netherlands, also based on a large number of
scintillation detectors, distributed in as many upper secondary schools.
• The LAAS (Large Area Air Shower) project, created in 1996 and consisting of
a dozen stations—each with a certain number (4–12) of scintillators operating
in coincidence—located at a great relative distance, in order to also search for
coincidences among distant showers [Ochi2003].
• The Extreme Energy Events [EEE] project, which exploits the peculiarities
provided by large area telescopes based on Multigap Resistive Plate Cham-
bers, capable of reconstructing the trajectories of individual muons with
excellent efficiency and resolution. The telescopes of the EEE project have
been distributed since 2007 in almost a hundred schools in the Italian terri-
tory and have been operational since that date, also searching for long-
distance correlations (hundreds/thousands of km), and contributing to a wide
range of educational and scientific activities [Abbrescia2011, Abbrescia2013,
Abbrescia2015, Abbrescia2016, Abbrescia2018, Abbrescia2021; Riggi2017;
LaRocca2019, LaRocca2020a, LaRocca2020b].
Other small local arrays, mainly for educational use, have been built and are
operating in various parts of Europe, from Portugal to Poland, Greece and Sweden.
The use of very small detector arrays, located within a restricted area, with easily
modifiable configurations, has also proved to be a useful tool for carrying out a
variety of educational activities [Pinto2020a; LaRocca2020c; Riggi2021a].
Chapter 9
The Primary Cosmic Radiation
Abstract A more detailed presentation of the nature of the primary cosmic radiation
is reported in this chapter. The all-particle energy spectrum of the primary radiation,
which extends over a huge number of orders of magnitude, both in energy and
in intensity, is commented, together with the most recent findings from modern
experiments, up to the still open problem of the highest energy part of the spectrum
and the possible existence of an energy cutoff. The composition of the hadronic
component and the abundance of light nuclei in the primary radiation, as compared
with the natural abundance of elements in the solar system is discussed. A detailed
presentation is also made of the electron/positron and of the gamma components, as
well as a discussion of the intensity of the primary radiation at different heights in
the Earth atmosphere.
9.1 Introduction
We have seen how understanding the nature of the primary radiation played a central
role during the first decades of the history of cosmic ray physics. In some ways
we can say that it still occupies an eminent role today, long after the first insights
into its composition, although the problems have shifted to other fronts or to more
specific details. It is known, in fact, that what we call primary cosmic radiation
consists of particles and radiations that cover an enormous range of energy. Among
the most important components recognized today, in particular those which produce
extensive air showers in the atmosphere, we certainly have hadrons, gamma rays and
neutrinos, as well as electrons, even if the latter do not contribute predominantly to the
formation of extensive showers. The detailed composition of the primary radiation, in
the different energy ranges, is however still being investigated today, as is the energy
spectrum in its most energetic component. Other open problems are obviously those
related to the origin and production mechanisms of the different components, as well
as the possible effects of anisotropy or the existence of point sources.
As previously remarked, up to a certain energy it is possible to investigate
the different components by means of direct measurements, carried out on board
sounding balloons or satellites. At higher energies, however, the flux of the primary
https://doi.org/10.1007/978-3-031-24762-0_9
174 9 The Primary Cosmic Radiation
cosmic radiation is reduced to such an extent that it is impossible in practical terms

to obtain meaningful measurements by direct methods. The energy border between
direct and indirect measurements is around an energy of 1014 eV for protons, but at
significantly lower values for electrons, around 1012 eV, and even less for gamma
rays, a few tens of GeV.
In the past, one of the problems concerned with the measurement of the energy
spectrum of the various components was the gap existing between the range
exploitable by direct measurements and that by indirect measurements. The more
energetic particles that could be measured directly, in fact, produced showers that
the large detector arrays found difficult to measure, as their design was oriented
towards the detection of those of even higher energy. In more recent times, the possi-
bility of extending direct measurements to higher energies (using sounding balloons
and longer measurement times), combined with the possibility of extending indirect
measurements to lower shower energies, have made it possible to fill this gap.
Beyond the usual distinction between primary and secondary radiation, which is
made in connection to the peculiar phenomena that occur in the Earth’s atmosphere,
we can somehow say that the “primary” radiation, which affects the top of the Earth’s
atmosphere, includes stable particles and nuclei with a sufficiently long average
life. From the astrophysical point of view, this primary radiation is associated with
particles or radiations emitted “directly” by astrophysical sources, while we may
consider as “secondary” in this context the particles, or the radiations emitted in the
interaction of the primary particles with the interstellar medium. From this point of
view, protons and other light nuclei synthesized in stars, in addition to electrons, can
be considered as a true primary component, while other nuclei that do not derive from
stellar nucleosynthesis and are rather the product of successive interactions with the
interstellar medium, can be considered as secondary.
9.2 The Hadronic Component and the Energy Spectrum
As regards the hadronic composition, we have already mentioned that the large
majority are protons, which make up approximately 90% of the primary radiation.
Particles such as neutrons and mesons cannot be constituents of the primary radiation,
due to their instability, which would not allow to travel long paths, though in the
case of neutrons a local component could exist. A certain fraction of the primary
component is instead made up of light nuclei or nuclei of intermediate mass. This
composition, and the ratio between the different components, however, depends on
the energy range considered.
The hadronic component covers a very high energy range, from values below
1 GeV up to energies in excess of 1020 eV, limited to low energy by the existence
of a geomagnetic cut-off, at least if it refers to measurements performed near the
Earth. The flux of low-energy hadrons is subject to large temporal variations, both
periodic and aperiodic, due to the influence of the Sun. Such variations, on the other
hand, become negligible as the energy of the primaries increases. The (chemical)
9.2 The Hadronic Component and the Energy Spectrum 175
composition depends in detail on the energetic region considered, but it can be said
that in general it includes from protons to heavier elements. Direct measurements of
this component were carried out up to energies of the order of 1014 eV, using various
types of instrumentation placed on board balloons, satellites and more recently also
in space stations. For example, the JACEE [JACEE1999] mission, based on emulsion
chambers launched aboard balloons into the stratosphere, with a series of launches
between 1979 and 1996, investigated the composition and energy spectra of the
primary components at the top of the atmosphere. The typical detection configura-
tion included a set of nuclear emulsions, alternating with layers of plastic or lead
material, and grouped into various sections that performed the different necessary
roles: a primary charged particle detector, a target volume (in which the shower
was generated), and a calorimeter for measuring the energy of the shower produced.
We should once more remember that such measurements were at that time the only
possibility—before the advent of very high-energy accelerators, such as the Large
Hadron Collider—to study nuclear interactions in that energy range.
High-altitude balloon launches have been and continue to be a powerful means of
investigating the cosmic radiation—as well as other physical quantities of interest—
and have a long history in research. A fairly detailed list of the launches made from
the mid-1940s to today is shown on the [STRATOCAT] website.
In addition to nuclear emulsions, to be recovered and analysed by visual observa-
tion after they have been exposed to cosmic rays, the instrumentation used on board
satellites, especially in more recent times, has also made use of magnetic spectrom-
eters and electronic detectors, to determine charge and energy, or mass and impulse
of the primary particles.
At higher energies, over 1014 eV, apart from the measurements still making use
of nuclear emulsions in the high mountains, up to 5450 m, most of the data were
obtained from detector arrays, from the sea level up to altitudes of 5000 m. The use
of these arrays, as we have seen, is not in itself able to determine the mass of the
primary particle, but only the energy. However, the combined use of particle detector
arrays with Cherenkov light observation or other hybrid techniques has also allowed
in many cases an approximate estimate of the mass of the primary, allowing at least
the selection of events generated by protons, by light and by heavy nuclei.
From a huge number of different experiments and measurements, it has been
possible to obtain an inclusive energy spectrum (unidentified particles) in a wide
energy range. Figure 9.1 shows a collection of data from several experiments; the
plot shows the flux of particles in the range from 1 GeV to 1020 eV, expressed in
number of particles per (m2 s sr GeV). Data below 1 GeV, which typically show a
decrease (dip) around 10 MeV and then a further rise at even lower energies, are
strongly dependent on temporal variations and cannot be considered—in absolute
value—as significant. An interactive database for the extraction of cosmic ray data
has been reported in recent years by Maurin et al. [Maurin2014].
This plot, reproduced in various forms, is one of the most important classical
results of today’s cosmic ray physics, and it is worth commenting on several of
its aspects. It covers a huge number of orders of magnitude (12 in energy and 32
in flux). To get an idea of the measurable flux in practical conditions, it is usual
Fig. 9.1 Energy spectrum of primary particles (All particle spectrum), starting from the energy of
1 GeV, up to about 1020 eV
to report in various areas of the plot how many primary particles per unit of time
and surface correspond to the flux shown. Thus, for example, at an energy around
1011 eV (100 GeV), the flux corresponds to about 1 particle per m2 per second, a
relatively high value. However, the flux is greatly reduced by going to higher and
higher energies, reaching values of 1 particle per m2 per year around 1015 eV, and 1
particle per km2 per year around 1019 eV.
The shape of this spectrum is well represented by a power law, of the E−γ type,
where the exponent γ assumes slightly different values in the first energy region,
up to 1015 eV, where γ is approximately 2.7, and in the subsequent region, up to
about 1019 eV, where γ is about 3.15. The intersection between these two regions,
around an energy of 3 × 1015 eV, has traditionally been called knee, and the first
evidence of a change in the spectrum shape in this region dates back to the late 1950s
[Kulikov1958]. A slope parameter γ of 2.7 implies a reduction of a factor of 500 in
the flux for each decade in energy, which explains the extremely low values of flux
at very high energies. Around 1019 eV a further change in the slope of the spectrum
is observed, with a parameter γ coming back to a value of about 2.7. This region is
traditionally called ankle.
The trend of the energy spectrum of the primary component, well represented by
a power law, albeit with a slightly different exponent in the lower and intermediate
9.2 The Hadronic Component and the Energy Spectrum 177
Fig. 9.2 Energy spectrum of primary particles (All particle spectrum), scaled by the factor E2.6
[PDG]. The different symbols show the results obtained from different experiments. Figure repro-
duced with permission from R. L. Workman et al. (Particle Data Group), Prog. Theor. Exp. Phys.
2022, 083C01 (2022)
energy regions, suggests representing the flux values multiplied by Eγ , choosing an

appropriate value of the exponent γ (between 2.5 and 3). With this representation, it
becomes easier from a graphical point of view to verify whether the trend of the data
actually follows that law, as the values should ideally be arranged along a horizontal
line. The current knowledge of the spectrum of primary particles in the energy region
from 1013 eV upwards is shown in Fig. 9.2, taken from the documentation reported
by the Particle Data Group [PDG], in which an exponent equal to 2.6 has been used.
In this representation we see how in the interval between 1013 eV and about 3 ×
15
10 eV, the spectrum shows a much smaller variation, due to the multiplication factor
with energy, compared to the original spectrum. This type of representation also has
the advantage of better highlighting from a graphical point of view the regions where
it is possible to notice a change in the slope.
Regarding the characteristics of this change in the slope of the spectrum in the
energy region of the “knee”, several problems are still open, in particular whether this
change in slope is discontinuous or occurs continuously in a certain energy range,
and what is the detailed cause of the change, whether it is due to the superposition
of the spectra of the different particles, which have slightly different trends, or to
differences already in the mechanisms of production at the origin. In the energy
region of the “knee” an enormous amount of experimental data is now available
[Grieder2001, Grieder2009], obtained in particular in the ‘80s and ‘90s, with different
techniques and from dozens of different experiments, making use of detector arrays
with dimensions intermediate between the first arrays used in the past and the current
huge arrays.
The region of the spectrum at higher energies, from 1018 eV onwards, is obviously
characterized by an enormously reduced flux, which required huge arrays (thousands
of km2 ), and data taking times of the order of years, for its study. One of the main
problems related to the study of extreme energy cosmic rays is, as it is known, the
existence of a limiting value in the energy (cutoff). The idea of a limiting energy for
the primary particles capable of arriving on Earth was introduced in the 1950s by
Zatsepin [Zatsepin1951] and revived in the 1960s by Greisen [Greisen1966], and by
Zatsepin and Kuzmin [Zatsepin1966, Kuzmin1968] and is commonly referred to as
the GZK (Greisen-Zatsepin-Kuzmin) cutoff. This limit would be introduced by the
interaction of primary cosmic rays with the microwave radiation background, discov-
ered by Penzias and Wilson in 1965 [Penzias1965]. Protons of very high energy,
starting from 1020 eV, would in fact interact with the photons of the microwave
background, resulting in the photoproduction of pions, which becomes the domi-
nant process, degrading their initial energy and effectively eliminating the higher
energy component of these particles over large distances. Estimates of the mean
time required for this loss of primary proton energy to occur were made by Stecker
in 1968 [Stecker1968].
Figure 9.3 shows the effect of this interaction for cosmic rays of energy equal to
or greater than 1020 eV, as a function of the distance crossed [Anchordoqui2019].
Unless we admit the existence of very high-energy cosmic sources within relatively
small distances (50–100 Mpc, against the size of the order of 5000 Mpc for the entire
universe), this would prohibit the arrival of protons with energy higher than a certain
limit.
From an experimental point of view, only the huge arrays such as AGASA, HiRes,
the Pierre Auger Observatory and the Telescope Array have the potential to observe
very high-energy events, albeit with different techniques, which makes the compar-
ison between the data obtained critical. The results obtained by Auger so far seem
to indicate the existence of a cutoff, with the observation of very few events beyond
the cutoff, interpreted as due to heavy nuclei rather than protons. In contrast, the
AGASA results initially seemed to indicate many more events in that region, with
the predominance of protons, which would suggest a violation of the GZK effect. A
recent comparison between the results reported by the Auger and Telescope Array
Fig. 9.3 Effect of the

interaction of primary
cosmic rays with microwave
background radiation, with
consequent photoproduction
of pions, as a function of the
distance crossed
9.3 Composition of the Hadronic Component 179
Fig. 9.4 Spectrum of the primary cosmic radiation, in the region from 1018 eV upwards, scaled by
the factor E2.6 , measured by the Auger and Telescope Array experiments [PDG]. Figure reproduced
with permission from R. L. Workman et al. (Particle Data Group), Prog. Theor. Exp. Phys. 2022,
083C01 (2022)
Collaborations again shows (Fig. 9.4) a discrepancy in the spectrum measured by

the two experiments at energies greater than 1019 eV [PDG].
9.3 Composition of the Hadronic Component
In a complementary way to the study of the inclusive spectrum in energy, it is of

interest to know a more detailed composition of the primary radiation, while limiting
itself to the hadronic component. Measures of this kind, aimed at establishing the
abundance of the various nuclei, the energy spectrum of the individual components,
or even the isotopic composition (ratio between the various isotopes of the same
chemical species), were carried out as early as the late 1940s, although more signifi-
cant results were obtained only many years later. More precise results on these aspects
have obviously been obtained in the region with lower energies or in the case of light
nuclei (H, He …), since the flux at higher energies or for heavier nuclei decreases
rapidly, making observations with high statistics difficult to achieve. For this reason,
the different nuclei found in the primary radiation are often grouped together into
mass regions, for example light elements (Z = 3–5), medium mass elements (Z =
6–9), and so on.
Once again, the first answers to these questions came from measurements carried
out with nuclear emulsions aboard sounding balloons, in the late 1940s [Freier1948a,
Freier1948b, Bradt1948], a period in which the existence of nuclei significantly

heavier than protons was established in the primary radiation. Subsequent measure-
ments were then carried out by the use of satellites, at much greater distances from
the Earth, where the influence of the Earth’s atmosphere was negligible. Review
papers discussing the nuclear composition of the primary radiation, dating back to
the 1970s, are for example that of Shapiro and Silberberg [Shapiro1970] and that
of Webber [Webber1974]. More recent reviews can be found in [Grieder2001]. A
recent review, also available in interactive mode, of data relating to the abundance of
the different nuclei is that of Maurin [Maurin2014], in addition to the documentation
provided by the Particle Data Group [PDG], which is constantly updated. Figure 9.5
shows a recent status regarding the energy spectrum and the abundance of the various
species of nuclei, as reported by [PDG], summarizing a series of experimental data
obtained in recent times and by different experiments.
The low energy region (below some GeV/nucleon) is strongly influenced by the
measurement conditions, in particular by the solar period in which the measurements
are made, and this explains why different experiments, performed in different periods,
obtain values of the absolute flux that can differ even by an order of magnitude. Above
some GeV/nucleon, however, the spectra have a well-defined behaviour, regardless
of solar conditions. As regards the abundance of the various species, it can be seen
how protons dominate the yield compared to other nuclei, especially those heavier
than He. However, the abundance of the latter is not negligible, amounting to about
20% of the proton component at low energies (about 1 GeV/nucleon), but decreasing
to a few percent of the protons at much higher energies.
The integrated abundance of the individual components in the primary radia-
tion, from light nuclei up to Fe (Z = 26) is shown in Fig. 9.6, introducing a
threshold of 5 GeV to exclude the effects due to the modulation produced by the Sun
[Grieder2001]. As it can be seen, the nuclei beyond Helium have abundances between
2 and 4 orders of magnitude lower than the dominant component, the protons.
Beyond Iron (Z > 26) the presence of nuclei in the primary cosmic radiation,
although highlighted by more recent measurements, is very small, and, in particular
for elements with Z > 30, it is about 6 orders of magnitude smaller. This causes the
Iron to represent in some way the upper extreme of the chemical composition of the
primary radiation. Of particular interest is also the comparison between the abun-
dance of these nuclei in the cosmic radiation and the natural abundance of elements
in the Solar System, as determined to a large extent by the study of meteorites.
Figure 9.7 shows the relationship between the abundances of the various species
found in the primary cosmic radiation—assuming a minimum threshold in energy
equal to 5 GeV—and those of the elements present in the Solar System. As can
be seen, for some species the ratio is close to 1 (in other words the abundance in
the cosmic radiation roughly follows the distribution of the elements in the Solar
System), while for some of them the abundance found in the cosmic radiation is
much greater than what is present in the Solar System. It can be seen in particular
that the elements with Z = 3 (Lithium), Z = 4 (Beryllium) and Z = 5 (Boron) are
enormously more present in the cosmic radiation (from 10,000 times to a million
times more abundant). This phenomenon is interpreted as due to nuclear spallation
9.3 Composition of the Hadronic Component 181
Fig. 9.5 Review of recent experimental results concerning the abundance and energy spectrum of
the different species of nuclei present in the primary cosmic radiation. The insert also shows the
relationship between the abundance of hydrogen and that of helium, as a function of the particle
rigidity [PDG]. Figure reproduced with permission from R. L. Workman et al. (Particle Data Group),
Prog. Theor. Exp. Phys. 2022, 083C01 (2022)
reactions of the original primary nuclei when crossing the interstellar space. The
same phenomenon, albeit with smaller factors, also occurs for elements with Z =
21–25. In other words, the particles arriving on Earth are not necessarily “primary”
but can also derive from these secondary processes that occur along the enormous
journey they make.
The flux of the primary nucleons, which to a large extent are protons or nucleons
belonging to the nuclei of Helium, can be parameterized, in a wide range of energy
(from some GeV to over 100 TeV) by the following relationship

IN (E) = 1.8 × 104 E−α (GeV)particles/ m2 s sr GeV
where E represents the energy per nucleon, and α is approximately 2.7.

Fig. 9.6 Relative abundance of the various species present in the primary cosmic radiation,
normalized to the abundance of Carbon (Z = 6), assumed to be 100. Data extracted from
[Grieder2001]
Fig. 9.7 Ratio between the abundance of the various species present in the primary cosmic radi-
ation, compared to that present in the Solar System, assuming a threshold equal to 5 GeV. Data
extracted from [Grieder2001]
The mass composition of the higher energy primary cosmic rays cannot be inves-
tigated by directly detecting primary particles, due to the very low flux, and must be
studied through the development of the extensive air showers produced by them,
generally employing hybrid techniques, as discussed. In the knee region, many
measurements were made in the 1980s and 1990s, although in most cases it was
9.4 Electrons and Positrons 183
only possible to determine an average mass of the primaries, or to distinguish simply

between protons and much heavier nuclei, such as Iron. In particular, data obtained
for primary energies between 1011 and 1015 eV, up to the ‘90s, indicated an increase
with the energy of the fraction of nuclei such as Iron, while the component due to
protons remained approximately constant [Ichimura1993]. At the end of the 1990s
it was however quite clear that up to energies of 1015 eV the mean value of the mass
of the primary particles was quite low (about 7), while a sharp variation could be
observed beyond 1016 eV, where the average mass value reached roughly a value of
30. Results obtained in recent decades, especially by large detector arrays coupled to
hybrid techniques, have allowed to establish that the average mass value decreases
again for energies around 1018 –1019 eV. The study of the mass composition for
the higher energy cosmic rays, from 1019 eV upwards, is still under investigation
[Aab2017a].
9.4 Electrons and Positrons
Apart from the initial hypotheses on the nature of the primary cosmic radiation as
constituted by electrons, already discussed in relation to the events around the mid-
1930s, after the general acceptance that it consisted largely of protons and other
light nuclei, the attempts to establish the quantitative existence of an “electronic”
component in the primary radiation continued, but unsuccessfully, until about 1960.
Measurements carried out, for example, in the early 1950s by Critchfield et al. [Critch-
field1951], using cloud chambers carried at high altitudes by sounding balloons, had
allowed to establish only a maximum limit to the presence of electrons, of the order
of 0.6% of the total flux. Only in 1961, by Earl [Earl1961] and by Meyer and Vogt
[Meyer1961], in two different experiments with measurements always carried out
on board balloons, could the existence of a primary component due to electrons be
confirmed.
Earl’s work justifies the possibility of making new measurements—about 10 years
after Critchfield’s previous experiments—due to the fact that the experimental tech-
niques making use of cloud chambers, in particular keeping them on board balloons at
high altitudes have in the meantime improved significantly. The existence of electrons
and gammas could be identified based on the electromagnetic showers generated in
the lead plates overlying the different sections of the (multiplate) cloud chamber and
discriminated against nuclear interaction events. Figure 9.8 shows one of the events
in which the track of an ionizing particle at the minimum of ionization coincides
with the axis of a shower that develops in the following sectors, taken from the work
of Earl [Earl1961].
The result of Earl’s work led to an unambiguous, albeit with low statistics, evalu-
ation of the number of events due to the arrival of primary electrons, with an estimate
equal to (3 ± 1) % with respect to the proton flux, against the upper limit of 0.6%
set by the Critchfield measurements carried out about 10 years earlier. This discrep-
ancy was interpreted as due to several factors: the different threshold in energy, the
Fig. 9.8 An event identified as being produced by a primary electron in a “multiplate” cloud
chamber [Earl1961]. Figure reproduced from the work of J. A. Earl, Cloud chamber observations
of primary cosmic ray electrons, Physical Review Letters 6 (1961)125. Copyright (1961) by the
American Physical Society, License No. RNP/22/JUN/054867
different absolute flux of cosmic rays between 1949 and 1960 and the different alti-
tude. From this measurement a first rough energy spectrum of the primary electrons
was also extracted, within 3 GeV, with a preponderance for those of lower energy,
between 0.5 and 1 GeV.
In another work, by Meyer and Vogt, practically contemporary to that of Earl—
since it was received and published by the same Journal about a month after—the data
obtained from three different balloon flights, carried out in 1960, were used with a
different experimental setup [Meyer1961]. The latter was based on scintillators read
by photomultipliers, arranged in a vertical telescopic configuration, with further
plastic scintillators arranged laterally and operating in anti-coincidence with the
9.4 Electrons and Positrons 185
others. This apparatus made it possible to easily identify the primary protons, and
therefore to search for events that were not due to these particles.
Measurements performed at various altitudes, from about 2 to 300 g/cm2 of atmo-
spheric depth, made it possible to observe how the flux of these events as a function
of atmospheric depth followed closely the behaviour of secondary electrons up to
the altitude corresponding to the maximum point, at about 100 g/cm2 . At higher
altitudes, where the contribution of secondary electrons should have decreased, a
significant residual contribution was instead observed, which even increased at the
top of the atmosphere, as shown in Fig. 9.9.
Even in this case a rough energy distribution of these events was estimated,
observing how the major fraction of events corresponded to electrons of relatively
low energies (some hundreds of MeV).
As regards the energy spectrum of the primary electronic component, a review
of measurements made with different techniques in the second half of the 1960s
has been reported by [Fanselow1969]. A review of the data currently available can
be found through the documentation offered by the Particle Data Group [PDG]. As
an example, Fig. 9.10 shows the spectrum of electrons and positrons, multiplied
Fig. 9.9 The flux of events measured at high altitudes, between about 2 and 300 g/cm2 of atmo-
spheric depth, not attributable to events generated by primary protons, measured during balloon
flights by Meyer and Vogt [Meyer1961]. The dotted line describes the expected contribution
for secondary electrons, which at high altitudes (small atmospheric depths) should be negligible
compared to that observed. Figure reproduced from the work of P. Meyer and R. Vogt, Electrons
in the primary cosmic radiation, Physical Review Letters 6(1961)193. Copyright (1961) by the
American Physical Society, License No. RNP/22/JUN/054868
Fig. 9.10 Recent results about the energy spectrum of primary electrons, obtained from a variety
of experiments [PDG]. Figure reproduced with permission from R. L. Workman et al. (Particle Data
Group), Prog. Theor. Exp. Phys. 2022, 083C01 (2022)
by the factor E3 , known today, following the observations produced by a variety of

experiments.
As for the sign of the charge of these electrons, i.e., the fraction of negative
electrons and positive electrons (positrons) in the primary radiation, the first esti-
mates were made in 1964 [DeShong1964] by means of a magnetic spectrometer,
capable of separating particles with different charge states. The results gave different
values, albeit with large statistical uncertainties, depending on the energy: in partic-
ular a fraction of the positrons with respect to the total (electrons + positron) of
31%, 38% and 16% for the energy intervals 50–100 MeV, 100–300 MeV and 300–
1000 MeV respectively. Positrons therefore have a lower abundance than electrons.
More sophisticated measurements date back to more recent times, from the 1990s
onwards, confirming that the positron fraction, especially at high energies (> 1 GeV),
is less than 10% of the total, reaching values slightly higher (about 20%) for lower
energies [Grieder2001].
Today the ratio between positrons and electrons is known with excellent precision
over a wide range of energy, up to TeV, and shows that the positrons fraction reaches
a minimum around the energy of 10 GeV (about 5%), and then rises again up to
about 15% in the energy range between 100 GeV and 1 TeV. Figure 9.11 shows the
current situation, as available from the Particle Data Group documentation [PDG].
The black line shows the prediction of positron abundance based on a pure secondary
production mechanism, which largely underestimates the observed contribution at
energies above 10 GeV, while the other lines indicate the prediction of different
models, based on more complex mechanisms of astrophysical interest.
9.5 Other Components in the Primary Radiation 187
Fig. 9.11 Recent results about the fraction of positrons with respect to the total of (electrons +
positrons), obtained from different experiments [PDG]. Figure reproduced with permission from
R. L. Workman et al. (Particle Data Group), Prog. Theor. Exp. Phys. 2022, 083C01 (2022)
9.5 Other Components in the Primary Radiation
In this context we will not discuss, if not for a very brief mention, the other compo-
nents found as part of the primary radiation. Although they represent an enormously
smaller fraction from the point of view of the flux, with respect to the hadronic and
electronic fractions, they are nevertheless the subject of current investigations, also
in relation to other important aspects of astroparticle physics.
Among these components we must certainly include gamma rays, which at the
beginning of the history of cosmic rays seemed to made up most of the primary
radiation.
Experiments aimed at studying the characteristics of gamma radiation in primary
cosmic rays have usually been carried out using instrumentation aboard satellites in
space, but also aboard balloons, or even from the ground, using different tools and
methodologies, also suitable for the specific energy range of gammas and for the
expected flux. The lower energy component, up to energies of a few tens of GeV,
is in fact studied mainly in space or aboard balloons at the top of the atmosphere,
due to the absorption in the Earth’s atmosphere, which precludes its detection by
experiments on the ground. The latter are mainly dedicated to the detection of high-
energy gammas, from 100 GeV upwards. Recent reviews concerning the contribution
of gammas to the primary radiation have been reported for example by Lorenz and
Wagner [Lorenz2012] and by De Angelis and Mallamaci [DeAngelis2018b]. The
gamma radiation as a component of the primary cosmic rays is also important because
to some extent it retains the original imprint of its origin, from the point of view of
direction and energy spectrum.
Of particular interest, for experiments in space, is the study of the distribution

of the arrival directions (gamma rays maps), in order to look for evidence of point
sources or possible anisotropies. In fact, it should be considered that, unlike charged
particles, deflected by magnetic fields, gamma rays can in principle keep information
on their original incoming direction, even if after a more or less long path, they can
interact with interstellar matter and with the existing photon field.
At lower energies, in the X-ray region, evidence of this radiation from the Sun had
already been discovered in the 1940s. In much more recent years, it has been seen
that the Sun is also a source of gamma radiations, at energies of the order of MeV,
emitted in the context of nuclear processes. Only in the early 1960s, however, there
was evidence of X-radiations coming from outside the Solar System, by Roberto
Giacconi, Nobel laureate in 2002 for this discovery [Giacconi1962]. The first X-
ray sources highlighted, by X-ray instrumentation mounted on board a rocket, were
well-defined sources, such as Scorpius X-1, a low-mass binary star (one of which
is a neutron star), known as the strongest X-ray source visible in the sky. The X-
ray emission of this source is about 104 times more intense than that in the visible
region, and greatly exceeds the emission by the Sun in all other wavelengths. In the
following years, in 1964, the emission of X-rays was also observed by other objects,
such as the Crab Nebula. In addition to the emission from defined sources, a diffuse
background of X-radiation is also observed, which appears to be isotropic.
In the gamma-ray region, the first observations of photons with energies of the
order of 100 MeV of extrasolar origin occurred in the late 1960s, with satellite
observations (OSO-III in 1968, SAS-2 in 1972 and, above all, COS-B in the years
1975–1982 [Swanenburg1981]). The gamma radiation detected by COS-B had ener-
gies between a few tens of MeV and 5 GeV, and it was possible to distinguish between
the diffuse component and the localized ones.
One of the first gamma ray maps of the Milky Way was provided by the COS-
B mission in the second half of the 1970s. Other elements of interest for gamma
astronomy, in addition to the study of point, or well-defined, sources, are represented
by gamma-ray bursts, as well as by the study of gamma rays with specific energies,
related to the nucleosynthesis of the elements.
Diffuse gamma radiation of galactic origin is mainly due to the interaction of the
primary nuclei of the cosmic radiation with interstellar matter, through three main
mechanisms: the decay of the neutral pions produced in the nuclear interactions of
the primary nuclei with the interstellar medium, the bremsstrahlung processes of
the electrons, or the reverse Compton scattering of electrons with photons of the
galactic radiation field. The energy spectrum of the gamma radiation diffused in
the low energy range, through measurements obtained on board of balloons, up to
20 MeV, is shown in Fig. 9.12 [Schönfelder1980]. A review of the results obtained
from various experiments in the 1970s up to energies of about 100 MeV is reported
in [Fichtel1978].
Today, gamma astronomy has become one of the most promising fields of investi-
gation for the study of numerous celestial objects, either within the Milky Way or in
extragalactic regions, and is accompanied by other probes (cosmic rays, neutrinos,
9.5 Other Components in the Primary Radiation 189
Fig. 9.12 Diffuse flux of primary photons up to energies of about 20 MeV. The energy value
reported refers to the average value of each measured energy interval. Data extracted from
[Schönfelder1980]
gravitational waves) in approaches that make more and more use of the concept of
multi-messenger astronomy.
As far as the contribution of neutrinos is concerned, a component of which in this
context we will limit ourselves only to mentioning the existence, if we neglect the
continuous flux of low-energy neutrinos coming from the Sun and deriving from the
reactions that take place in it, high-energy neutrinos can be produced in a variety of
objects of astrophysical interest, and their study represents one of the leading aspects
of modern, rapidly evolving, astroparticle physics experiments. Neutrinos can be
produced in general by thermonuclear processes within stars, during their lifetime,
as well as in the collapse of stars that can lead to the formation of supernovae. From
this point of view, the observation, for the first time, of a neutrino burst associated
with the explosion of the supernova SN1987a, has aroused enormous interest in the
study of this type of processes.
As they interact weakly with matter, neutrinos carry direct information on the
processes that generated them and on their location. Of course, just because of the
very low probability of interaction, the possibility of detecting them is linked to the
availability of detectors of enormous volume, which has become possible in recent
times, also taking advantage of the possibility that they produce events of interest
for their detection in enormous quantities of “natural” matter, such as ice and sea
water, environments in which different types of equipment have been installed in
recent decades for the detection of neutrinos. For a review of the recent historical
steps regarding the creation of a true neutrino astronomy, a recent contribution has
been made available by Spiering [Spiering2012].
One of the other issues of considerable interest about the components present in
primary cosmic radiation concerns the presence of antinuclei, such as antiprotons and
antideutons. The first laboratory observations of antiprotons, using particle acceler-

ators, date back to 1955, by Chamberlain and collaborators [Chamberlain1955]. In
the same period, the hypothesis that such antiparticles could also be present in the
primary cosmic radiation was also discussed [Fradkin1956]. It is believed that such
antiparticles can be produced by energetic collisions of the primary radiation with
the interstellar medium; in this sense they would not really be “primary”. A possible
primordial origin for such antiparticles, for example as deriving from relevant
portions of antimatter present in the universe, are currently highly speculative. After
several unsuccessful attempts, between the end of the 1950s and the beginning of
the 1970s [Haskin1959, Aizu1959, Aizu1961, Grigorov1961, Apparao1967], using
nuclear emulsions, or with the aid of a permanent magnet [Bogomolov1971], the
first evidence was reported by Golden and collaborators [Golden1978, Golden1979a,
Golden1979b] and by Bogomolov and collaborators [Bogomolov1979], by means
of measurements carried out aboard stratospheric balloons, capable of reaching alti-
tudes corresponding to a few g/cm2 of residual atmosphere. In the measurement
reported by Golden [Golden1979a, Golden1979b], carried out with a magnetic spec-
trometer and a Cherenkov detector, some dozen events attributable to antiprotons
were observed, estimating a ratio between antiprotons and protons of (5.2 ± 1.5) ×
10–4 . The first Bogomolov measurements with evidence of events due to antiprotons
[Bogomolov1979] reported a ratio of (6 ± 4) × 10–4 in the energy range between 2
and 5 GeV. Values of the same order of magnitude, depending on the energy range
explored, from 0.1 GeV to over 10 GeV, were obtained in the ‘80s and ‘90s by the
same authors. In general, it was established that this ratio is slightly greater for higher
energies.
Today, modern experiments dedicated to the search for antimatter in cosmic radi-
ation use sophisticated equipment, in some cases with superconducting magnetic
spectrometers mounted on board satellites or the International Space Station (ISS).
Figure 9.13 summarizes recent results about the relationship between antiprotons
and protons, as well as between antiprotons and positrons, obtained by the AMS
Collaboration, which uses an apparatus mounted on board the ISS [Aguilar2016].
9.6 The Intensity of the Primary Radiation at Different

Altitudes
We have already mentioned in Chap. 7 the experimental studies to measure the

abundance of extensive air showers at various altitudes. This is somehow related to
the flux of primary radiation. More specific and direct measurements of the intensity
of primary radiation as a function of altitude were made as early as the second half
of the 1930s [Pfotzer1936, Schein1941].
In the work of Schein and collaborators [Schein1941], the result of which is
shown in Fig. 9.14, the intensity of the vertical radiation capable of passing through
a considerable thickness of lead, without however producing showers (curve A in the
9.6 The Intensity of the Primary Radiation at Different Altitudes 191
Fig. 9.13 Values of the antiproton/proton ratio as a function of rigidity. The antiproton/positron
and proton/positron ratios are also shown below. Figure reproduced from the work of M. Aguilar
et al. (AMS Collaboration) [Aguilar2016] (Creative Commons Attribution 3.0 License)
figure) showed that this radiation could not be constituted by electrons and not even
by mesons (due to their decay) but rather by protons, and that this intensity increased
continuously with altitude, unlike the trend of the total intensity—essentially elec-
trons and gamma—of the secondary radiation measured without absorbers (curve B,
by Pfotzer).
The flux of vertical protons decreases with atmospheric depth, following an
exponential attenuation law, with the consequence that the abundance of protons
at sea level, which represents the residue of primary radiation, is very small, with an
integrated intensity, for energies above 1 GeV, of the order of 0.9/(m2 s sr) [PDG].
The point where the interaction of primary cosmic rays first occurs in the Earth’s
atmosphere (first interaction point) assumes a distribution as a function of the
atmospheric depth X of the type:
dN 1
= e−X/Λ p−air
dX Λ p−air
where Λp-air represents the free average path of interaction of protons in the air, linked
to the cross section of interaction of protons with the nuclei of the Earth’s atmosphere.
Measurements of these cross sections were obtained from experiments carried out
with accelerators of gradually increasing energy, up to the current maximum energy
range covered by the Large Hadron Collider at CERN, which corresponds to energies
of the primary protons of the order of 1017 eV. On the other hand, at energies higher
than those obtainable in the laboratory, are rather the data from cosmic ray physics
to suggest the values of the interaction cross sections.
Fig. 9.14 Intensity of the penetrating radiation (identified mainly as consisting of primary protons,
curve A), compared with the intensity of the secondary radiation (curve B). Figure reproduced from
the work of M. Schein et al., The nature of the primary cosmic radiation and the origin of the
mesotron, Physical Review 59(1941)615. Copyright (1941) American Physical Society, License
RNP/22/JUN/054914
9.7 Possible Anisotropy Effects in the Primary Radiation
The primary high-energy cosmic rays, coming from outside the Solar System, have a
fairly isotropic distribution of the arrival directions, reflecting the diffuse propagation
in the galactic magnetic field along very long paths. At slightly lower energies, some
small effect of anisotropy can be expected, due to local irregularities of the magnetic
field or to other factors. The so-called Compton-Getting effect, introduced in 1935
[Compton1935] provides for a dipolar component in the distribution of the arrival
directions, due to the motion of the observer with respect to a reference system based
on an isotropic plasma of cosmic rays. For example, the anisotropy resulting from
the motion of the Earth around the Sun, predicted to be of the order of 10–4 , has been
observed in recent times [Aglietta1996, Amenomori2004].
Precision measurements of primary cosmic rays, in the energy range from a few
tens of GeV up to PeV (1015 eV), revealed small but measurable anisotropies in the
distribution of cosmic arrival directions, at the level of 10–5 –10–3 [Amenomori2006,
Guillian2007, Abdo2009, Abbasi2010]. As easily understood, such a small value
of anisotropy (1 part in 1000, or one part in 100,000) requires very high-statistics
measurements, in order to reduce the statistical fluctuations below the desired effects.
9.7 Possible Anisotropy Effects in the Primary Radiation 193
Just as an example, with 10,000 events measured in a certain angular interval, there
would be a statistical accuracy of 1%, and therefore only effects much more consis-
tent than 1% could be highlighted. Assuming that statistical accuracy improves
with the square root of the number of events measured, according to the Poisson
distribution, it is easy to understand that looking for variations of 0.1% (10–3 )
would require a measurement with several million events in any given angular
interval, and so on. If one wants to obtain a two-dimensional map in equatorial
(or galactic) coordinates of the intensity of cosmic radiation, the same statistical
reasoning must be repeated for each angular interval in two dimensions, making
this type of measurements extremely long and difficult, with a statistic that can
only be reached after years of data taking by large area experiments. However,
this is the case of some large current experiments [Amenomori2010, Abbasi2010,
Aartsen2016, Amenomori2017, Abeysekara2019], which are capable of measuring
1010 –1011 events/year, with the possibility of exploring anisotropies between 10–5
and 10–3 .
In the context of the search for deviations from the isotropic distribution, it is
then necessary to distinguish between the search for point sources, with an excess of
events coming from the same direction, and the search for widespread anisotropies
on a large scale, involving large regions of the sky. This issue has become particularly
relevant in recent years for extreme energy events, where some results have already
been obtained [Aab2017b].
Chapter 10
The Secondary Cosmic Radiation
Abstract The interaction of a primary particle at the top of the atmosphere produces
a complex secondary radiation, whose composition and properties are discussed in
this chapter. Electrons and positrons, gammas, muons, charged hadrons and neutrons
are the main components of the secondary cosmic ray radiation. The relative abun-
dance of these species at the various altitudes along the atmosphere, their intensity
profile, as well as the energy and angular distribution of each component constitute
an important part of the cosmic ray physics on Earth. A collection of experimental
data originating from pioneer measurements of these quantities, compared to most
recent findings is presented along the chapter. Different parameterizations of the
muon flux at the sea level are also discussed and compared in Appendix L.
10.1 Composition of the Secondary Radiation
As we have discussed, the development of extensive air showers in the Earth’s atmo-
sphere, generated by the interaction of primary high-energy particles (mostly protons
or light nuclei), determines the composition and properties of the secondary radiation,
i.e., the set of particles and electromagnetic radiation present at any given altitude
above the sea level, together with their energy and angular distributions. The devel-
opment of the shower is in turn determined not only by the possible interactions of
the secondary particles during their propagation in the atmosphere, but also by their
possible decays. It is therefore clear that the composition of the secondary radiation
itself and its properties will be different at the different altitudes with respect to the
sea level. We know, for example, that muons are subject to decay, with a lifetime at
rest of about 2.2 μs. Despite their speed, close to that of light, which allows them to
reach the sea level, due to the relativistic dilation of time, a certain fraction of muons
decay along their path in the atmosphere, altering the proportion between muons and
electrons at the various altitudes. Particles that have significantly shorter lifetimes
than muons have a negligible probability of crossing the atmosphere to reach sea
level.
Of particular interest for the physics of cosmic rays are the properties of the
cosmic radiation observable at sea level or at moderate altitudes, where most of the
https://doi.org/10.1007/978-3-031-24762-0_10
196 10 The Secondary Cosmic Radiation
experiments capable of detecting the secondary products of an extensive air shower

take place.
10.2 Muons
Cosmic muons mainly originate from the decay process of charged pions, and, to a
much lesser extent, from that of charged kaons, processes that occur in the atmosphere
during the development of an extensive air shower. Only very high-energy muons
(larger than TeV) can originate in processes involving other particles. The main
results concerning the muonic component in the atmosphere and at the sea level
concern the dependence of their intensity on the altitude, the abundance of the two
charge states, hence the ratio between positive and negative muons, their energy, or
momentum, spectrum and their angular distribution.
Measures aimed at investigating these properties have already been carried out
since a long time, as we have partially discussed in previous chapters, using counter
telescopes, magnetic spectrometers and other kind of detectors.
Referring first to the study of this component in the atmosphere, i.e., at altitudes
significantly above the sea level, we can divide the experiments into two broad cate-
gories: those carried out with the experimental setup fixed to the ground—even if at
high altitudes—and those carried out with some apparatus aboard balloons. Measure-
ments of the first type are for example those reported by Blokh and collaborators
in the 1970s [Blokh1977] at various altitudes (atmospheric pressures between 650
and 950 mbar, corresponding to altitudes between just over 500 and 3600 m), where
an intensity equal to about 1.8 cm−2 min−1 sr−1 at the highest altitude and about
0.3 cm−2 min−1 sr−1 at an altitude of 500 m was measured. In more recent years high
altitude measurements have been obtained by means of detectors placed on board
stratospheric balloons (MASS, IMAX, CAPRICE experiments in the ‘90s), able to
obtain the intensity profile of the muonic component in the atmosphere, from an
atmospheric depth of a few g/cm2 to the sea level, also separating the two charge
states.
The trend of the intensity vs atmospheric depth has shown that the profile grows
starting from low values at the top of the atmosphere to reach a maximum at atmo-
spheric depths around 200 g/cm2 (altitudes of 12 km), although with some differences
depending on the selected muon energy range. Figure 10.1 shows as an example the
intensity profile reported by Bellotti and collaborators [Bellotti1999] for negative
muons in the overall momentum range from 0.3 to 40 GeV/c.
High altitude measurements, at about 3000 m, of the muon energy spec-
trum were carried out from the second half of the 1950s [Kocharian1956,
Kocharian1959, Allkofer1964], and in the following years also at an altitude of
5260 m [Allkofer1965]. At even higher altitudes, measurements of the energy spec-
trum were obtained, again in those years, by means of sounding balloons and on
board airplanes [Baradzei1959]. More recent measurements were obviously carried
out in the ‘90s and beyond, with instrumentation installed on board stratospheric
10.2 Muons 197
Fig. 10.1 Intensity of the muon component at different atmospheric depths. Data extracted from
[Bellotti1999]
balloons up to an altitude of 36 km. In some of the measures mentioned, the charge

state of the muons was identified, obtaining a value of the ratio between positive
and negative muons, which, although with large fluctuations, also depending on the
muon momentum range, is close to the value of 1.3.
If we exclude low-energy particles, considering an energy threshold of the order
of GeV, muons are the most abundant charged component observed at sea level.
Most muons are produced in the upper part of the atmosphere (10–15 km of alti-
tude) following the processes already described. The muons produced continue their
travel in the atmosphere, losing part of their energy through ionization processes,
or possibly decaying. The energy lost by a muon in crossing the atmosphere can be
estimated at about 2 GeV, since they behave as particles at their minimum ionization
[with a specific energy loss of about 2 MeV/(g cm2 )] and must pass through an atmo-
spheric thickness of about 1000 g/cm2 . As a consequence of the different processes
(production, energy loss, decay) the muons have an energy distribution at sea level,
with an average value, for the vertical ones, around 3–4 GeV, and an angular distri-
bution that can be represented to a first approximation by a dependence of the type
cosn θ, where the exponent n is close to 2. The distribution in energy (or momentum
p) is however very different depending on the zenithal angle of arrival, as shown in
Fig. 10.2, which shows a comparison between the momentum spectra (multiplied by
p2.7 ) of the approximately vertical and highly inclined muons (zenithal angle 75°).
As can be seen from Fig. 10.2, which reports the results of a series of measure-
ments carried out by different experiments [PDG], the spectrum of highly inclined
muons, i.e., close to the horizontal, is shifted towards higher momenta. The path of
these muons, in fact, involves crossing a greater thickness of the atmosphere, with
significant differences in terms of the loss of energy suffered and decays.
Fig. 10.2 Momentum spectra of secondary muons observed at sea level, multiplied by the factor
p2.7 , in the case of vertical muons (solid symbols) and with a zenithal angle of 75° (empty symbols)
[PDG]. Figure reproduced with permission from R. L. Workman et al. (Particle Data Group), Prog.
Theor. Exp. Phys. 2022, 083C01 (2022)
The absolute value of the intensity of vertical muons at sea level is shown in
Fig. 10.3 in a wide momentum interval, from 0.2 GeV/c up to about 1000 GeV/c, as
reported by Allkofer and collaborators [Allkofer1971] from various measurements
carried out between the end of the ‘60s and the beginning of the ‘70s, also in order
to evaluate possible discrepancies with the data available in that period. As it is seen,
the spectrum presents a maximum for momenta slightly lower than 1 GeV/c and a
trend that to a first approximation, starting from a few GeV/c, follows a relationship
similar to E−γ , even if in Allkofer’s work the spectrum was parameterized with
slightly more complex functions, having different parameters obtained from the fit
of the data. Detailed measurements up to zenithal angles close to the horizontal have
also been reported by Crookes and Rastin [Crookes1972] in the same years.
Due to the enormous variety of experimental data concerning the muon spectrum,
it is useful in many cases to use simple analytical forms (parameterizations) that
best represent the trend of the data and can be used in calculations, for example in
simulation procedures, or for the evaluation of the response of the detectors. There
are different parameterizations of the muon spectrum, depending on their energy (or
momentum) and their zenithal angle. One of the most used, valid at high energies
and for zenithal angle not too large, so as to neglect the effect of the curvature of the
Earth, is the following, due to Gaisser [Gaisser1990], in which the energy of muons
is expressed in GeV:

dN 1 0.054
= 0.14E −2.7 cos θ
− 1.1E cos θ
cm−2 s−1 sr−1 GeV−1
d Ed 1 + 1.1E
115 GeV
1 + 850 GeV
10.2 Muons 199
Fig. 10.3 Vertical intensity of muons, extracted from various experiments reported by Allkofer
and collaborators between the end of the ‘60s and the beginning of the ‘70s. Data extracted from
[Allkofer1971]
Figure 10.4 shows a typical plot of the Gaisser parametrization, for vertical muons,
in which the vertical scale is multiplied by the factor E2.7 .
Fig. 10.4 Spectrum of vertical muons, multiplied by the factor E2.7 , according to the Gaisser
parameterization described in the text
Various other parameterizations have been proposed in more recent years, also
to extend the range of applicability of the Gaisser formula. Appendix L discusses
some of them and shows numerical examples of spectra calculated from these
parameterizations.
Measurements of the muon flux as a function of the zenithal angle, extended
to values close to the horizontal [Crookes1972] in the 1970s, made it possible to
establish that the intensity can be represented by a relationship of the type
I(θ) = I(θ = 0) cosn θ
with an exponent n slightly dependent on the momentum range considered, with an

average value around 1.85. Table 10.1 shows a summary of the values of the exponent
n extracted from different experiments [Riggi2020].
While the primary radiation is made up almost exclusively of positively charged
particles, in the secondary muon component it is possible to find both positive and
negative muons, with an excess in favour of the positive ones. The ratio between
positive muons and negative muons is of the order of 1.3–1.5 and depends not only
on the energy of the muons, but also—particularly for low-momentum muons—
on the geomagnetic latitude of the observation site and on the angle of inclination
of muons. Measurements of the charge ratio of muons had been made as early as
Table 10.1 Values of the exponent n relating to the relation I(θ) = I(θ = 0)cosn θ which describes
the trend of the angular distribution of muons at sea level or at moderate altitudes [Riggi2020]
Exponent n Experimental conditions References
2.1 > 0.33 GeV/c, sea level, 50° N [Greisen1942]
1.6 ± 0.1 > ∼ 10 GeV/c, underground, 38.3° N [Sheldon1963]
1.96 ± 0.22 > 0.7 GeV/c, sea level, 53° N [Judge1965]
2.16 ± 0.01 > 0.35 GeV/c, sea level, 53° N [Crookes1972]
2.2 > 0.35 GeV/c, 120 m a.s.l., 16° N [Karmakar1973]
1.91 ± 0.1 > 0.4 GeV/c, sea level, 12° N [Bhattacharyya1974]
1.85 ± 0.11 > 1 GeV/c, sea level, 12° N [Bhattacharyya1974]
1.75 ± 0.11 > 2 GeV/c, sea level, 12° N [Bhattacharyya1974]
2.23 ± 0.33 417 hg cm−2 underground, 13.0° N [Bhat1978]
1.95 ± 0.13 1200 m a.s.l., 35.4° N [Bahmanabadi2005]
1.884 ± 0.005 1.5 GeV, ∼ 150 m a.s.l., 55.6° N [Dmietreva2006]
2.15 ± 0.01 > 0.28 GeV/c, sea level, 19° N [Pal2012]
1.91 ± 0.07 1200 m a.s.l., 35.4° N [Abdhollai2013]
2.00 ± 0.14 (syst) > 0.11 GeV/c, 160 m a.s.l., 9.9° N [Pethurai2017]
2.33 ± 0.11 1200 m a.s.l., 35.4° N [Bahmanabadi2019]
1.91 ± 0.15 (syst) Sea level, 24.5° N [Arneodo2019]
1.83 ± 0.13 > ∼ 0.2 GeV/c, 3100 m a.s.l., 37.7° N [Riggi2020]
10.3 Electrons 201
the 1950s and over time they were extended to wider energy intervals, with greater
precision, in different locations (geomagnetic latitudes) and for inclination angles
even close to the horizontal. A large amount of data is now available, obtained under
various conditions. For vertical muons the Nμ+ /Nμ− ratio stands at values around
1.3, although with large fluctuations between different experiments, in a wide muon
momentum interval, from about 5 GeV/c up to 1000 GeV/c, with slightly lower
values for low momentum muons (< 5 GeV/c). For muons close to the horizontal
(inclination angles from about 80° upwards) the charge ratio Nμ+ /Nμ− increases
enormously, until it reaches values of the order of 10–20 for low momentum muons
close to the horizontal [Allkofer1979]. Finally, due to the East-West asymmetry, for
a given intermediate value of the zenithal angle (for example 45°) and for muons of
low momentum, below a few GeV/c, the ratio Nμ+ /Nμ− differs, showing values even
lower than 1 along the East direction, and values up to 1.5 along the West direction
[Allkofer1967].
10.3 Electrons
As already discussed, the secondary cosmic radiation also includes a “soft” compo-
nent, which unlike the “hard” one, made up of muons, is absorbed by even relatively
small thicknesses of solid material. Generally, by convention, a lead layer of 15 cm
(167 g/cm2 ) is assumed to define the demarcation threshold between the radiation
capable of penetrating through this thickness (hard component) and that which is
instead arrested (soft component). The soft component actually consists not only of
electrons and positrons, but also of gammas, whose properties are described in more
detail in the next section. The abundance of the soft component, or, more specifically,
that of electrons/positrons, with respect to the muonic component, strongly depends
on the selected energy range.
In the atmosphere, the electronic component is also partially affected by the pres-
ence of primary electrons. Primary electrons, however, constitute a non-negligible
fraction of the total electrons present only at very high energies and at the top of the
atmosphere (atmospheric depths up to a few tens of g/cm2 ). As an example, in the
energy range between about 50 and 100 MeV the primary electrons represent about
10% of the total electrons present up to atmospheric depths of 100–200 g/cm2 , while
they are negligible at greater atmospheric depths (lower altitudes). Only in the energy
range of the order of GeV the primary electrons are comparable in abundance with
the secondary ones. A review of numerous results obtained from various experiments
in the 1970s using high-altitude balloons, is reported by Grieder [Grieder2001].
The intensity profile of secondary electrons in the atmosphere also has a behaviour
to a first approximation similar to that of muons. At high altitudes in the atmosphere
(atmospheric depths starting from a few g/cm2 ) the intensity gradually increases
with the atmospheric depth, up to about 100–200 g/cm2 , where the intensity reaches
a maximum, then decrease again going towards the sea level. Figure 10.5 shows
a typical behaviour, measured by Fulks and Mayer in the 1970s [Fulks1974] by
Fig. 10.5 Intensity profile of secondary electrons in the energy range (54–85) MeV. Data extracted
from [Fulks1974]. The different symbols refer to different balloon ascents
instrumentation aboard balloons, for energies in the range (54–85) MeV, in which
the contribution of secondary electrons is preponderant with respect to that of the
primary. For gradually increasing energies, at the top of the atmosphere and up to
a depth of about 100 g/cm2 , the contribution of the secondaries gradually becomes
less significant than that of the primary ones.
The energy spectrum of electrons in the atmosphere, also measured in numerous
experiments in the 1970s at various atmospheric depths [Fulks1974], generally shows
a pattern of the E−γ type, where the exponent γ is approximately equal to 2 at low
atmospheric depths and assumes higher values going towards the sea level. A large
fraction of the electrons therefore has very low energies, below 100 MeV, even at
high altitudes in the atmosphere. Figure 10.6 shows the energy spectrum of electrons,
at two different atmospheric depths (291 and 637 g/cm2 ) [Fulks1974].
At sea level, comparative measurements of the soft and hard components had
been already carried out in the 1940s, for example by Greisen [Greisen1942], using
a telescope—orientable in the zenithal angle—with several Geiger counters in coin-
cidence, even separated by lead layers. The comparison between the number of
coincidences obtained with and without the interposed lead layers made it possible
to evaluate the proportion between the two components.
At an altitude close to sea level, in the case of vertical particles, the relative
intensities measured for the hard and soft components were respectively 2.27 and
0.59, thus corresponding to a fraction of the soft component of about 26% compared
to that of the hard component. Under the Greisen measurement conditions, the so-
called soft component included electrons with an energy greater than about 20 MeV.
This same apparatus had been used to carry out similar measurements, at altitudes of
1600 m, 3200 m and 4300 m, obtaining in this case fractions equal to 38%, 58% and
77% respectively. It is clear, therefore, that the abundance of electrons compared to
muons is lower at sea level, compared to what occurs at higher altitudes.
10.4 Gammas 203
Fig. 10.6 Energy spectrum of secondary electrons at two different atmospheric depths: 291 g/cm2
(empty triangles) and 637 g/cm2 (solid symbols). Data extracted from [Fulks1974]
The energy spectrum of secondary electrons, even at sea level, is highly concen-
trated at low energies compared to that of muons. Measurements of the energy
spectrum of electrons were obtained in the 1960s and 1970s by various authors
[Wibberenz1962; Allkofer1970]. Figure 10.7 shows a series of data reporting
the vertical intensity of electrons (positive and negative) obtained from various
measurements made at sea level [Grieder2001].
10.4 Gammas
The presence of gammas in extensive air showers is linked to a variety of physical

processes, including mainly the decay of neutral pions and other unstable particles,
the bremsstrahlung processes of electrons and positrons and their annihilation.
Although high-energy gamma rays have also been measured, beyond TeV, most of
the gamma radiation has a low energy, and the energy distribution follows a power law
of the E−γ type, with an exponent close to 1.8. The gamma intensity profile increases
with atmospheric depth, reaching—for ranges close to the vertical—a maximum at
atmospheric depths around 100–200 g/cm2 , and then decreasing as it goes towards
high atmospheric depths, close to sea level, as shown in Fig. 10.8, which reports
Thompson’s results obtained in the mid-1970s [Thompson1974].
Fig. 10.7 Vertical electron intensity, obtained from measurements made at sea level by Beedle and
Webber [Beedle1970] (crosses) and by Beuermann and Wibberenz [Beuermann1968] (triangles)
Fig. 10.8 Intensity profile of the gamma (with energy E > 30 MeV and inclination close to the
vertical), as a function of atmospheric depth. Data extracted from [Thompson1974]
10.4 Gammas 205
At sea level, the gamma spectrum has often been measured together with that of
electrons and positrons, without discriminating between the different components
of the soft radiation. Figure 10.9 shows an example of a gamma spectrum separated
from the electronic component [Beuermann1968], obtained at an atmospheric depth
of 760 g/cm2 , for energies from 100 MeV upwards.
At lower energies, from a few MeV to tens of MeV, gamma spectra measured at
different atmospheric depths have been reported by Ryan [Ryan1979], as shown in
Fig. 10.10, which reports some of these data.
Fig. 10.9 Gamma spectrum measured at atmospheric depth of 760 g/cm2 . Data extracted from
[Beuermann1968]
Fig. 10.10 Gamma spectrum at sea level and at atmospheric depth of 160 g/cm2 . Data extracted
from [Ryan1979]
Therefore, the measurable gammas at the sea level, even if abundant, have an
energy distribution mainly below 10 MeV, so that even a moderate shielding is able
to greatly reduce their contribution compared to that of muons.
10.5 Charged Hadrons
The flux of secondary charged hadrons, in particular protons and pions, and the corre-
sponding energy spectrum, has been measured in a variety of experiments carried
out at high altitudes, in the mountains, and with instrumentation aboard aircraft or
balloons. The first results were obtained by nuclear emulsions, while more complex
experiments, also with the aim of separating protons from pions—the two most abun-
dant components—performed in more recent years, used combinations of magnetic
spectrometers, calorimeters and time of flight measurements.
The dependence of the proton flux (of energy greater than 1 GeV) on atmospheric
depth, as emerges from a series of measures summarized by the work of Shopper
and collaborators [Schopper1967], is shown in Fig. 10.11.
As it can be seen, the flux decreases by several orders of magnitude going from
the top of the atmosphere towards sea level, where this component is practically
negligible compared to the penetrating component.
Although the spectrum in energy extends up to very high values (TeV), as
evidenced by recent experiments, most protons have relatively low energies, with
Fig. 10.11 Profile of the flux of vertical protons (energy E > 1 GeV) as a function of atmo-
spheric depth, from different experiments carried out between 1948 and 1955. Data extracted from
[Schopper1967]
10.6 Neutrons 207
Fig. 10.12 Energy spectrum of vertical protons, measured at an altitude of 3200 m. The different
symbols refer to different datasets. Data extracted from [Kocharian1956, Kocharian1959]
a maximum around a hundred MeV and a sharp decrease in flux going towards the
higher energies. Measurements of the flux of vertical protons, already carried out
around the end of the 1950s, had made it possible to extract the energy spectrum
of the protons, as for example represented in Fig. 10.12, which reports some data
from the measurements obtained by Kocharian and collaborators [Kocharian1956,
Kocharian1959] by the use of magnetic spectrometers combined with calorimeters.
Further sets of measures of the energy spectrum of protons, measured under various
conditions, are available in the Grieder review [Grieder2001].
At sea level the abundance of protons is very low compared to that of muons, of the
order of percent, especially at energies higher than GeV [Brooke1964]. Figure 10.13
shows the value of the proton to muon ratio in the momentum interval between about
1 and 80 GeV/c. Only for very low energies, less than about GeV, the abundance of
protons exceeds a few percent compared to that of muons.
From the same work by Brooke and Wolfendale [Brooke1964] it is also possible
to extract the spectrum of protons at sea level, which is shown in Fig. 10.14.
10.6 Neutrons
Neutrons represent an important component of the secondary cosmic radiation

present in the atmosphere and also at sea level. They can be generated as evapora-
tion products from the de-excitation of nuclei in the atmosphere, with low energies,
typically below 10 MeV, or as a result of high-energy nuclear collisions, in which
case they also have high energies. Low-energy neutrons, in particular, are affected by
solar events, both in terms of periodic variations due to solar cycles and catastrophic
events, such as solar flares, which we will discuss in a later chapter.
Fig. 10.13 Percentage abundance of protons compared to that of muons, evaluated at sea level, as
a function of the momentum of the particle. Data extracted from [Brooke1964]
Fig. 10.14 Intensity of protons at sea level, as a function of their momentum. Data extracted from
[Brooke1964]
The intensity profile of neutrons at different atmospheric depths, reported by

Armstrong and collaborators in the 1970s [Armstrong1973] is shown in Fig. 10.15,
between thermal energies and 20 MeV.
As it can be seen from the figure, the maximum intensity is reached around atmo-
spheric depths of about 100 g/cm2 , then decreasing by over two orders of magnitude
at sea level.
The energy spectrum of neutrons, measured at different altitudes and with different
techniques, shows that most neutrons have very low energies, from thermal neutrons
10.6 Neutrons 209
Fig. 10.15 Neutron intensity profile (from thermal energies to 20 MeV) at different atmospheric
depths [Armstrong1973]
(energies of the order of eV) up to energies of several GeV, with a shape that from
about 10 MeV can be represented with a power law, of the E−γ type. A detailed
measurement of the neutron spectrum in the upper atmosphere, at 12.5 km latitude,
in a wide energy range, from 0.01 eV up to GeV, was reported at the conclusion of
a series of aerial flights with instrumentation on board, by Hewitt and collaborators
[Hewitt1978], and is shown in Fig. 10.16.
In more recent times, studies of the flux of cosmic neutrons at various altitudes
and under different conditions have been carried out, even by the use of passive
detectors, mainly for the purpose of evaluating the possible effects on the absorbed
dose [Zanini2001].
Fig. 10.16 Energy spectrum of neutrons, measured at high altitude (12.5 km), in a wide range,
from about 0.01 eV up to about 1 GeV. Data extracted from [Hewitt1978]
Fig. 10.17 Abundance of light nuclei in measurements obtained during a series of Concorde flights
at high altitude (13–17 km) [Zhou1999]
10.7 Nuclei
Although in the primary radiation we have the presence of light and medium mass
nuclei, as seen in the previous chapter, the probability that such nuclei can propagate
intact in the atmosphere is enormously small, due to nuclear interactions. This is also
generally true for any secondary nuclei created in these interactions. The average free
path for nuclear interaction in air is in fact a few tens of g/cm2 (from 75 g/cm2 for
protons to 14 g/cm2 for heavier nuclei, such as Iron), so the flux, even for light nuclei
such as Helium, practically reduces to zero after about 100 g/cm2 of atmospheric
depth.
Measurements of the yield of light nuclei in the atmosphere were, for example,
carried out using instrumentation installed on board the Concorde during a series of
high-altitude flights (between 13 and 17 km) in the 1990s [Zhou1999]. Figure 10.17
shows the yield of light nuclei, (from Z = 2 to Z = 8) at these altitudes, obtained
from an observation period of several hundred hours of high-altitude flight.
As expected, based on the previous considerations about the probability of nuclear
interaction during their path in the atmosphere, the presence of nuclei, even the
light ones, at sea level, is practically negligible. Measurements aimed at estab-
lishing the presence of nuclei associated with cosmic radiation, or at establishing
an upper limit for their flux, have been carried out in recent decades. In Yock’s work
[Yock1986], for example, only some events due to nuclei heavier than protons (d,
t, 3 He), were observed, using a range telescope and a measurement time of about
5000 h, establishing an upper limit on the presence of 4 He.
Chapter 11
The Influence of the Earth
Abstract The atmosphere and the magnetic field of the Earth exert an important
influence on the flux of primary and secondary cosmic rays. These effects have been
observed since the early history of cosmic rays and were an important part of the
understanding the general properties of the cosmic radiation. Meteorological effects
due to the presence of the atmosphere, in particular the temperature effect and the
anticorrelation between the cosmic ray flux and the atmospheric pressure, resulting in
a barometric coefficient to be determined for any specific component are discussed,
with various quantitative examples. The influence of the Earth magnetic field and
the associated latitude and East–West effect are also presented. The latter had an
important role since the early investigations of the cosmic ray flux in the mid-1930s
and still gives an opportunity to carry out educational experiments in cosmic ray
physics.
11.1 Introduction
The Earth, with its atmosphere and its magnetic field, influences in various ways
the properties of the cosmic radiation, both primary and secondary, existing in the
vicinity of our planet. The Earth’s magnetic field, whose effects are important even
at large distances from the Earth’s surface, in fact strongly modify the propagation of
charged particles that are part of the primary cosmic radiation. These effects are also
important on the secondary particles produced in the atmosphere by the development
of extensive air showers. The first evidence of these effects, which lead to a different
flux of cosmic rays over different geomagnetic coordinates (geomagnetic latitude
and longitude of the observation point), dates back, as we have already seen, to the
first decades of the history of cosmic ray physics. Another important aspect through
which the Earth influences the propagation of the secondary cosmic radiation is
obviously the existence of the Earth’s atmosphere itself, with its composition and the
variations that characterize it, in terms of temperature and pressure. The presence of
the Earth’s atmosphere and its variations, together with the presence of the magnetic
field, result in an intensity which depends on the orientation (zenithal and azimuthal
angles) for the various secondary components. Variations in pressure and temperature
https://doi.org/10.1007/978-3-031-24762-0_11
212 11 The Influence of the Earth
also determine variations in the secondary flux observed locally, with the need to
make corrections to this, which then allow for further analyses of other physical
phenomena, for example those related to the Sun or to the properties of the primary
radiation.
We shall discuss some of these aspects throughout this chapter. On the other hand,
we have already discussed, both in relation to the properties of the primary radiation
and of the various secondary components, the effect of altitude with respect to sea
level on the measured intensity, which is largely determined by the presence of the
atmosphere and by the processes that take place in it. Measurements of the intensity
of the cosmic radiation in the Earth’s subsoil, under the ice or underwater, will be
discussed separately in a specific chapter.
11.2 The Interaction with the Atmosphere

and Meteorological Effects
Meteorological effects on cosmic rays are of interest both because their study makes
it possible to evaluate the corresponding corrections that must be introduced in
the experimental data in order to subsequently proceed with the study of extra-
atmospheric effects, and because the observed effects can contribute in themselves
to a better understanding of the changes taking place in the Earth’s atmosphere.
Basically, there are two effects on the variations in the observed flux of cosmic
rays, related to atmospheric causes: the barometric effect and the temperature effect.
A very complete treatment of these effects, in the context of the different sources of
variation of the cosmic ray flux, is reported in a comprehensive textbook by Dorman
[Dorman1974].
The atmospheric pressure is subject to temporal variations of various kinds, both
periodic and aperiodic. Periodic variations are those related, for instance, to the
day/night cycle (daily variations) and those on an annual basis.
The daily variations, basically linked to the atmospheric tidal phenomena and
the expansion of the air that accompanies the increase in temperature caused by the
solar radiation, have two maxima within 24 h, one around 10 o’clock in the morning,
more pronounced, and a second one around 10 p.m., and two minima, the first around
4 a.m., the second, more pronounced, around 4 p.m. In these considerations the times
are expressed in local time. The amplitude of these variations is greater during the
day than during the night. The difference between the high value in the morning and
the low value in the afternoon can therefore be taken as an estimate of the maximum
daily variation. The amplitude of these variations is of the order of at most a few
mbar, however it depends on the season and on the location of the observation site. It
is also known that planetary waves (Rossby waves), naturally occurring in rotating
fluids, may also be observed in the atmosphere of a planet, due to its rotation. These
waves are characterized by a low-pressure amplitude, and by a period in the order of
a few days.
11.2 The Interaction with the Atmosphere and Meteorological Effects 213
In addition to regular (periodic) variations, the atmospheric pressure is also

subject, as we know, to aperiodic variations, linked to the overall weather condi-
tions of that location. These variations can have a much greater amplitude, even
10–20 mbar, and may last even several days, so that their effect is able sometimes to
mask the regular variations.
By an example, Fig. 11.1 shows the trend of atmospheric pressure, expressed in
mm Hg, monitored with a pressure sensor connected to a data logger system during
a period of approximately 3 months, at intervals of 30, , except for some short breaks.
An expanded region of the previous graph is shown in Fig. 11.2, in a time interval
during which the overall weather conditions were sufficiently stable (average pressure
value only slightly increasing over a week). It is possible to see the daily variations,
with a period of 24 h, while the second daily maximum is hardly seen.
Fig. 11.1 Atmospheric pressure trend, monitored over a period of about 3 months, at regular
intervals of 30, . Both the daily variations, of a periodic nature, and the aperiodic ones, of greater
amplitude, linked to the overall weather conditions, are visible
Fig. 11.2 Expanded region of the previous graph, in a time interval of about a week, with the
presence of the daily maxima, and an average pressure trend only slightly increasing over time, due
to the overall weather conditions
The position of the maxima in the daily cycle is however strongly influenced
by the geographical location of the observation site, by its altitude and the overall
climate of that region (e.g., continental or desert climate). Depending on these specific
conditions, displacements of several hours in the time of the maximum and minimum
pressure have also been observed.
Quantitative analyses of these pressure data, measured as a function of time for
long periods, can be performed with methods used in general for the analysis of
time series. The use of correlograms, in particular, allows to accurately establish the
period of the main cycles and the existence of secondary periodicities. The existence
of a second daily maximum in the pressure trend has been, still in recent times, the
subject of discussion and detailed interpretations (resonance phenomena, absorption
of UV radiation by ozone, periodic wind trend) [Janeselli1968].
On an annual scale, further small variations in atmospheric pressure are observed,
linked to the cycle of the seasons and to the effects of heating and cooling of the soil.
They are smaller in amplitude, and strongly dependent on the particular geographical
area of observation.
The aperiodic variations in pressure are instead linked to the overall trend of
atmospheric weather, which in turn depends on numerous parameters, such as the
temperature of the air layers, humidity conditions and winds. The pressure condi-
tions are different from area to area, and determine, together with many other param-
eters, the weather conditions also in the following days. In this context, the maps
of isobars (lines that connect geographical points with the same pressure value),
frequently shown during weather forecasts, are of considerable importance as a tool
for understanding the evolution of atmospheric weather in the following days.
Catastrophic events, such as large volcano explosions, are also able to produce
shock waves in the Earth’s atmosphere, which can be observed even at large distance
from the explosion. Sometimes these waves, which may produce local pressure vari-
ations up to a few mbar, are able to go around the Earth even two or more times. A
recent example was offered by the explosion of the Hunga-Tonga volcano in 2022,
whose shock waves were observed in different parts of the Earth, and also monitored
by one of the EEE stations located in the Svalbard archipelago [Abbrescia2022].
Considering the effect of the pressure p at sea level, the variation dI in the intensity
of a specific component of the secondary cosmic radiation can be expressed as
d I = −μI dp
where μ is the absorption coefficient for that specific component. In the hypothesis
that this coefficient is constant, the previous equation admits a solution of the type
I = I0 e−μ(P−P0 )
where I0 is the intensity at pressure P0 . If the pressure variations (P − P0 ) are small

enough, the intensity variation can be expressed, in the approximation to the first
order, as a linear function of the pressure variation:
ΔI = β I Δp
where β represents a barometric coefficient, usually expressed in %/mbar. The sign

of this coefficient is negative, as the pressure variation is anti-correlated with the
intensity variation. In other words, the higher the pressure, the lower the intensity
of that specific component. Generally, for the muonic component the barometric
coefficient values are in the order of 0.1–0.2%/mbar and the linear approximation
is quite reasonable. For the neutron component, however, the observed variations
are much greater, with coefficients of the order of 0.7–0.9%/mbar, and therefore
using a linear approximation is not correct, requiring in this case to use the original
exponential dependence.
Having clarified this in a synthetic way, it must however be remembered that the
barometric coefficient, as it is linked to the absorption of a given component of the
secondary radiation in the atmosphere, depends on many factors, in particular on the
energy composition of that given component, which in turn is different depending
on the geographical position and altitude of the place of observation. It also depends
on the characteristics of the detector (for example its geometrical acceptance and
its efficiency at different energies). These considerations therefore suggest that each
detection apparatus is characterized by its own barometric coefficient, which should
be determined experimentally and subsequently used to correct the data measured
with that apparatus.
As an example, Fig. 11.3 shows the trend of atmospheric pressure (top) and the
corresponding counting rate (bottom) in a cosmic ray detector (telescope based on
MRPC chambers of the EEE Project) over a period of time of about 20 days. As it
can be seen, in correspondence with a strong decrease in the atmospheric pressure,
determined by the overall weather conditions, an increase in the event rate is observed.
Figure 11.4 shows the correlation plot between the two quantities, which shows
an almost linear trend.
To quantitatively obtain the value of the barometric coefficient, using in this case
a linear approximation, a linear fit of the pairs of values (pressure, event rate) can
be performed, obtaining the angular coefficient of the best-fit line. This coefficient
expresses the variation in the number of counts per second with respect to a unitary
variation in pressure (for example one millibar). To obtain the percentage change,
the angular coefficient determined by the fit can be divided by the average value of
the count rate in the data set considered.
The determination of the barometric coefficient for each particular experimental
apparatus presents however some critical aspects, as in the considered period all the
other effects that may lead to a variation in the observed flux must be regarded as
negligible. For this reason, the barometric coefficients are generally extracted from
limited data collection periods (for example a few weeks), possibly by averaging the
values obtained over different but compatible periods.
Depending on the specific secondary component (muons, electrons, neutrons),
and the energy range considered for these particles, the values of the barometric
coefficients can be quite different, as already mentioned.
Fig. 11.3 Atmospheric pressure trend (top) and the corresponding event rate measured in one of
the MRPC telescopes of the EEE project
For the muonic component, detailed measurements were made starting from the
1950s, with different types of detectors and different amounts of shielding mate-
rial placed above the detectors, observing values between 0.13 and 0.2%/mbar. As
expected, the muon flux observed in the subsoil, therefore with a considerable amount
of rock absorber material, has a much lower barometric coefficient than that observed
on the surface. Typical values deduced from measurements performed at equivalent
Fig. 11.4 Correlation plot between atmospheric pressure and event rate measured in an MRPC
telescope of the EEE project, over a period of about 20 days. The solid line in the plot shows the
result of a linear fit to the data
water depths of 40–60 m show barometric coefficients between 0.03 and 0.06%/mbar
[Dorman1974].
Reviews of recent data relating to the muonic component, measured by detec-
tors located in different locations around the world [Mendonca2019] have made it
possible to study the barometric coefficients and the temperature effect as a function
of the geomagnetic cutoff and the muon arrival angle. For example, it is observed that
the barometric coefficient for such detectors is lower in areas of low geomagnetic
cutoff, increasing towards the higher cutoff values.
For the soft component of the secondary cosmic radiation, generally higher
values are observed, of the order of 0.3%/mbar, although these values strongly
depend, as expected, on the amount of the absorber thickness placed above the
detector. Measurements carried out with an ionization chamber surmounted by
various amounts of material [Shamos1966] have shown for example values of
the barometric coefficient decreasing from 0.3%/mbar (in the absence of external
absorbers) down to values of 0.04%/mbar with absorbers larger than 2000 g/cm2 .
In the case of the neutron component, the barometric coefficient values are around
0.7–0.9%/mbar, also depending on the geomagnetic latitude.
If we consider instead the effect of atmospheric pressure on extensive air showers
as a whole, the dependence of the observed shower rate on atmospheric pressure
shows coefficients varying between 0.5 and 1%/mbar, depending on the energy of
the shower and the altitude of the observation site [Dorman1974]. Measurements of
the coincidence rate between individual detectors (therefore signalling the arrival of
simultaneous particles associated with a shower) were already made in the second half
Fig. 11.5 Dependence of the number of showers observed (ratio with respect to the average value
of the shower rate) on varying atmospheric pressure. Data reproduced from [Cosyns1940]. The
showers were identified in this context by the triple coincidence between detectors placed within a
distance of 18 m
of the 1930s, using for example three Geiger counters operated in coincidence and
observing how the number of these coincidences varied in an anti-correlated way with
the atmospheric pressure, with a significantly higher correlation coefficient than that
associated with the detection of individual particles in the shower [Stevenson1935,
Cosyns1940].
An example of the dependence of the number of showers observed on the atmo-
spheric pressure is shown in Fig. 11.5 [Cosyns1940], which shows the ratio between
the observed shower rate and its average value, for different pressure values. In this
context, to evaluate the number of showers, the coincidence between three detectors
positioned within a distance of 18 m was employed.
Using a set of detectors located at a certain distance, in other words a small array,
it is possible to determine how the barometric coefficient depends qualitatively on
the energy of the shower, selecting the events with a gradually increasing particle
density in each detector of the array. An example of such a measurement, carried out
in the 1960s by Jones [Jones1965] shows that as the density of particles increases,
the barometric coefficient also increases, from values of about 0.2%/mbar to values
of about 0.6%/mbar.
Even in the absence of variations in the atmospheric pressure at ground level,
variations in the local density of the air column above the observation site can also
determine an effect on the flux of secondary cosmic rays measured. This effect, due
to temperature, is the result of two causes, one due to interactions and one due to the
decays of secondary particles (pions and kaons) through the atmosphere, capable of
producing muons. An increase in temperature reduces the density, while at the same
time slightly varying the probability of interaction. On the other hand, the effects
due to decay increase, and the net result is an increase in the rate as the temperature
increases (positive correlation between temperature and observed rate). The other
effect (with negative correlation between temperature and observed rate) is linked
11.3 The Influence of the Earth Magnetic Field 219
to the phenomenon of the decay of the muons themselves. With the increase in
temperature and the consequent reduction in density, muons must travel longer paths
before decaying, thus producing a greater flux of muons observed at ground level.
Which of the two effects, that of positive correlation or that of negative correlation
with temperature, is preponderant? It turns out that everything depends on the energy
of the particles. In the case of high-energy particles, the positive correlation effect is
dominant, while the opposite happens for low-energy particles.
The variations related to the temperature effects of the atmosphere are however
different from those related to pressure and are approximately an order of magnitude
smaller.
Furthermore, while the variations related to pressure are interpreted in relation
to the atmospheric pressure measured on the ground, those related to temperature
require the knowledge of the temperature profile Ti along the different layers i = 1,
2, … N of the atmosphere itself.
Data of this kind, which are more difficult to obtain, generally require the use of
sounding balloons launched regularly. From these it is however possible to obtain
a single variable, an effective temperature Teff , averaged over the mass, in terms of
which to report the variations in intensity corresponding to a unit variation of this
temperature [Dorman1974, Riadigos2020]:
ΔI = α I ΔTe f f
where α represents a temperature coefficient, with values of the order of 0.2–0.3%/K.
11.3 The Influence of the Earth Magnetic Field
The Earth affects the flux of cosmic rays, even before the phenomena that occur in
the atmosphere, already at a large distance, through the magnetic field it generates.
Terrestrial magnetism, or geomagnetism, is one of the phenomena intimately linked
to cosmic radiation. We have already discussed how the hypotheses on the corpuscular
or radiative nature of the primary cosmic radiation could be verified precisely on the
basis of measurements performed at different geomagnetic latitudes. Even more
deeply, the flux of primary charged particles is strongly influenced by the Earth’s
magnetic field, which acts in some way as a screen for a large fraction of them, with
consequences of primary importance also on the very existence of life on Earth.
The Earth’s magnetic field extends from the inside of the Earth to very large
distances from the Earth’s surface and is generated by electric currents due to the
convective motions of the components present in the liquid state in the central part
of the planet, in turn caused by the flow of heat towards the outermost layers. The
intensity of the magnetic field at the Earth’s surface varies between about 30 and
60 μT (0.3–0.6 gauss), and as a first approximation it can be represented as the field
produced by a simple dipole inclined with respect to the Earth’s rotation axis. This
leads to a difference between the geographic coordinate system (latitude and longi-
tude), based on the rotation axis and the geographic equator, and the geomagnetic
coordinate system, as described in more detail in Appendix C. On a time scale of
the order of years, the position of the magnetic poles moves over the Earth’s surface
(with displacements of up to 40 km per year), while—but on a time scale of hundreds
of thousands of years—the whole Earth’s magnetic field reverses, as evidenced by
paleomagnetic studies on rocks. Paleomagnetism is precisely the study of the Earth’s
magnetic field in the past, which provides evidence of the movement of continents.
The effect of the Earth’s magnetic field extends up to distances of tens of thousands
of km with respect to the Earth’s surface (several times the Earth’s radius), where
the presence of the atmosphere is already negligible. Although of low intensity at
these large distances, the existence of a magnetic field extended over a very large
region makes the effect exerted on the motion of the charged particles in this region
important. In particular, this effect is noticeable on lower energy charged particles,
such as the so-called “solar wind”, a stream of particles released from the outermost
part of the Sun, the solar corona. Although during solar flares high-energy—even
GeV—particles may be emitted, for the most part these particles (mainly electrons,
protons and alpha particles, together with small percentages of heavier nuclei) have
very low energies—of the order of keV—but sufficient to escape the gravitational
pull of the Sun. We will discuss further aspects of cosmic rays due to the Sun in the
next chapter, while we are concerned here with understanding how the Earth, with
its magnetic field, can act as a screen against these particles.
The magnetosphere, or region in which the Earth’s magnetic field exerts its influ-
ence, has a complex structure. The ideal dipolar form that it would have is in fact
distorted by the presence of the solar wind, which itself shapes the outer solar field
in the interplanetary magnetic field. The magnetosphere is therefore asymmetrical,
with dimensions of the order of 10 times the Earth’s radius on the side of the Sun,
but with much larger dimensions, of the order of 100 times the Earth’s radius, on the
opposite side. Within the magnetosphere, a region—extending from 60 km above
the ground up to a few times the Earth’s radius—contains low energy ionized parti-
cles. In this region there is also the ionosphere and the Van Allen belts, discovered
following explorations with the first satellites in the late 1950s, which contain parti-
cles of higher energy, even of several MeV. The first of these bands is located at a
distance from the Earth equal to 1–2 terrestrial radii, the second at greater distances,
between 4 and 7 terrestrial radii. Figure 11.6 shows a sketch of the structure of the
Earth’s magnetosphere.
It must be remembered that many other objects in the solar system—but not all of
them—have a magnetosphere. Among these the planets Jupiter and Saturn, while for
example Venus and Mars do not have a magnetic field. The magnetic field of Jupiter
is greater than that of the Earth and enormously more extended, about 7 million km
on the side of the Sun and up to the orbit of Saturn on the opposite side.
The existence or absence of a magnetic field also implies a different geolog-
ical history for each planet, as well as the existence of certain conditions (such as,
for example, the presence of water or the atmosphere itself), which can make life
possible.
Fig. 11.6 Sketch of the Earth’s magnetosphere. Source Wikimedia Commons, Creative Commons
Attribution-Share Alike 4.0 International License
This is an important condition for the Earth, in which the existence of the Earth’s
magnetic field largely protects it from radiations. In addition to deflecting low energy
particles that make up the solar wind, the Earth’s magnetic field also acts on high-
energy particles, cosmic rays that come from outside the solar system. A previous
action on these cosmic rays is already exerted by the Sun with its magnetic field, as
we will see in the next chapter. As for the Earth, the trajectories of particles, even
of relatively high energy, and above all the possibility that they actually arrive in the
atmosphere, is strongly influenced by the geomagnetic field. In turn, the overall condi-
tions of the magnetosphere undergo changes due to solar conditions, particularly in
connection with catastrophic events, such as flares or coronal mass ejections (CMEs).
A recent example of a change in the conditions of the magnetosphere occurred in
June 2015, causing a sudden increase in the observed flux of muons [Mohanty2016],
which correlated with a large variation in the interplanetary magnetic field.
The overall effect of the Earth’s magnetic field, in terms of shielding with respect
to the primary cosmic radiation, in particular for high-energy particles of extrasolar
origin, is to deflect the particles below a certain threshold (geomagnetic cutoff) which
therefore cannot reach the Earth. Variations of this cutoff during a magnetic storm
can produce an increase in the observed flux of cosmic rays.
The geomagnetic cutoff can be calculated in principle from the detailed knowledge
of the magnetic field map. The general method consists in solving the equations of
motion of the particles, tracing the possible trajectories in the magnetosphere of each
particle of given energy or momentum, a complex problem from the point of view
of calculation.
In a simplified approach, the basic equations that regulate the motion of a charged
particle in a dipolar magnetic field allow us to understand the behaviour of a charged
particle approaching the Earth with its magnetosphere. Imagining the trajectory seen
by an observer located above the North Pole, if the particle has a very high energy, it
will not be appreciably deflected by the magnetic field and will be able to reach the top
of the Earth’s atmosphere. Particles with low energy, below a certain threshold value,
will be deflected backwards in interplanetary space, failing to penetrate the Earth’s
atmosphere. Particles of intermediate energy, higher than the threshold value, will be
curved more or less strongly by the magnetic field, penetrating the atmosphere along
a curvilinear path. However, what happens depends on the latitude of the place and
on the inclination of the arrival direction of the particle with respect to the magne-
tosphere. Around the poles, the magnetic field lines are radial in first approximation
and the particles arriving radially will have easier access to the atmosphere; in the
equatorial region the shielding effect of the magnetic field is greater and the threshold
value higher. Particles that have an energy just above the threshold can follow very
complicated trajectories.
The Norwegian mathematician and astrophysicist Carl Størmer was the first to
tackle this problem analytically, as early as at the beginning of 1900s, starting from
the observation of the northern lights. Størmer’s theory starts from the first calcula-
tions, carried out in 1907 and related to the motion of charged particles in a dipolar
magnetic field, in order to understand under which conditions electrons coming from
the Sun could reach the Earth, producing those phenomena known as Northern Lights
[Størmer1907]. Only in the 1930s did Størmer apply this method to the treatment
of the motion of particles of much higher energy, the cosmic rays of extrasolar
origin [Størmer1930]. Without going into the complex treatment of this theory, the
main consequence that follows is that the trajectories of the particles are confined in
permitted regions, while other (forbidden) regions are not compatible with trajecto-
ries that satisfy the equations of motion. Using Størmer’s theory it became possible to
understand the dependence of the observed flux of particles on latitude, as well as the
effects of East–West asymmetry. Subsequently, the possibility that charged particles
remained trapped in the Earth’s magnetic field was also studied on the basis of devel-
opments in this theory, a possibility that was confirmed in 1958 with the observation
of the Van Allen belts. Given the complexity of the calculations involved, even with
modern computing resources, numerical solutions of the equations are often used
to evaluate the geomagnetic cutoff at any given point, especially to estimate the
dose of radiation absorbed in space (for example in the International Space Station)
[Smart2005].
Figure 11.7 shows a sketch of some possible trajectories of charged particles of
different energy (magnetic rigidities) [Smart2005, Copeland2020]. Cosmic rays with
a high magnetic rigidity are less affected by the influence of the magnetic field and
follow relatively simple orbits (trajectory 1). For particles that have gradually lower
magnetic rigidity, the trajectories are curved more by the action of the field and are
more complex (trajectories with increasing numbers, 2, 3 … up to 15), even with the
presence of loops.
The distribution of the geomagnetic cutoff values can be obtained in any given
historical period by means of a numerical calculation of the possible trajectories based
Fig. 11.7 Possible trajectories of charged particles with different magnetic rigidity in the geomag-
netic field [Smart2005, Copeland2020]. The trajectories from “1” to “15” correspond respectively
to rigidities from the largest ones (more “resistant” to geomagnetic deflection) to the smallest ones
Fig. 11.8 Geomagnetic cutoff map [Johnson1938], with the level curves connecting the zones
with equal cutoff value. Figure reproduced by T. H. Johnson, Cosmic ray intensity and geomag-
netic effects, Review of Modern Physics 10(1938)194. Copyright (1938) by the American Physical
Society, License No. RNP/22/JUN/055239
on a given model of the geomagnetic field. Figure 11.8 shows a historical example,
dating back to the late 1930s of the geomagnetic cutoff map, with the level curves
that connect the areas of equal value to the cutoff. Recent data on the distribution of
vertical geomagnetic cutoff values are for example reported in [Gerontidou2021]. As
it can be seen, at high latitudes the geomagnetic cutoff is lower, which means that even
protons with relatively low energy can reach the top of the atmosphere, producing
extensive atmospheric showers. Going towards the equator, the geomagnetic cutoff
increases, and in the equatorial regions minimum energies up to about 15 GeV are
required for a vertical proton to penetrate the atmosphere. The contours of the regions
with the same cutoff are curved since the axis of the magnetic field does not coincide
with the axis of rotation of the Earth.
A very detailed report on geomagnetic cutoff calculations, with tables and graphs
for the period 1965–2010 is the one edited by Copeland [Copeland2020].
11.4 Angular Distribution of Muons and the East–West

Effect
Although we have already discussed some of these aspects as they were first observed,
it is worth mentioning explicitly in this context how some characteristics of the
secondary cosmic radiation are attributable to the influence of the Earth system.
In this section in particular we will examine the phenomenology concerning the
East–West effect.
While the angular distribution in the zenithal angle of the particles of the secondary
radiation is determined to a large extent by the influence of the Earth’s atmosphere,
based on the actual path that these particles take to reach the place of observation
and the processes (loss of energy, decays) that occur along this path, the dependence
on the azimuth angle in a given place is strongly linked to the charge of the particles
and to the effect of the magnetic field.
As far as the dependence on the azimuthal angle ϕ is concerned, anisotropic
effects between the flux of particles coming from the East compared to that of the
particles coming from the West, the so-called East–West effect, had been observed,
since 1930s by Johnson and Street [Johnson1933a, Johnson1933b] and by Rossi
[Rossi1934a]. In the work that reports the main results of the study by Johnson and
Street [Johnson1933b], reference is made to sets of measurements obtained with
Geiger counter telescopes, operated in such a way as to be able to observe, for the
same interval of zenithal angles, the number of associated events to cosmic rays
from the East and West directions. The measurements were carried out for different
zenithal angles, from 25 to 75°, with measurement periods of the order of ten hours
for each measurement, which made it possible to record a number of events of a few
thousand in each condition. The location of the experiment was in Mexico City, at
an altitude of 2250 m and a geomagnetic latitude of 29°.
11.4 Angular Distribution of Muons and the East–West Effect 225
Fig. 11.9 East–West asymmetry ratio observed by Johnson and Street in a series of measurements
obtained in the early 1930s at different zenithal angles and at an altitude of 2250 m. Data reproduced
from [Johnson1933b]
Figure 11.9 shows the results of these measurements, in particular the value of
the asymmetry ratio
IW − I E
R=2
IW + I E
where the quantities IW , IE represent the intensity relative to the two directions
of origin of the particles. This ratio therefore expresses the percentage difference
between the two intensities with respect to their average value.
As it can be seen from the previous figure, the percentage difference between the
intensities measured along the East and West directions amounts in these conditions
to about 10% for the intermediate zenithal angles, where it has a maximum around
40–50°, while the asymmetry decreases for inclinations close to the horizontal or to
the vertical.
In Rossi’s measurements [Rossi1934a], carried out in Africa, at a geomagnetic
latitude around 11° and at an altitude of 2370 m, it was found, for example, that the
W/E ratio between the flux of particles coming from the West (W) with respect to
those coming from East (E), was around 1.11 for θ = 15°, 1.16 for θ = 30°, and
about 1.19 at θ = 45°.
The measurements done in that period typically used Geiger counters and were not
always able to discriminate—based on the amount of shielding provided above the
detectors—between the soft and hard component of the secondary cosmic radiation.
As we discussed in the introductory chapters, such experiments were the basis for
considerations about the nature of the detected particles, as being charged, long
before the specific characteristics of these particles were clarified.
More detailed measurements, carried out in the following years [Johnson1934,
Johnson1935, Johnson1941], also showed that the amount of this effect depends
on the latitude and altitude of the observation site. The result of this campaign of
measurements, carried out using an apparatus with many Geiger counters arranged on
two concentric crowns, which could rotate around an axis, and a system of multiple
coincidences between the different counters, gave rise in those years to measurements
in different locations, at various altitudes (at sea level, between 2000 and 3000 m
and at the maximum altitude of 4300 m) and at different geomagnetic latitudes.
Since that period, East–West asymmetry measurements have been carried out by
many experiments capable of measuring the directions of arrival of cosmic particles,
in various locations around the world, at different geomagnetic altitudes and latitudes.
We cite as an example those carried out by Barber in 1949, at high altitude, using
airplanes [Barber1949], a review of the 1950s [Winckler1950] and many other works.
Measurements of the East–West effect today also represent a classic example
of possible educational activities in cosmic ray physics, which can also be carried
out with relatively simple equipment. An example of such activity is reported in
[Blanco2008c], which discusses some measurements made with a telescope of small
scintillators that can be oriented both in azimuth and zenithal angle.
11.5 The Latitude Effect
As we have already discussed with regard to the first worldwide measurement

campaigns conducted in the 1930s by Compton and Millikan, as well as by other
authors, the dependence of the cosmic ray flux on latitude, in particular on geomag-
netic latitude, was of great help for establishing the corpuscular nature of the radia-
tion. These campaigns in fact showed, as we saw in detail in Chap. 3, a dependence of
the observed flux on geomagnetic latitude, with an increase in intensity going from
the equator towards the poles, up to a certain latitude, around 50° N, beyond which
the intensity remained almost constant. The interpretation of these measurements, as
well as being based on the nature of the primary radiation as essentially constituted
by charged particles, was related to the characteristics of the Earth’s magnetic field,
capable of introducing a geomagnetic cutoff, i.e., a minimum energy that the primary
particles (vertical) must possess in order to penetrate into the Earth’s atmosphere and
therefore give rise to the creation of extensive atmospheric showers. This geomag-
netic cutoff is dependent on the particular geographic location (geomagnetic latitude
and longitude), according to some maps, examples of which we reported in Chap. 3.
Although further measurement campaigns—following those dating back to the
1930s—were organized sporadically in the following years, detailed and organized
measurements of the flux of cosmic rays in geographic locations spread out on a
large scale on the surface of the Earth have not apparently received the same atten-
tion over the following decades, with the exception of the use of neutron detection
stations, which are in any case located in fixed positions. In particular, it is difficult to
find examples of measurement campaigns carried out with the same detector trans-
ported over long distances, thus avoiding problems of comparison and normalization
between detectors having different characteristics.
11.6 Other Influences on the Cosmic Ray Flux Due to the Earth Environment 227
Only in very recent years have some measurement campaigns based on exploiting
the cosmic ray flux over different latitudes been organized by various Institutions,
quite often in the context of large-scale educational and scientific projects. An
example is represented by the measurement campaign carried out with a neutron
detector based on 3 He [Pyle1999]. Another example is the recent campaign of
measurements done on board an oceanographic vessel equipped with an MRPC
camera telescope, the TRISTAN detector [Saraiva2020], traveling between Spain
and Chile. A further recent example is the campaign of measurements undertaken
within the scope of the scientific and educational activities of the EEE project,
which used a scintillator telescope, the POLA detector, installed on board a boat
on a voyage between Iceland and the Svalbard Archipelago, up to a latitude of
82° N [Abbrescia2020]. The same telescope was also used for a further campaign of
measurements of the cosmic ray flux from latitude 35° N, thus covering a large part
of the entire range between 35 and 82° N [Abbrescia2023].
11.6 Other Influences on the Cosmic Ray Flux Due

to the Earth Environment
Due to the close connection between the observed flux of cosmic rays and the geomag-
netic field, it is reasonable to expect that variations in the Earth’s magnetic field will
produce variations in the measured flux of cosmic rays. Variations in the geomag-
netic field may be due to processes occurring inside the Earth, to processes in the
ionosphere or in the radiation belts around the Earth, or to processes in interplanetary
space near the Earth.
As already mentioned, long-term variations of the Earth’s magnetic field due to
internal causes have been highlighted by paleomagnetism studies, which have also
shown the polarity changes of the magnetic field on scales of hundreds of thou-
sands of years. It is plausible that in correspondence to these polarity inversions the
magnetic field undergoes considerable variations, leading to corresponding varia-
tions in the cosmic flux, for example to an increase in flux at the polarity inversion.
The characteristics of these polarity reversals are still the subject of experimental
studies and theoretical considerations, in particular as regards the duration of the
reversal periods, with some estimate between 1000 and 10,000 years, although there
are estimates of much shorter duration, about 100 years, compatible with the average
life span of people. It is known that the last reversal of the Earth’s magnetic field
occurred 780,000 years ago, with a duration of about 22,000 years. Inversions gener-
ally occur randomly, governed—according to some scientists—by Poisson processes,
by more complex statistical distributions or with chaotic characteristics. The average
interval between successive reversals of the polarity has however been estimated at
450,000 years, based on the reversals that have occurred in the last 83 million years.
As for the variations due to magnetic storms, these were highlighted in the late
1950s [Yoshida1959]. Magnetic storms are sudden variations in the magnetic field,
generally distributed throughout the planet, with a duration that can range from a
few hours to a few days.
Other aspects related to the structure of the magnetosphere are related to the “tail”
on the opposite side of the Sun, which also extends to very large distances, greater
than the radius of the Moon’s orbit. It has been shown that if the extension of this tail
is large enough, low energy particles can penetrate by diffusion inside it and reach
the polar caps, decreasing the value of the geomagnetic cutoff. The effect of this tail
was also evaluated for particles with slightly higher energy, for example protons with
energy around 1 MeV, to explain the (anomalous) fact that they can reach locations
at low latitudes.
It should also be remembered that to evaluate the variations in the measured flux
of cosmic rays due to extraterrestrial origin, it is necessary to take into account the
conditions under which the flux is measured, depending on the motion of the Earth
around the Sun, on the inclination of its rotation axis with respect to the ecliptic
plane and on the distance from the Sun. These aspects require the evaluation of
appropriate geometrical correction factors, which must be applied to the observed
flux, in addition to those due to the influence of the pressure and temperature of the
atmosphere, before investigating other causes of variation. A detailed treatment of
these aspects has been reported by Dorman [Dorman1974].
Chapter 12
The Secondary Cosmic Radiation
and the Influence of the Sun
Abstract The influence of the Sun, especially on the secondary components of

the cosmic ray flux, is discussed in this chapter. The Sun is subject to periodic
phenomena, among which the 11-years cycles and the 27-days periodicity related to
its rotation, and to aperiodic, catastrophic effects, such as the solar flares and coronal
mass ejections. Both have an effect on the observed flux of cosmic rays on Earth. The
modulation of the cosmic ray flux as determined by the solar cycle is presented, with
quantitative plots showing the number of sunspots and the corresponding cosmic ray
flux. An important aspect of the influence of solar events on the cosmic ray flux is
provided by the Forbush variations, discovered in the mid-1930s, sudden decreases of
the cosmic ray intensity followed by a slower recovery to its average value. Historical
data are provided about Forbush events, together with more recent data showing
examples of such effects on the neutron and muon flux. A brief mention is also given
on other solar effects, including ground level enhancements in the intensity of the
cosmic ray flux.
12.1 Introduction
The Sun exerts its influence on the surrounding environment through various phys-
ical processes, the study of which exceeds the topics most closely related to the
physics of cosmic rays. The region in which these phenomena are mainly localized
is called the heliosphere, and extends to a large distance from the Sun, exceeding 100
astronomical units, therefore well beyond the orbit of the outer planets. One of the
main characteristics is represented by the magnetic field present in the heliosphere,
by which cosmic rays of extra-solar origin are affected when passing through this
region. Depending on their point of arrival with respect to the heliosphere and on their
energy, the flux of particles of the cosmic radiation can undergo temporal variations,
thus encountering a modulation of solar origin.
Solar activity mainly affects particles with an energy lower than about
10 GeV/nucleon. This activity has periodic components, such as those represented
by the 11-year cycle of sunspots, and by the rotation of the Sun, with a period of
27 days. However, there are also aperiodic phenomena, such as solar flares and
https://doi.org/10.1007/978-3-031-24762-0_12
230 12 The Secondary Cosmic Radiation and the Influence of the Sun
coronal mass ejections that occur without a fixed periodicity, but are in some way
related to the number of sunspots, therefore to the eleven-year cycle of the Sun. These
local disturbances cause irregularities in the solar wind and associated magnetic
field. The Sun’s magnetic field reverses its polarity every 11 years, with effects on
the phenomena related to the transit of cosmic rays in the heliosphere. The geom-
etry and intensity of the magnetic field in the heliosphere have been modelled by
numerous authors (see for example [Gray1980]) based on four different components
(“dynamo”, “sunspot”, “dipole” and “ring”). The solar and terrestrial magnetic fields,
discussed in the previous chapter, together with the solar wind, constitute a complex
environment, subject to temporal variations, which also determine the flux of cosmic
rays observed on Earth.
12.2 Periodic Phenomena in the Sun and Solar Cycles
The solar cycle, or cycle of the magnetic activity of the Sun, is a cycle with a period
equal to 11 years, the main characteristic of which is highlighted in the number
of sunspots observed on the surface of the Sun. The phenomenon of sunspots was
observed quantitatively already at the beginning of 1600, with Galilei and Scheiner.
Starting from the second half of the 1700s, a periodicity in the average number of
observed spots was then noted. A more detailed study and a historical reconstruction
of the sunspots and observed periodicities were carried out by the Swiss astronomer
Wolf, around the mid-nineteenth century, who also introduced a quantitative index
for measuring the number of sunspots as well as the names of the various cycles,
starting from that relating to the period 1755–1766, conventionally called cycle 1.
The cycle in progress at the date of 2022 is cycle 25, which began at the end of 2019.
At the beginning of 1900 the deep link between the spots present in the Sun and
the phenomena of solar magnetism was recognized, establishing that within the 11-
year cycle the polarity remains unchanged, changing instead from one cycle to the
next, and returning to the value of departure after 22 years. Although the basic cycle
is considered to have a period of 11 years (and not the 22 years required by a full
magnetic cycle), the two halves of the 22-year period are not exactly equivalent in
terms of the number of spots.
Today the Sun’s activity is constantly monitored from many points of view,
with data available online for the various parameters that describe its activity
[SPACEWEATHER]. Figure 12.1 shows a recent image of the Sun, with some group
of sunspots in evidence.
Figure 12.2 shows instead the periodicity in the number of sunspots observed over
the last few centuries, from about 1750, the period from which the monthly averages
of the number of observed spots were recorded in a quantitative way. The inversion
of the solar magnetic field occurs in correspondence with the periods of maximum
number of sunspots.
The detailed trend for the last cycles is further shown in Fig. 12.3.
12.3 Modulation of the Cosmic Ray Flux Due to the Sun 231
Fig. 12.1 A recent image (February 28, 2022) of the Sun, showing several groups of spots
[SPACEWEATHER]. Also seen close to the borders of the image are photospheric faculae, bright
extended structures
Fig. 12.2 The number of sunspots observed over the last few centuries, since around 1750. Source
Wikimedia Commons
12.3 Modulation of the Cosmic Ray Flux Due to the Sun
When a greater number of spots are observed on the surface of the Sun, it exhibits
a phase of greater activity and emits more energy into the surrounding space. In the
minimum periods, even for intervals of the order of weeks, the presence of any spot
is not noticed; on the contrary, in peak periods, the simultaneous presence of several
Fig. 12.3 The number of sunspots observed over the last four cycles. Source [CLIMATE]
groups of spots can be observed. Although other phenomena, such as solar flares and
coronal mass ejections, are not exactly periodic phenomena, their average frequency
is somehow correlated with the 11-year cycle.
Without going into the detail, it can certainly be said that variations in the solar
activity affect many aspects of the conditions observed on Earth, such as the structure
of the geomagnetic field, the Northern Lights, and, in some ways, also the climate.
In this context we will mainly discuss the effect on the flux of cosmic rays.
When solar activity increases, so does the intensity of the solar wind, the flux
of particles that propagate in space and the associated magnetic field. This in turn
deflects the energetic particles arriving from outer space, resulting in a reduction in
the observed flux of cosmic rays to Earth.
Figure 12.4 shows the anti-correlated trend of solar activity and the flux of cosmic
rays, monitored through neutron monitoring stations, particularly sensitive to the
effects of solar variations. The observation period shown in this figure includes an
interval of about 50 years, in which, for five consecutive solar cycles, the anticorre-
lated trends of these two quantities, albeit slightly different from cycle to cycle, are
clearly visible.
12.4 Forbush Variations 233
Fig. 12.4 Number of observed sunspots and the corresponding cosmic ray flux, measured at the
Kiel neutron monitoring station. Source [CLIMATE]
12.4 Forbush Variations
The effect of solar phenomena on the observed flux of cosmic rays in the Earth
manifests itself not only in the periodic behaviour of sunspots, but also in aperiodic
events, during which abrupt changes in the intensity of cosmic rays can be observed.
One type of such events is the so-called Forbush variations, which were high-
lighted by the American physicist and geophysicist Scott Ellsworth Forbush (1904–
1984) in the mid-1930s [Forbush1937, Forbush1938]. Since that period, a few
hundred events of this type have been observed so far, in most cases by measurements
performed at ground, although in more recent times also by observations carried out
on board satellites (Fig. 12.5).
Forbush’s observations fall within the framework of continuous monitoring of the
intensity of secondary cosmic radiation. Already in the early 1930s, some stations for
detecting cosmic radiation had been equipped, in various locations, for the continued
observation of the intensity of this radiation, also on the proposal of Millikan and
Compton. The first examples were the survey stations built in Cheltenham (Maryland,
USA), Huancayo (Peru) and Christchurch (New Zealand), which were followed by
many others. It was precisely by observing and analysing the data obtained from these
stations, using a detection apparatus based on a special type of ionization chamber
made by Compton, Wollan and Bennett in 1934, that Forbush found evidence of a
significant variation in the intensity, visible in a correlated way by two distant stations,
Fig. 12.5 Scott Ellsworth

Forbush (1904–1984).
Source American Institute of
Physics, Emilio Segre Visual
Archive
the one located in Cheltenham and the one located in Huancayo [Forbush1937].
The (negative) variation observed in the two stations had occurred rapidly, within
a few hours, on April 25, 1937, with a slower recovery phase, over a few days, to
the previous intensity values, as shown in Fig. 12.6, taken from the original work
by Forbush. The data in the figure show, with two different symbols, the intensity
measured at the two stations. As clearly seen, the trend is very similar in the two
cases, and since the two stations are enormously distant, the effect could not be due
to a local cause, but evidently involved a cause at least at a planetary level. The other
important aspect of this observation is the correlation of this trend in the intensity
of cosmic radiation, measured at the two stations, with the values of the horizontal
component of the Earth’s magnetic field, also shown in the same figure. Both the
intensity and the magnetic field values are averaged in these results using intervals
equal to 2 h, intervals that allow at the same time to have sufficient statistical precision
and allow for the observation of any variations, even on a relatively short time scale.
The intensities measured by the two stations during this period, from 23 to 30 April
1937, showed—after correcting for the effect of atmospheric pressure—successive
variations, up to a maximum of about 4% of the intensity observed in the previous
days.
Already in this work Forbush identified what would become the salient features
of these variations: a sudden increase (SC, Sudden Commencement) for a period of
a few hours, followed by a rapid decrease and a phase of slow recovery of the initial
conditions, lasting a few days.
Subsequent observations, with data from additional survey stations [Forbush1938]
showed that the intensities observed in any two of the distant stations, after correcting
for local barometric effects, were in general highly correlated with each other over
time. In other words, the most significant variations in the intensity of the cosmic
radiation were phenomena of a planetary nature, and this was true not only for long
time-scale variations, such as seasonal variations, but also for sudden variations, such
Fig. 12.6 Trend of the intensity of cosmic radiation measured continuously at two distant survey
stations (Cheltenham in USA and Huancayo in Peru, data indicated by empty and solid symbols
respectively), compared with the trend of the horizontal component of the Earth’s magnetic field
(solid line). Figure reproduced by S. E. Forbush, On the Effects in Cosmic-Ray Intensity Observed
During the Recent Magnetic Storm, Physical Review 51(1937)1108. Copyright (1937) American
Physical Society, License RNP/22/JUL/055430
as those observed in correlation with large variations in the magnetic field. Similar
observations and considerations were also made in the same period by Hess and
collaborators [Hess1937, Hess1938].
At that time, considerations about the cause of these variations in the intensity of
the radiation were limited to the existence of a magnetic storm, the extent of which
was estimated by Forbush to be at least twice the Earth’s radius.
Forbush variations are now constantly monitored through the many neutron
component detection stations distributed worldwide.
Figure 12.7 shows the trend of the intensity measured by one of the neutron
monitoring stations, the Moscow Neutron Monitor Station, located near Moscow
[IZMIRAN], during the month of November 2004, during which a typical Forbush
decrease of high amplitude and duration is clearly observed. As an evidence that these
phenomena occur on a planetary scale, it is possible to see a comparison, Fig. 12.8
[LaRocca2005], between the data measured by this station and those reported by
another station, the Oulu Neutron Station, in Finland [OULU].
Forbush variations can also be observed for the muon component, as shown in
Fig. 12.9, which shows the temporal correlation observed between the data obtained
during the month of November 2004 from the Moscow Neutron Station and from a
muon detector operating at the time in Adelaide (Australia).
According to some authors, the exact definition of what is meant by Forbush
variation presents some ambiguity. Just to assume that these are a change in the
Fig. 12.7 Percentage change in cosmic ray flux, measured by the Moscow Neutron Monitor Station
[IZMIRAN], during November 2004
Fig. 12.8 Comparison between the intensity measured by two different neutron monitor stations,
located in Moscow and Oulu (Finland) for a few days, starting on November 9, 2004 [LaRocca2005]
flux of cosmic rays during a geomagnetic storm—as mentioned in some clas-

sical textbooks—or sudden changes (a few hours), at least 10%, in the intensity
of galactic cosmic rays, observed by neutron monitor stations, as defined by other
sites, overlooks notable aspects of these variations, as observed in more recent times.
Forbush variations are not necessarily sudden, as in some cases they have been
observed to occur gradually. The amount of these variations can reach 10%, but
these are very rare events, and in most cases the variation is only a few percent.
Forbush variations in cosmic flux have been observed not only through neutron
monitor stations, but also through muon-sensitive detectors, both at sea level and
underground or at high altitudes. Finally, the variations were observed not only for
the galactic component of cosmic rays, but also for that of solar origin.
Fig. 12.9 Comparison between the intensity measured by a neutron monitor station located in
Moscow and a muon detector operating in Adelaide (Australia), during the month of November
2004 [LaRocca2005]
Forbush variations are the net result on the intensity of cosmic rays of solar flares
or coronal mass ejections and/or of solar wind particle flow. Solar flares are intense
bursts of radiation connected to the release of energy stored into solar magnetic fields.
The emission of radiation from a solar flare involves all wavelengths of electromag-
netic radiation, even in the X-ray band, and therefore appears as a flow of energy that
propagates at the speed of light, and—if it occurs in the part of the Sun looking to
the Earth—it can reach our planet in about 8 min. Coronal mass ejections produce
the emission of material in the form of plasma, consisting mainly of electrons and
protons, by the solar corona, the outermost part of the solar atmosphere. A coronal
mass ejection can produce up to a billion tons of matter, emitted at speeds of millions
of km/h, which can reach our planet after a period of a few days. Although these are
two distinct types of events, solar flares and coronal mass ejections can also occur
together.
The topology of the Forbush variations is multiple: the time scale can be faster or
slower, the amplitude can vary within wide limits, the time profile can be simple, with
a single decrease followed by a recovery phase, or multiple, with several minima.
Fig. 12.10 Annual number

of Forbush decreases,
exceeding a certain
magnitude (3%), over the
years (data extracted from
[Belov2009])
These characteristics can also be different, depending on the place of observation on

Earth.
Most of the Forbush variations are sporadic and related to coronal mass ejections.
In these cases, the variation in the intensity of the cosmic rays is linked to a region of
the solar wind perturbed by strong magnetic fields. Often the decrease in intensity is
preceded by a slight increase. The actual Forbush decrease begins when the coronal
mass ejection impacts the Earth magnetosphere. Finally, there is a recovery phase
when the CME propagates beyond the orbit of the Earth.
Although Forbush events occur sporadically, there is a certain correlation between
the amplitude of Forbush decreases or the number of events greater than a certain
amplitude and the eleven-year cycle of the Sun, as shown in Fig. 12.10 [Belov2009].
In general, much statistical information is now available on Forbush variations
observed over the past 70 years by continuously operating neutron monitor stations.
Some of these stations make their data available through their respective websites
which can be addressed for further analysis. The possibility of correlating these events
with additional databases containing information related to solar parameters and
geomagnetic disturbances makes these analyses an important tool for understanding
the properties of these events.
An analysis of several Forbush events between 1957 and 2006, reported by
[Belov2009] showed for example that the distribution of the amplitudes A of these
variations, starting from amplitudes greater than (1–1.5)%, follows a power law
dN/dA ~ A−k , with k = 3.1. Variations with amplitudes greater than 3% correspond
to intense geomagnetic storms.
Finally, it should be noted that Forbush decreases have been observed in recent
times also in the space around the Earth, for example on the International Space
Station, through the instrumentation of the AMS Collaboration, on the planet Mars
[Forstner2018] and also at the borders of the Solar System, through the Voyager 1
and 2 probes.
12.5 Other Effects Related to Solar Activity 239
Fig. 12.11 Increased intensity observed at ground by different neutron monitoring stations, corre-
sponding to the GLE event of 23 February 1956. Figure reproduced from the GLE database of the
Oulu Cosmic Ray Station [OULU]
12.5 Other Effects Related to Solar Activity
Another phenomenon particularly linked to solar activity is represented by Ground

Level Enhancements (GLE), sudden increases in the intensity of cosmic radiation
measured at ground, in particular by neutron monitors. They are linked to the occur-
rence of solar flares, although the mechanism by which highly energetic particles of
the order of GeV can be produced on these occasions is not fully known. Although
the average number of these events is typically of the order of one per year, in some
periods of intense solar activity, several of these events have occurred. The increase
in the observed intensity ranges from a few percent up to the extremely high value
of 45 times the average intensity, in the case of the event observed on February 23,
1956 [Belov2005], as shown in Fig. 12.11, which reports the data obtained for this
event from different stations far from each other.
A currently accepted formal definition for a GLE-type event is the occurrence of
a statistically significant and simultaneous increase in at least two distant neutron
monitoring stations, of which at least one is located at altitudes close to sea level,
and a corresponding increase in proton flux measured by instrumentation in space
[Poluianov2017].
Chapter 13
Interaction of Muons with Matter
Abstract This chapter briefly recalls the most important aspects of the muon inter-
action in matter, since they have a crucial role in understanding various basic and
applied phenomena concerned with the physics of cosmic rays. The various mech-
anisms contributing to the energy loss of muons are considered, also providing the
reader with quantitative plots describing the specific energy loss and the muon range
in a solid material. Effects due to the multiple scattering of these particles in various
elements are also discussed in the chapter.
13.1 Introduction
The interaction of the particles that make up the cosmic radiation with matter is
an important aspect in the study of the properties of this radiation and its possible
applications. In general, the interaction of electrons, of “heavy” charged particles,
those having a mass significantly greater than that of electrons, as well as the interac-
tion of photons, are all aspects commonly and extensively treated in several specific
textbooks on the subject [Leo1987, Knoll2000, Leroy2004, Grupen2008], to which
we refer. In this chapter we only recall some brief considerations that specifically
concern the interaction of muons with matter. The soft component of the secondary
radiation, as we have discussed in previous chapters, can in fact be absorbed by a rela-
tively limited layer thickness of solid material, while the more penetrating charged
component, muons, is capable of going deep inside a solid material, even crossing
layers of considerable thickness. For this reason, the possible interactions of muons
with matter, in particular with solid rocks, constitute an important aspect of cosmic
ray physics, especially for the evaluation of shielding effects and for the application
aspects in which muons play a significant role that we will examine in a specific
chapter.
In propagating inside a material, muons are subject to different mechanisms: the
main ones concern the phenomena of ionization and atomic excitation, production
of e+ e− pairs, bremsstrahlung and nuclear interactions.
https://doi.org/10.1007/978-3-031-24762-0_13
242 13 Interaction of Muons with Matter
13.2 Energy Loss of Muons
The average energy loss per unit of path can generally be written in the case of muons
as the sum of two terms:
dE
= a(E) + b(E)E
dx
where the first term, a(E), represents the contribution of the ionization and atomic
excitation processes, usually described, for not too high energies, by the Bethe-Bloch
relation used for heavy charged particles:

dE Z 1 2m e c2 γ 2 β 2 δ
− = 4π N A re2 m e c2 z 2 ln − β 2
−
dx A β2 I 2
where NA is the Avogadro number, r the classical radius of the electron, m its mass
at rest, c the speed of light in vacuum, z the electric charge of the particle, Z/A the
ratio between the atomic number and the atomic mass of the medium√in which the
muon propagates, β = v/c (v being the velocity of the particle), γ = 1/ 1 − β2 is the
Lorentz factor and I (in eV) represents the ionization potential. The last term in the
previous formula represents a correction factor due to the density, often approximated
with 2 ln γ + k, with k constant. The traversed thickness x is expressed in terms of
the surface density (g/cm2 ). In energy units, the constant quantities in the previous
relation are equivalent to a factor
4π N A re2 m e c2 = 0.3071 MeV g/cm2
The ionization potential can be parameterized in different ways, either by the

choice of using a value common to all the elements, by a simple proportionality
between I and Z (or a power of Z with exponent close to 1), by semi-empirical
formulae which link I and Z or even by customized tables of values for each
element. Further details on the various terms are available in general textbooks on
the interaction of radiation with matter, already quoted, or in specialized reports
[Groom2001].
The second term, b(E), includes the contribution to energy loss resulting from
bremsstrahlung processes, pair production and nuclear interactions. Each of these
processes depends on the energy of the muons, and this dependence, complex from a
mathematical point of view, can be expressed in graphical form as in Fig. 13.1, taken
from the Particle Data Group [PDG]. To evaluate these values, the so-called “standard
rock” was considered, in which an average density of 2.65 g/cm3 is assumed, and
an average atomic number of 11, with Z/A = 0.5 and an average value of Z2 /A
equal to 5.5. It is also assumed that the average ionization energy and the parameters
describing the density effects correspond to those of calcium carbonate [Groom2001].
The overall energy loss of muons in standard rock, together with the contribution
due to ionization alone, is shown in Fig. 13.2. As can be seen, up to energies of the
13.2 Energy Loss of Muons 243
Fig. 13.1 Trend of individual contributions due to bremsstrahlung, production of pairs and nuclear
interactions in the factor b(E) which expresses the energy loss of muons in standard rock (Z/A =
0.5, density 2.65 g/cm3 ). Data extracted from the Particle Data Group [PDG]
order of 10 GeV, the mechanism of energy loss is essentially that due to ionization,
while above 100 GeV the contribution due to the other mechanisms begins to become
relevant. It can also be noted that the contribution due to ionization does not vary
much over a wide range of values of the incident energy.
Quantitative data on the interaction of muons in different materials, including
water, different chemical elements, and the most commonly used materials, can also
be extracted interactively from the Particle Data Group Web site [PDG] for practical
purposes or to evaluate the absorption of these particles.
Fig. 13.2 Total energy loss of muons in the standard rock (upper blue line), together with the
contribution due to ionization alone (lower red symbols), plotted as a function of the muon energy
13.3 Range of Muons in Matter
Of particular interest, also for the purposes of possible applications, is the evaluation
of the muon range in a given material, that is, the path that muons of a given energy
can travel within a material before being stopped. The mean range R(E) can be
expressed in g/cm2 and is related to the specific energy loss by the general relation:
0 −1
dE
R Eμ = dE
dx
Eμ
For very high energies it can be assumed as a first approximation that the terms a(E)
and b(E) in the relation dE/dx = a(E) + b(E)E are constant (not strongly dependent
on E); we can then write:
0
1 1 a + bE μ 1 Eμ
R Eμ = d E = ln = ln 1 +
a + bE b a b Ec
Eμ
where Ec = a/b represents the critical energy, that is, the energy at which the loss of
energy due to electronic processes (ionization and excitation) becomes equal to the
loss of energy due to radiative effects. From the previous equation it is also possible
to derive the relationship between the muon energy Eμ and its range:
a bR
Eμ = e −1
b
Figure 13.3 shows the expected range values, expressed in g/cm2 , corresponding
to a given muon energy, in the interval between a few hundred MeV and some TeV.
As can be seen from this figure, muons with energy 2–4 GeV, equal to the average
energy that these particles have at sea level, are able to cross a thickness between
1000 and 2000 g/cm2 of standard rock, equal to linear paths approximately between
3.5 and 7 m.
13.4 Multiple Scattering
Muons, like other charged particles, in addition to losing energy when crossing a
medium, can undergo scattering effects due to the Coulomb interaction. In the case
of a single process, diffusion can be described by Rutherford’s formula. This rela-
tionship had, as it is known, a decisive role in the understanding of the phenomenon
of diffusion of alpha particles by thin sheets of gold, leading to the formulation of
13.4 Multiple Scattering 245
Fig. 13.3 Expected range of

muons, in g/cm2 , in standard
rock, as a function of the
muon energy
atomic models based on the existence of a massive central core. Since the proba-
bility of having a deflection of an angle θ, in Rutherford’s formula, is proportional
to 1/sin4 (θ/2), it is expected that in most cases, the angular deflection will be very
small. The path of the charged particle in the medium will therefore follow a “jagged”
trajectory, with small changes of direction in each interaction. However, the possi-
bility that a particle undergoes a large number of individual scattering processes in
crossing a considerable thickness of material can lead to a non-negligible value of
the overall deflection (Fig. 13.4).
In the treatment of these processes a distinction is generally made between three
different possibilities, depending on the thickness crossed and therefore on the
average number of individual scattering processes expected.
(a) If the thickness is very small, and therefore the probability that the particle
undergoes more than a single scattering process along its path in the material
is negligible, the phenomenon can be described as due to a single Rutherford
deflection (single scattering).
Fig. 13.4 Schematic

description of the multiple
scattering phenomenon. A
charged particle of
momentum p can undergo, in
crossing a certain thickness
X of matter, a series of small
deflections that lead to a
finite deviation from the
initial direction of motion
(b) If the thickness is very large, so much that the number of single scattering
processes is very high, greater than a few tens, and the overall energy loss
is still small, the phenomenology can be treated with statistical methods, to
obtain, according to various approximations, the probability of having a given
overall deflection. In this case we speak of multiple scattering, and it is the most
commonly treated case, even in practical applications.
(c) If the thickness has an intermediate value, such that the number of individual
scattering processes is of the order of ten, we speak of plural scattering. This is a
difficult situation to deal with, since the process cannot be described neither by a
single interaction nor by the combined effect of many interactions. A treatments
of this kind of situation has been reported for example by Keil and collaborators
[Keil1960].
However, even in the case of multiple scattering it is possible to find different
approximations to the general solution, which is very complex. Among the most
used approximations are those provided by Molière [Bethe1953] and by Snyder and
Scott [Snyder1949], equivalent to each other and valid for values of the deflection
angle that are not too high.
For small scattering angles a Gaussian approximation to the probability distribu-
tion can be used, with a zero mean value and a standard deviation (in the plane) θ0
given by
/
13.6MeV X X
θ0 = 1 + 0.0038 ln
βcp X0 X0
where βc is the speed of the particle, p its momentum, X is the thickness crossed and
X0 the radiation length of the material. The latter is a characteristic quantity of the
material, also linked to the development of electromagnetic showers, and which can
be expressed in terms of the chemical properties (Z, A) of the material according to
various parameterizations, for example [Grupen2008]:
A
X0 =
4α N A Z 2 re2 ln 183Z − 3
1
The standard deviations of the scattering angle in the plane and in space are linked
by the relationship:
1
θ0 = √ θspace
2
As it can be seen from the previous relations, the scattering angle depends on
the momentum of the particle, being smaller for higher momenta, and it depends a
lot on the atomic number of the material. What has been said so far also applies to
cosmic muons and electrons, as charged particles of high energy. The phenomenon
13.4 Multiple Scattering 247
Fig. 13.5 Standard

deviation in the scattering
angle of muons of various
momentum, in the case of
crossing a standard 10 cm
thick solid rock
Table 13.1 Standard deviation of the multiple scattering angle distribution, for muons of 3 GeV/c
traversing a linear thickness of 10 cm in various materials
Element Air Rock Aluminum Iron Lead Uranium
Std. Dev. (°) 0.005 0.26 0.28 0.62 1.11 1.48
of multiple scattering must be appropriately taken into account in various aspects

concerning the physics of cosmic rays and its applications. The tracking accuracy of
charged particles, including muons and electrons present in the secondary radiation, is
sometimes limited just by the amount of multiple scattering that the particles undergo
in passing through any material placed above the detection apparatus (absorbers, roof
or floors of the building, …). Particularly underground experiments can be influenced
by this aspect, although in general the particles of lower energy are absorbed and for
those of high energy the effect of multiple scattering is reduced.
Figure 13.5 shows as a practical example the value of the standard deviation θ0
as a function of the muon momentum, assuming a standard rock thickness equal to
10 cm.
For different materials, in particular those with a high atomic number, the scat-
tering angle can be significantly higher, as can be seen from Table 13.1, which reports
the values of the standard deviation (θ0 ) of the distribution of the scattering angles
for various materials, with the same thickness of 10 cm. This constitutes the basis, as
we shall see, also for possible applications in the field of muon tomography, which
will be discussed in Chap. 16.
Chapter 14
Cosmic Radiations Underground, Under
Water and Under the Ice
Abstract This chapter is concerned with measurements of the cosmic ray intensity
underground, under water and under the ice. Both historical experiments providing
the first measurements of this type, and more recent results obtained to large depths
are presented and discussed, with quantitative plots showing the absolute value of
the muon flux under a large thickness of material, together with different parameter-
izations which summarize the results. Possible parameterizations of the muon flux
underground are further discussed in Appendix M.
14.1 Introduction
The particles of the secondary cosmic radiation that manage to reach the ground
can further interact with the soil (or with water), according to the same mechanisms
with which they interacted with the air of the atmosphere. In a solid or liquid mate-
rial, however, which has a density much greater than that of air, unstable particles
are subject to a higher probability of interaction than the probability of decaying,
contrary to what happens in air. As for the type of particles, while electrons and
hadrons are easily absorbed because of these interaction processes, the most pene-
trating components—muons and neutrinos—can also reach great depths in the under-
ground rock, in the sea or in the ice. The interaction of muons with materials can also
produce additional secondary particles of various kinds. We will not discuss in this
context neutrinos, which, due to their interaction cross section, have a very low prob-
ability of interacting with even a considerable amount of solid material, most of them
managing to cross the entire Earth. The detection of neutrinos is today of consider-
able importance in astroparticle physics, but it requires enormous detection volumes
to compensate for the very low probability of interaction and to collect, despite this,
a significant number of events. On the contrary, muons, as charged particles, easily
interact with rock, sea water or ice, according to the mechanisms described in the
previous chapter, and their flux progressively attenuates in the subsoil, making their
detection possible basically with the same equipment and the same techniques used
on the surface.
https://doi.org/10.1007/978-3-031-24762-0_14
250 14 Cosmic Radiations Underground, Under Water and Under the Ice
Measurements of the muon flux at different depths in the Earth’s subsoil, or

underwater, have been carried out in various conditions, up to great depths, in excess
of 106 g/cm2 (10 km.w.e.), depths at which the atmospheric muon flux becomes
comparable with the background value of muons created by neutrinos. Recall that
the entire thickness of the atmosphere corresponds to about 1000 g/cm2 ; therefore,
the depths involved in these measurements are up to 1000 times greater. In the case of
measurements carried out under water, the composition of the liquid is quite known,
apart from small variations in density or the presence of dissolved substances, while in
the case of solid materials such as rocks, consistent variations in the composition and
density of the rocks, as muons penetrate deeply, make data analysis more complex.
The measurement of the intensity of muons underground has numerous aspects
of interest, due to the possibility of studying in detail the interaction mechanisms
of high-energy particles, as well as from an application point of view, either in the
design of underground laboratories with a low cosmic background noise, such as
many of the astroparticle physics experiments require, or in those applications that
use cosmic muons for the tomography of large natural structures, such as volcanoes.
14.2 Underground Measurements
The first measurements of the intensity of the cosmic radiation in the subsoil, or
under a certain depth of water, date back to the history of the discovery of cosmic
radiation, as we discussed at the beginning of this text. In that case, it was a question
of establishing the very nature of the radiation, whether it came from the outside or
from the radioactivity of terrestrial rocks. We now turn instead to the more systematic
measurements, carried out in the following years, once the nature of these radiations
had been clarified, and in particular of the measurements concerned with the most
penetrating component, the muons. A review of the main results and interpretations
of these first measurements, carried out up to the 1950s, can be found, for example,
in a work of 1952 by Barrett and collaborators [Barrett1952], which in addition to
reporting their original measurements, carried out in a mine located at the depth of
1574 m.w.e. (1.574 × 105 g/cm2 ), reviewed experiments performed up to that time
by other authors.
At that time it was already well established—from measurements carried out in
previous years at various depths in the subsoil—that the penetrating particles, capable
of propagating and being detected even under a consistent amount of rock, were
muons; however, it was precisely the underground measurements that were seen as a
precious investigative tool to test the overall representation of extensive air showers,
with their fraction of penetrating particles, as well as the mechanisms of interaction
of these particles, including the secondary reactions to which muons could give rise.
At the end of the 1930s, different types of measurements were already available,
organized inside mines, with detectors of various kinds (including Geiger counters
operating in coincidence). Examples of such experiments are reported, among others,
14.2 Underground Measurements 251
Fig. 14.1 Vertical intensity of muons, obtained from various experiments, each with a different
symbol. The values indicated with the symbols in red, corresponding for the most part to greater
depths crossed, indicate the measurements extrapolated from inclined trajectories of the muons.
Data extracted from [Grieder2001]
by Barnóthy and Forró [Barnóthy1939], by Clay [Clay1939], Wilson and Highes

[Wilson1943], Nishina [Nishina1941] and Miyazaki [Miyazaki1949].
A review of the vertical intensity measurements of muons at various depths,
as reported by different experiments (each shown with a different symbol)
[Grieder2001] is shown in Fig. 14.1. The corresponding measurements at higher
depths, indicated with red triangles in the figure, were obtained from experiments in
which muons with inclined trajectories were also detected, evaluating the thickness
of the rock actually crossed.
As can be seen from these results, the flux of vertical muons in the subsoil
decreases by a factor of 10 after a depth of approximately 250 hg/cm2 , equivalent to
25 m.w.e. In the measurements of Barrett et al. [Barrett1952], the flux measured at
a depth of 1574 m.w.e. was approximately a factor 1000 smaller than that measured
at the surface, as can also be seen from the overall set of results in Fig. 14.1. Going
further in depth, as it has been done in more recent times, in order to evaluate the flux
of muons in sites dedicated to the installation of astroparticle physics experiments,
which require environments with a much reduced background of cosmic muons, they
have shown a further decrease of 3 orders of magnitude from a depth of 1500 m.w.e.
to a depth of about 8500 m.w.e., like that corresponding to the LVD experiment at the
Gran Sasso underground laboratories. We therefore understand the growing difficulty
in measuring the muon flux at ever greater depths, in terms of the measurement times
required or the geometric dimensions of the detectors, in order to have a significant
number of events. For a review of recent results on underground muon flux, see the
section provided by the Particle Data Group [PDG] on this topic.
The relationship between cosmic muon flux and depth in the subsoil—expressed
by the results shown in Fig. 14.1—has been parameterized according to various
semi-empirical formulas. A frequently used relation has been for example proposed
by Miyake [Miyake1963], which expresses the vertical intensity I(X) as a function
of the depth X as
K
I (X ) = (X + a)−α e−β X
X+H
where the parameters a, K, H, α and β are obtained from a fit of the experimental
data and the depth X is expressed in m.w.e. (or hg/cm2 ). Appendix M shows the
intensity as a function of depth X, with a series of numerical parameters used to
reproduce the first measurements available at the time, together with a simplified
formula that was sometimes used instead of the previous equation. At very large
depths, the behaviour of intensity as a function of depth can be represented by a
simple exponential relationship, such as
I (X ) = Ae−X/Λ
where the two parameters A and Λ, obtained from the data, vary according to
the range of depths explored. As an example, measurements carried out by Meyer
[Meyer1970] between 4000 and 9000 hg/cm2 , can be represented by the previous
relation, assuming A = 1.04 × 10–6 and Λ = 804 hg/cm2 .
A more detailed list of the intensity measurements of muons, carried out under-
ground since the 1950s by various experiments, is reported by Grieder [Grieder2001].
It should be noted that some experiments directly measure the flux of vertical muons,
while others, also sensitive to muons with inclined trajectories, make corrections to
the data in order to obtain the vertical flux. The detection apparatus in these measure-
ments used sets of Geiger counters operating in coincidence, photographic emulsions
or telescopes of plastic scintillators.
The results of the first underground measurements had allowed, among other
things, to extract not only the intensity of the vertical component, but also the angular
distribution of muons in the subsoil, a trend of the cosn θ type, with an exponent n
whose value was however different at various depths. In Barrett’s work [Barrett1952],
for example, with measurements made at 1574 m.w.e. the shape of the angular distri-
bution of muons, shown in Fig. 14.2, could be represented using a value of n equal
to 3.
Neglecting the Earth’s curvature, at least for inclination angles not too far from the
vertical, a vertical depth X corresponds, for muons arriving with a zenith angle θ, to
an effective depth (slant depth) equal to XS = X sec(θ). If we assume an exponential
attenuation law for the muon flux, they will be attenuated by a factor e−X sec θ/Λ as
they penetrate to a vertical depth X with an inclination angle θ. For greater zenith
angles, the approximation of a “flat” Earth surface is no longer valid, and the effect
of the Earth’s curvature on the distances actually crossed must be considered.
Even in more recent times, the zenithal angular distribution of underground muons
has been represented with a dependence of the cosn θ type, with an exponent n varying
with the depth, roughly like [Miyake1973]:
n = 1.53 + 8 × 10−4 X
14.2 Underground Measurements 253
Fig. 14.2 Angular distribution of muons measured at a depth of 1574 m.w.e. Data extracted from
[Barrett1952]. The solid line shows the shape of the cosn θ function that fits the data, with a value
of the coefficient n = 3
where X is the depth expressed in hg/cm2 . At large depths the exponent n increases
significantly, and this makes the zenithal angular distribution more pronounced in
correspondence with the vertical, with a gradually more negligible contribution from
the inclined muons. In Miyake’s work [Miyake1973], more sophisticated parame-
terizations are also proposed to describe the zenithal angular distribution at large
depths. The distribution of the azimuth angle, on the other hand, is uniform, since
the East–West effects perceived on the surface are mainly due to the lower energy
muons, sensitive to the magnetic field, while at large depths only highly energetic
muons are able to penetrate, on which the geomagnetic effect is negligible (Fig. 14.3).
Fig. 14.3 Values of the exponent n in the relationship that expresses the intensity of underground
muons as cosn θ, obtained from experiments performed at various depths in the subsoil. Values
extracted from [Grieder2001]
Fig. 14.4 Differential intensity of muons, in an underground experiment carried out at a depth of
70 hg/cm2 of rock by means of a magnetic spectrometer [Murdoch1960]
The first attempts to measure the energy distribution of muons at a certain depth in
the subsoil date back to the 1950s and 1960s, with measurements made with nuclear
emulsions [George1956], cloud chambers [Nash1956] and magnetic spectrometers
[Daion1959, Ashton1960, Murdoch1960], at a depth of a few tens of m.w.e. below
ground level. Measurements at greater depths (850 hg/cm2 ) were carried out in the
1970s, with emulsion chambers, up to energies of several TeV [Mizutani1979]. The
result of the measurements carried out at the end of the 1950s had already made it
possible to obtain the spectrum of the differential intensity of muons up to energies
of the order of 100 GeV (Fig. 14.4).
From the analysis of these first measurements, a power law was suggested for the
relationship between flux and energy, of the type E−w , with w = 1.8 for energies
greater than 1010 eV.
By increasing the depth, the amount of the rock above acts as a filter, gradually
selecting muons of increasingly higher average energy, as shown in Fig. 14.5, which
summarizes data obtained in the course of various experiments.
Carrying out measurements at a certain depth therefore means selecting muons
with a given average energy, depending on the depth. Assuming an approximately
constant rate of muon energy loss, equal to 200 MeV per meter of water equivalent
(m.w.e.), the average muon energy at a depth h (expressed in m.w.e.) is equal to
200 h/(w − 1) in MeV. For example, at a depth h equal to 1000 m.w.e., the average
energy would be 2.5 × 1011 eV, or 250 GeV.
The main characteristics of the energy and angular distributions of muons
measured in depth have already been extracted from the experiments performed
several decades ago. In more recent times, starting from the late 1970s, several
experiments in astroparticle physics, dedicated to neutrino physics, looking for
14.3 Measurements Underwater and Under the Ice 255
Fig. 14.5 Average energy of vertical muons measured at a certain depth in the subsoil, as a function
of depth, expressed in equivalent km of water (km w.e.). Data reproduced by [Alexeev1973] (red
triangles), by [Castagnoli1997] (blue squares), and by [Ambrosio1999] (green dots)
possible proton decays or for dark matter, have installed large underground detectors,
with good detection capabilities for muons, which allowed to obtain more detailed
information on the distribution of muons at large depths.
14.3 Measurements Underwater and Under the Ice
Apart from the first measurements of historical interest carried out at the begin-
ning of the history of cosmic rays, such as the measurements made by Pacini, and
already mentioned in Chap. 1, measurements at great depths underwater in the marine
environment were carried out starting from the 1960s by Higashi and collaborators
[Higashi1966] up to a depth of 1380 m, and subsequently, by Davitaev and collab-
orators [Davitaev1970] up to 3190 m, by Vavilov and collaborators [Vavilov1970]
up to 3190 m, by Fyodorov and collaborators [Fyodorov1986] up to about 5000 m,
from Babson et al. [Babson1990] up to 4150 m. Figure 14.6 shows the set of these
results, which show a well-defined behaviour, similar to that obtained under a certain
thickness of rock and already shown in Fig. 14.1.
Measurements in a lake environment (pure water instead of salt water, as in the
seas) were also carried out in subsequent years, obviously at shallower depths. For
example, Rogers and Tristam [Rogers1984] used, up to a depth of 175 m, in the
Geneva Lake, a cylindrical scintillator telescope read by photomultipliers, completely
autonomous (from the point of view of power supply, control and data saving), placed
inside a suitable sealed container. Their results are shown in Fig. 14.7. Measurements
in Lake Baikal, at larger depths (1170 m) were carried out in the 1990s.
Fig. 14.6 Vertical intensity

of muons, obtained from
various experiments
organized underwater in the
sea, each shown with a
different symbol. Data
extracted from several
experiments and summarized
in Grieder’s review work
[Grieder2001]
Fig. 14.7 Vertical intensity

of muons, obtained from
measurements made by
Rogers and Tristam
[Rogers1984] in the Geneva
Lake, up to a depth of 175 m
In even more recent years, large-scale experiments for neutrino physics have been
set up or are being set up underwater or under the ice, providing new experimental data
regarding muons at very large depths, particularly for the high-energy component.
Measurements carried out at large depths under water or under the ice have the
advantage of being able to know more precisely the thickness of the material traversed
than those performed under rock thicknesses whose composition is never known in
detail along the entire path of the muons. Such measurements can therefore be used
for more quantitative tests of particle propagation and energy loss.
As regards the flux of muons under a given thickness of ice, the first measurements
under the ice were carried out in Greenland, at a depth of about 200 m, by the
AMANDA Collaboration [Lowder1991]. Recent measurements, at depths of up to
about 2 km under the ice of Antarctica, have been carried out very recently as part
of the activities of the IceCube Collaboration.
Chapter 15
The Origin of Cosmic Rays
Abstract The question of the origin of the cosmic radiation has been raised since
the early evidence of the existence of this radiation and it is still an important aspect
addressed by several recent investigations. A few historical considerations about
the acceleration mechanisms and the origin of cosmic rays are provided along the
chapter, with a brief discussion of the Fermi acceleration mechanism. The role of the
supernovae in the understanding high energy cosmic rays is also presented, together
with a brief mention concerned with the extra-galactic high energy components.
15.1 Introduction
Although the question of the origin of cosmic rays has been raised since the earliest
evidence of the existence of an extraterrestrial radiation, after more than a century the
problem still has a number of open aspects, despite significant progress has been made
in its understanding. Although these pages are not expressly dedicated to presenting
the astroparticle issues inherent in the question of the origin of cosmic radiation,
we cannot avoid mentioning at least some historical steps, the general problem and
some recent results. General considerations of historical interest about the origin of
cosmic rays can be found not only in the original papers by various authors, but also
in textbooks dating back to the ‘60s [Ginzburg1964, Sandström1965].
Meanwhile, we mention that the energy spectrum of the primary cosmic radiation,
which we discussed in one of the previous chapters, does not show evident discon-
tinuities, except for a slight change in the slope, along an enormous energy interval,
which covers from a few MeV to over 1020 eV. This, however, does not mean that
the particles of such different energies all have the same origin, or are accelerated by
the same mechanisms. Already a few decades after the first evidence of the extrater-
restrial origin of this radiation it was considered reasonable that the processes that
gave rise to the primary particles took place in different environments of space, from
regions close to our solar system to regions outside the Milky Way. The consider-
ations about the origin of the primary cosmic radiation generally refer, in fact, to a
“local” origin (in which it is assumed that this is produced—if not within the Solar
System—in a region of negligible dimensions compared to those of our Galaxy), to
https://doi.org/10.1007/978-3-031-24762-0_15
258 15 The Origin of Cosmic Rays
a “galactic” origin (to denote those components that are produced and accelerated in
regions comparable to the size of the Milky Way) and finally to an “extragalactic”
origin (to indicate the components originating from outside our Galaxy).
The phenomena of the production of particles and their subsequent acceleration
cannot be completely unrelated, and the question of where the particles in ques-
tion originate is also dependent on the possible mechanism hypothesized for their
acceleration. The evidence that in a first approximation the primary cosmic radia-
tion is isotropic, that is the flux is independent of the incoming orientation, would
lead to hypothesize either a large number of “local” sources uniformly distributed
throughout the space, or a “storage” mechanism of very long duration, through which
the particles reach a sort of equilibrium that causes them to come from all directions.
As was recognized many decades ago, discussions about the origin of cosmic radi-
ation include at least four different types of problems: the production (injection),
acceleration, localization and storage of particles.
In general, observational evidence should be used to answer such questions.
However, there has not been much evidence from the past capable of directly telling
us something about the origin of the radiation and providing an answer to the type
of problems listed. Most of the characteristics of the primary cosmic radiation, for
many decades, have been obtained from measurements performed on Earth or in the
Earth’s atmosphere. Only since the 1960s has a considerable amount of data been
obtained, also through observations by satellites, first in orbit around the Earth, and,
subsequently, in interplanetary space. The probes which are at the largest distance
from Earth, such as Voyager 1, have only recently passed what can be qualitatively
defined the boundaries of the Solar System, giving in principle experimental infor-
mation on what is happening in the interstellar space. It should also be remembered
that in recent decades it has been possible to extend the observation of the universe
beyond the visible band, both in infrared and radio waves and in the X and gamma ray
regions. This has enormously increased the amount of observational data, some of
which can also give valuable information about the origin and distribution of cosmic
radiation.
If we were to say which experimental evidence of the past could give direct
information on some of the problems listed above, they mainly concern the variations
related to solar phenomena, as discussed in one of the previous chapters. In this case,
for example for solar flares, the question of origin is well defined, but it concerns
only a small component of the energy spectrum of primary radiation, usually the
low energy region. However, it is not known exactly what the upper limit of the
component of cosmic radiation associated with the Sun may be, because there is also
evidence of protons with an energy equal to a few tens of GeV associated with the
Sun or with interplanetary space in the Solar System, without a clear identification
of acceleration mechanisms.
The energy spectrum of this radiation shows, as we have discussed, a variation in
the exponent of the power law depending on the specific energy region. However, it is
not easy to separate this spectrum into separate individual components, corresponding
to different origins and acceleration mechanisms. In any case, it is reasonable to
assume that all regions corresponding to different origins or mechanisms would
15.1 Introduction 259
have some overlap between them, smoothing the transition areas between one region
and another.
As regards the composition in charge of the primary nuclei, it has been established
that it presents differences with respect to the distribution of the elements in the
universe. However, even in this case, the knowledge of both distributions is not as
detailed, and the interpretation of these differences is not immediate. For example,
the presence of light elements in the cosmic radiation is lower than that of the natural
abundance of elements in the universe, and this suggests that the cosmic radiation is
relatively “young”, that is, it has not had sufficient time to go through a fragmentation
of the heavier nuclei through interstellar space, having suffered a relatively limited
number of collisions.
Today it is quite accepted that most of the primary cosmic radiation, starting from
energies of the order of GeV/nucleon, up to energies of the order of 1015 eV, is
of “galactic” origin. The recent gamma observations produced by dedicated high-
quality satellite measurements also contributed to this belief. They have shown that
the flux of energetic gamma rays, produced to a large extent by the decay of neutral
pions, in turn generated in hadronic collisions with nuclei present in the interstellar
medium, is distributed throughout the entire Galaxy, with a greater intensity in its
central area and gradually decreasing towards the outermost regions of the disc. This
would go against a hypothesis of uniform distribution in intergalactic space, with
entry into the volume of the Galaxy by the outside.
Starting from a certain energy onwards, the so-called “knee” region around 3 ×
1015 eV, we generally speak of the presence of extragalactic components, particularly
for those energies above the “ankle” region at 3 × 1018 eV.
What are the arguments in favour of an extragalactic origin of these extreme
energy particles?
First, charged particles, assuming they were produced within our Galaxy, could
not remain forever confined by the weak galactic magnetic field. In general, if we
assume an average value of the magnetic field B within a given region, we can estimate
the maximum energy of the particles that could remain confined in that region, as
they would move on circular orbits with a radius smaller than the size of the region.
The Larmor radius corresponding to a given magnetic field B is given by R = pT /qB,
where pT is the momentum of the particle perpendicular to the direction of the field.
Table 15.1 shows the energy value of the protons that could be confined within typical
regions of interest for the origin and propagation of cosmic rays, together with the
known value of the magnetic field present in those regions.
Another argument in favour of a different origin lies just in the change in the
observed power law, with an exponent that at 3 × 1015 eV passes from a value of
2.7 to a value greater than 3. Although this difference may seem small, it implies
a difference in the reduction of the flux by at least a factor of 2 for each decade of
energy. An exponent value equal to 2.7 in fact implies a flux that is reduced by a
factor 500 for an increase in energy by a factor 10, while an exponent equal to 3 leads
to a reduction in the flux of a factor 1000 for the same increase in energy.
Finally, very recent measurements seem to indicate a geometric correlation
between the incoming direction of the extreme energy particles and the large-scale
Table 15.1 Typical dimensions of the regions of interest for the origin, propagation and possible
confinement of cosmic rays
Region Typical dimensions Average magnetic Max energy of protons
field
Solar system 100 A.U. = 1.5 × 10–9 T 4 × 1012 eV
1013 m
Interstellar medium 100 pc = 3 × 1018 m 0.5 × 10–9 T 4 × 1017 eV
(mean distance
between stars)
Disc thickness of 2000 l.y. = 2 × 0.3 × 10–9 T 2 × 1018 eV
Milky Way 1019 m
Milky Way diameter 105 l.y. = 9.5 × 0.3 × 10–9 T 8 × 1019 eV
1020 m
distribution of matter in the nearby universe. Associating the arrival direction of the
primary particles with specific areas of the universe is possible only in the case of
neutral radiation (gamma, neutrinos), or for charged particles with very high energy,
which are little affected by the magnetic field.
15.2 Some Historical Considerations About

the Acceleration Mechanisms and the Origin
of Cosmic Rays
As we have seen in the first chapters, from the very beginning the question of the origin
of cosmic radiation has been one of the crucial issues behind the enormous variety
of experimental observations and specific phenomenological aspects. It has been
correctly noted that almost everyone who contributed to the first studies concerning
cosmic radiation also attempted to get a theoretical representation of what the origin
of this radiation could be, even before its commonly accepted extraterrestrial origin.
In many cases these first hypotheses proved to be without foundation over time, even
when they were proposed by some protagonists of the research of the time. We have
already mentioned Millikan’s attempts to associate gamma radiation—hypothesized
as the preponderant part of primary radiation—with nuclear processes originated by
the birth of the elements. But in that period—since the late 1920s—there was no
lack even of other exotic hypotheses, such as those of C. T. R. Wilson and Halliday,
who discussed the role of lightning in the production of cosmic radiation particles,
or those of Bothe and Kolhöster, who considered the role of the gravitational forces
in the acceleration of particles.
Between the 1930s and 1960s a variety of approaches attempted to address the
problem of the origin of cosmic rays and of the mechanisms capable of justifying the
acceleration of these particles, also in relation to the different regions of the energy
15.2 Some Historical Considerations About the Acceleration Mechanisms … 261
spectrum of the primary radiation, from a few hundred MeV to energies in excess of
1019 eV.
Among the approaches that referred to a solar origin and in general to the inter-
planetary environment, we should mention that of Swann [Swann1933], Dauvil-
lier [Dauvillier1934], Menzel and Salisbury [Menzel1948], Richtmeyer and Teller
[Richtmeyer1949], Alfvén [Alfvén1949, Alfvén1959]. One of the first models based
on electrodynamic acceleration processes was developed by Swann [Swann1933],
assuming that these processes took place near sunspots. The mechanisms hypothe-
sized in this approach are, however, generally applicable to the existence of variable
magnetic fields. This approach was able to explain the origin of the low energy
component, with an injection of particles by explosions in the active regions of the
Sun and a subsequent acceleration. However, no predictions could be made about
the spectrum in energy, and the mechanism was closely related to regions in the
vicinity of stars. In other approaches, such as those proposed by Alfvén, the acceler-
ation mechanism was based on plasma beams produced by explosions in the active
regions of the Sun, also linked to the interpretation of Forbush variations. Assuming
that low energy particles were present in abundance in the solar corona, as happens,
they could have been accelerated by this mechanism, increasing their energy in a short
time. The existence of cosmic rays of energy of the order of 1011 –1012 eV could thus
be explained, but not that of particles of even higher energy, which, as we have
seen, could not have been confined within the Solar System. Although these models
initially made a qualitative account of some experimental evidence, a deeper analysis
in subsequent years showed that several specific observations, concerning the abun-
dance of low energy particles during periods of intense solar activity, or the charge
distribution of particles, were not reconcilable with this production mechanism.
Another group of theoretical models referred to the interstellar medium as the
source and place of the acceleration of cosmic radiation. This group includes
the model proposed by Fermi [Fermi1949, Fermi1954] which we will discuss
in more detail in the next section, as well as the theoretical approaches of Fan
[Fan1951], Chandrasekhar and Fermi [Chandrasekhar1953], Morrison, Olbert and
Rossi [Morrison1954] and Davis [Davis1956].
The origin of cosmic radiation in reference to intergalactic (or extragalactic)
space was discussed as early as the 1950s [Bierman1949, Bagge1950, Teller1954,
Cocconi1956]. A “universal” mechanism based on what is called “magnetic pump-
ing” was also proposed in those years [Swann1954, Alfvén1959]. Finally, theo-
ries that explicitly referred to supernovae are also present as early as the 1950s
[TerHaar1950, Colgate1960]. The hypothesis of supernovae as sources of the most
energetic cosmic radiation was strongly revived in subsequent years, also based on
observational evidence not available before and will be discussed in more detail at
the end of this chapter.
15.3 Fermi Acceleration Mechanism
A mechanism capable of accelerating charged particles moving within the Galaxy

was proposed by Fermi in 1949 [Fermi1949]. This mechanism is based on the exis-
tence of magnetic fields moving in interstellar space, generated by clouds of ionized
gas moving with respect to the reference system of the Galaxy. The average density
of matter in interstellar space is of the order of 10–24 g/cm3 , roughly corresponding
to one atom per cubic centimetre. However, it is not uniform throughout the space
within the Galaxy, and there may be regions where the density is considerably greater,
10–100 times greater than the average density. The regions concerned have very large
dimensions, of the order of several light years, and overall, it has been estimated that
they can occupy 5% of the entire volume of the Galaxy. Due to these dimensions,
they are characterized by great stability.
Charged particles with a certain energy can move between these regions in motion,
and occasionally penetrate them, colliding with one of the particles in the cloud or
interacting with an irregularity in the magnetic field. The interaction with these
clouds, in particular with the irregularities of the magnetic field, can produce varia-
tions (both in excess and in defect) in the energy of the particle. Due to the stability of
the magnetic region and the mass enormously greater than that of the single particle,
the process can be treated as a collision with a large mass “reflecting” body.
According to this approach, the probability of a net energy gain by the particle is
higher than the probability of an energy loss, at least starting from a certain threshold
energy onwards, of the order of 200 MeV for protons. The argument in favour of this
consideration is based on the statistical balance that must be achieved between the
degrees of freedom of the moving magnetic fields and those of the particle.
A classic analogy to describe this mechanism is to imagine the interaction of
the charged particle, in motion with a velocity v, with the magnetic cloud, as an
interaction with a body of enormously greater mass, which can move with velocity
V directed against the particle or in the same direction of motion as the particle. In
the first case the particle would acquire energy and move with speed v + 2 V, while
in the second it would lose it, moving with speed v − 2 V. The crucial point of this
mechanism is that interactions of the first type, in which the magnetic cloud moves
against the particle, are on average more frequent than those in which the cloud and
the particle move in the same direction. Some have suggested a representation taken
from everyday life to understand the phenomenon: imagine a car moving along a
road, in the presence of equal traffic in both directions. The car will meet on average
more cars moving in the opposite direction than cars moving in the same direction (in
which case it will overtake them if it has a higher speed). If the car were stationary
on the side of the road, on average, the same number of cars would pass in both
directions. It is therefore a consequence of the relative motion.
In general, since the motion is random in all directions, there will be an angle θ
between the direction of motion of the particle and that of the cloud (Fig. 15.1) and
the probability of a head-on interaction will be proportional to v + V cos θ, while
15.3 Fermi Acceleration Mechanism 263
Fig. 15.1 Schematic representation of the Fermi mechanism. A particle with initial velocity v can
interact with the moving magnetic cloud with velocity V, with a component of its velocity directed
in the same direction as the velocity of the cloud (right) or with a component directed in the opposite
direction (left). In this second case, more likely on average, the particle can gain energy in each
interaction
the probability for the interaction in the same direction will be proportional to v −
V cos θ.
If the cloud moves with average speed V, and the total energy of the particle is W,
the average energy gain for each collision is
2
V
dW = W = β2W
c
where β = V/c, and after N collisions, the energy of the particle will be
2
W = mc2 e Nβ
It can be shown that starting from these considerations, the energy distribution of
the particles, accelerated by this mechanism, would follow a power law, even if the
exponent would only be comparable with that observed experimentally.
Although the Fermi mechanism described (often referred to as the second order
Fermi mechanism, since it depends on the factor β2 ) can justify an acceleration of
particles within galactic space, the time it takes for the particles to reach such high
energies observed experimentally would be too long. Furthermore, the exponent of
the power law that would be derived from these considerations is only qualitatively
similar to that of the observed energy distribution. These, and other considerations,
led in later times to not consider this mechanism valid as a complete and quantitative
description of the acceleration mechanism of cosmic radiation. However, subse-
quent developments were based exactly on Fermi’s original ideas, which therefore
constitute a point of reference in the understanding of these phenomena (Fig. 15.2).
Particles could of course also be absorbed, interacting with matter. However,
assuming a density of 10–24 g/cm3 , and the absorption cross section measured in the
atmosphere, the average free path of these particles, before being absorbed, would be
7 × 1023 m. This distance would be bridged, for particles moving at speed c, in a time
of 2 × 1015 s, equivalent to about 60 million years. These values ensure that there is
Fig. 15.2 Enrico Fermi.

Source AIP Emilio Segré
Visual Archive
sufficient time during the life of the particle for it to be accelerated by the mechanism
described. It can be estimated, in fact, from the distances between the regions with
moving magnetic fields, that the average time interval between successive passages
of the particle in these regions would be 1.3 years, and therefore the number of
interactions with these regions over the course of its average life could be 107 .
As for the possibility that the gain in energy exceeds the losses, their average
values were estimated by Fermi as a function of the initial energy of the particles,
finding a threshold value above which the gain exceeds the loss, and the acceleration
mechanism becomes effective. This value is about 200 MeV for protons, a value that
can still be considered reasonable, assuming sources capable of producing protons
of this energy. However, the threshold is much higher for heavier nuclei, increasing
considerably with the atomic number of the nucleus considered. As an example, the
threshold would be 20 GeV for oxygen nuclei and 300 GeV for iron nuclei, values
that make this mechanism unlikely for a coherent explanation of the acceleration
mechanism, valid for all nuclei. Fermi himself was aware of this limit, considering
this mechanism as inadequate to explain the abundance of heavy nuclei in the cosmic
radiation.
15.4 The Role of Supernovae
Some considerations about the possible role of supernovae in the production or

acceleration of cosmic rays had already been made between the 1950s and 1960s
[TerHaar1950, Colgate1960]. In the work of Colgate and Johnson [Colgate1960]
15.4 The Role of Supernovae 265
reference is made, for example, to the material ejected from the surface of an
exploding supernova as the origin of cosmic rays, and to the fact that the nuclei
at the surface of the star can acquire a much greater energy than the average energy
released in the explosion, due to the different density between the innermost zone
(core) and the outer mantle. The shock wave generated by the rapid increase in
pressure in the core would be capable of imparting extremely high energies to the
outermost layers.
Stellar events that are now called supernovae were observed long before their
nature was understood. The sudden appearance of a very bright star in an area where
no object was visible (stella nova in Latin) represents an event that even the peoples
of the past had observed and recorded. Probably the first recorded observation (the
object now called SN185) dates back to the year 185 AD, by Chinese astronomers.
Very bright supernovae were observed in the year 1006 (SN1006, an object so bright
as to be visible in broad daylight) and in the year 1054 (SN1054, in the Crab Nebula,
an object comparable or even superior in brightness to that of the planet Venus). More
precise observations of such events were then made by astronomers of 1500–1600, for
example by Tycho Brahe in 1572 (SN1572) and by Kepler in 1604 (SN1604). Tycho
Brahe could establish that the object had no relative motion with respect to other
stars, and therefore could not be a nearby object, a matter of primary importance
at that time, since celestial phenomena beyond a certain distance from the Earth
were considered to belong to the place of immutability. These last two objects today
represent the most recent supernovae directly observed within the Milky Way. The
study of the residues of a supernova explosion allows, however, to date also events
from the past that have not been directly observed.
The understanding of these events from an astrophysical point of view, however,
came in much more recent times, when, around the end of the 1800s, the study of the
spectra of the radiation emitted by novae stars turned out to be different from those
of ordinary stars.
To date, the most recent supernova located within the Milky Way dates back to
1868, with a dating made only a few years ago. As for observations of supernovae
in other galaxies, the first was SN1885A, observed by astronomers of the time in the
Andromeda galaxy, although the nature of Andromeda, as an independent galaxy,
located about 2.5 million light years from Earth, was established only in 1925. Obser-
vations of supernovae in other galaxies similar to ours have allowed us to estimate
that the average rate at which these events occur in a galaxy like the Milky Way
is about 3 per century. It should be noted however that entire regions of the Milky
Way are not easily visible due to interstellar dust, so the immediate observation of a
supernova in the visible can also escape.
A detailed observation from the astronomical point of view of a relatively “close”
supernova (SN1987), observed in February 1987 in the Large Magellanic Cloud,
about 160,000 light years away from us, allowed to confirm many of the hypotheses
about the production of nuclei in these events.
If the term nova generically refers to a new bright star, the term supernova seems
to have been used as early as 1931 by Baade and Zwicki, although other authors
had used the same concept even before, with other names, such as “giant novae”
or “exceptional novae” or ’hauptnovae”, in any case to denote energetic explosions

much greater than ordinary novae.
In the case of supernovae, it has been estimated that the energy released is of the
order of 1043 –1044 J, with a maximum brightness that can reach a billion times that
of the Sun. Based on the light curves (trend of brightness over time) a classification
of supernovae into two types (I and II) was introduced. Type I supernovae have
a characteristic, well-defined, light curve, with a maximum reaching an absolute
magnitude of 19.8. The rapid increase in brightness before the maximum is followed
by a decrease at the same rate and then by a slower descent. The speed of the material
ejected in the explosive process is of the order of 107 m/s. Type II supernovae, which
reach a maximum absolute brightness of 16.5, do not have a characteristic light curve,
and the ejection rate is slower.
The explosion phenomena of supernovae have been understood, at least partially,
within the overall evolutionary framework of the stars. The phenomena of nuclear
fusion of light elements, present in the core of a star, allow to balance the gravitational
force of the star. When the lighter elements, which serve as nuclear fuel, begin to
run out, the star begins to contract, and its subsequent evolution depends on the
star initial mass. For not too large mass values (roughly 1.5 times the mass of the
Sun), the star contracts until it reaches enormously large densities (106 g/cm3 ) and
a pressure capable of balancing the gravitational contraction, making the whole a
compact stable system. For masses greater than a certain limit (Chandrasekhar limit)
the process of gravitational contraction does not stop, and the star continues its
evolution towards a violent implosive collapse, with a reflection wave that produces
the expulsion towards the outside of a substantial part of the star core.
The remnants of a supernova explosion (supernova remnants) will consist of
a small super-dense collapsed core (a neutron star or a black hole) and a rapidly
expanding gas nebula. Figure 15.3 shows a spectacular image of the Crab Nebula,
the remnant (6 light years wide) of the explosion of the supernova SN1054, taken
by the Hubble Space Telescope. The orange filaments are the tattered fragments of
the star and consist mainly of hydrogen; blue light comes from electrons spinning, at
speeds close to that of light, around magnetic field lines created by the pulsar located
in the center, an ultra-compact remnant of the exploded star. The different colours
reflect the various elements expelled during the explosion.
The amount of energy required per unit of time, necessary to maintain the level
of cosmic radiation in the entire Galaxy, has been estimated in values of the order of
1034 W. Estimates of this order of magnitude had already been reported in the years
’60 by Ginzburg, and further refined in more recent years [Strong2010], such as (0.7
± 0.1) 1034 W.
If we ask ourselves what could be the source capable of providing such power,
the answer, now discussed for several decades, is precisely that of the explosion
of supernovae. If an energy of the order of 1044 J is released in the explosion of a
supernova, and events of this kind occur on average every 30 years (9.5 × 108 s),
the resulting power is 1044 J/(9.5 × 108 s) ~ 1035 W, compatible with the above
estimate, if we assume for example that 10% of the available energy is converted
into the energy of the accelerated particles. Other sources, such as pulsars, are much
15.4 The Role of Supernovae 267
Fig. 15.3 Crab Nebula, the

remnant (6 light years wide)
of the explosion of the
supernova SN1054, taken by
the Hubble Space Telescope.
Source NASA, ESA, J.
Hester and A. Loll (Arizona
State University),
HubbleSite: gallery, release,
Public https://commons.wik
imedia.org/w/index.php?
curid=516106)
less energetic; they could be responsible, according to recent considerations, for a

fraction of the electrons and positrons present in the high-energy cosmic radiation,
but not for the prevailing part of the cosmic radiation, consisting of nuclei.
Regarding the question of the acceleration mechanism and the location of this
acceleration process, significant progress has been made in recent decades, starting
from the late 1970s [Axford1977, Bell1978, Blandford1978], on the basis of which it
has been accepted that a version of the Fermi acceleration mechanism based on shock
waves, has the characteristics necessary to explain the origin of most cosmic rays
of a galactic nature. In this process, now commonly referred to as ’Diffusive Shock
Acceleration” (DSA), it is assumed that the particles undergo numerous passages
through the shock wave, gaining energy to each passage. The detailed description of
this process, which goes beyond the scope of this chapter, can be found in several
recent works on the subject [Drury1983, Blandford1987, Jones1991, Jones1994,
Malkov2001]. We can only remember that this description leads to an energy distri-
bution of the particles produced governed by a power law, with an exponent very
similar to the one experimentally observed, without resorting to fine adjustments of
the parameters involved. Furthermore, it does not require pre-acceleration phases of
the particles, which can also originate with low energies.
In this context, there are aspects that limit the maximum available energy. First
of all, these limitations derive from the size of the region in which the process takes
place, due to the confinement of the particles up to a certain maximum energy, as we
have seen in the introductory part of this chapter, and from other factors related to the
diffusion of the particles. Another important parameter is the time, which must be
long enough for the particles to be accelerated from very low energies to the required
energy. Using reasonable values for these parameters, maximum achievable energies
are estimated, for protons, of the order of 1014 eV, which is even less than the energy
corresponding to the knee in the energy spectrum of the primaries. It must be said,
however, that if the higher energy part had a notable contribution from heavier nuclei,
such as Iron, the achievable energies would be higher.
Another aspect that has been underlined [Bell1978] is that the value of the
magnetic field in the region of interest for the acceleration of the particles could
be significantly higher than the average galactic magnetic field, estimated to be
0.3 nT. Preliminary evidence of possible magnetic field increases in these regions
has recently been obtained. This could further increase the maximum achievable
energy value, up to values close to the “ankle” region.
For electrons, it must be considered that—although this mechanism may be valid
for these particles—they significantly lose energy due to radiative processes, which
limits the energy that can be reached. The electron injection phase in the acceleration
process must also be hypothesized differently from that assumed for protons and
nuclei.
A plausible synthesis of the origin of cosmic rays of a galactic nature therefore
seems to be possible, assuming this acceleration mechanism in the vicinity of the
residues of the supernova explosion. Using some of the aforementioned hypotheses, it
is possible to justify the achievement of magnetic rigidity values of 1017 V; assuming a
non-negligible presence of even heavier ions, up to Iron, this rigidity could correspond
to energies up to about 3 × 1018 eV.
15.5 The High-Energy Extragalactic Component
Based on the previous considerations, it is reasonable to assume that the highest

energy component in the primary radiation spectrum, certainly that beyond the ankle
region, cannot be of galactic origin. Knowledge of this part of the energy spectrum has
significantly improved in recent years thanks to measurements obtained by the use of
large detector arrays (AGASA, The Pierre Auger Observatory). The measurements
themselves and their detailed interpretation are still ongoing, with discrepancies
between the values measured by the two experiments at the very end of the spectrum,
as discussed in one of the previous chapters. This may be due to the possible presence
of the long hypothesized GZK cutoff, or to different mechanisms in that region.
Certainly not a significant number of events due to high-energy showers are observed
beyond the cutoff. Some retain as plausible that the same mechanism described in
relation to the residues of supernova explosions could also be applied to Active
Galactic Nuclei (AGN) [Hardcastle2010], although other possible mechanisms have
recently been hypothesized, as well as the role of other phenomena with large energy
emission, including gamma-ray bursts.
As regards the localization of the arrival directions of these extreme energy events,
the study of which is possible in principle because at these energies the particles are
not significantly deflected by the magnetic fields, attempts are underway to associate
15.5 The High-Energy Extragalactic Component 269
the direction of arrival of these energetic particles to known sources of astrophys-

ical interest. The first promising evidence of a significant correlation between the
direction of these highly energetic primaries and the distribution of matter, obtained
by the Pierre Auger Collaboration [Abreu2010], have been dampened by subse-
quent measurements, with a correlation that, although existing, is less significant
than previously expected initially. However, the still not very precise knowledge of
the intergalactic magnetic field can partially obscure any possible correlations.
The study of the origin and acceleration mechanisms of cosmic rays, both for
the more energetic component—predictably of extragalactic origin—and for that
of lower energy, discussed in the context of supernova explosions, is in any case
a research field still very active today, with recent or pending measurements, and
theoretical interpretations under discussion.
Some experiments in particular will certainly contribute to the understanding
of these aspects, with detection facilities and measurement campaigns that will
significantly extend the current set of observations available at the moment. Exam-
ples (not exhaustive) are the expansion of large arrays for the detection—even
by hybrid techniques—of very high-energy atmospheric showers, high-energy
gamma measurements from satellites and the detection of neutrinos in underwater
laboratories.
Chapter 16
The Impact of Cosmic Rays
in Applications and in Daily Life
Abstract This chapter provides an extensive list of different fields of application of

cosmic ray muons in daily life and in applied research. These include: the production
of radioactive isotopes in the Earth atmosphere and the associated dating techniques,
also outside our Planet; the radiation dose delivered to living beings by the cosmic
radiation; possible effects of cosmic rays on consumer electronics; muon tomography
techniques based either on muon absorption, now employed for the investigation of
large structures such as volcanoes, or on muon scattering, such as those widely
employed for the scanning of containers in search of potentially dangerous fissile
materials; the use of cosmic muons as probes to monitor the long-term stability of
civil buildings. Additional topics concerned with cosmic ray muons, including those
related to prospections of other natural and man-made structures, the role of cosmic
rays in cloud formation, as well as recent proposals to use the arrival of cosmic
showers as a tool to synchronize distant clocks are presented.
16.1 Introduction
The list of issues about the possible impact of the radiation of cosmic origin and its
applications in various aspects of daily life, discussed in this chapter—although
extensive—cannot be considered exhaustive. In recent years, an ever-increasing
number of articles, technical reports, contributions to conferences, discussion forums
on the Web have appeared on these issues, with the involvement of experts from
different areas. For some of these issues, specialized conferences and workshops
are today held periodically, others have also given rise to industrial applications.
Despite this wide diffusion, or perhaps because of this, there is still a large territory
in which to look for new applications, improve existing techniques or establish new
links between research and application areas, all limited only by the imagination and
creativity of the people involved.
https://doi.org/10.1007/978-3-031-24762-0_16
272 16 The Impact of Cosmic Rays in Applications and in Daily Life
16.2 Production of Radioactive Isotopes by Cosmic Rays

and Dating Techniques
Particles in the primary radiation, in particular the most abundant ones (protons,
helium nuclei …) interact, as we have seen, in the Earth’s atmosphere, giving rise to
showers of secondary particles. The interaction of these particles can also produce
new isotopes: some of them are stable, others radioactive. Today we know that it is
possible, through appropriate laboratory techniques, to detect even small quantities
of these isotopes and to use these products, generated by cosmic rays, in various
applications, especially as clocks capable of dating the age of objects. The concen-
tration of an isotope in a material sample can in fact be used to determine the age of
that material.
The best-known example in physics is the so-called radio carbon (or Carbon
14) dating, which is produced by cosmic rays in the atmosphere of the Earth, and
which is found in every living organism (either animals or plants). The measurement
of Carbon 14 in organic substances has revolutionized the archaeological dating
technique: being able to date an artifact is extremely important for the study of past
civilizations and in archaeological applications. Dating techniques based on the use
of other isotopes can answer questions about how old the Earth is, how it formed,
how and when certain living organisms first appeared or when they evolved into
different forms. The age of the Earth can now be quantitatively estimated on the
basis of measurements of the isotopes of lead and the content of radioactive uranium
and thorium, whose relative proportions depend on the elapsed time and change on a
time scale of billions of years. For dating applications on shorter time scales, such as
those related to the history of civilizations and archeology, where times are measured
on a scale of hundreds or thousands of years, dating can be obtained by studying
the amount of Carbon 14. The Carbon 14 dating technique was introduced in the
1940s by Willard Libby [Libby1949a, Libby1949b], a chemist at the University of
Chicago, Nobel laureate in 1960.
The carbon that we find on Earth consists almost exclusively of a mixture of two
isotopes: Carbon 12 (at 98.9%) and Carbon 13 (at 1.1%). Other carbon isotopes
with average lifetimes of minutes or even shorter can be created in the laboratory
but are of no particular geological interest. Carbon 14, however, which is produced
by cosmic rays in the Earth’s atmosphere, is of special importance. To evaluate the
importance of Carbon 14 dating we must consider that carbon is a critical element
for life in all its forms, as all living organisms consist of a variety of carbon-based
molecules; long chains of carbon atoms make up, together with other atoms, the DNA
and complex molecules that are necessary for life and reproduction. The exchange
of carbon dioxide between plants, animals, the atmosphere, the oceans is a necessary
part of the cycle of life.
When cosmic rays penetrate the atmosphere, they interact with the air at the highest
altitudes and already at a height of about 15 km from the surface of the ground, most
of them have given rise to nuclear reactions in which neutrons can also be produced.
Neutrons lose energy to the point where they can be captured by nitrogen nuclei to
16.2 Production of Radioactive Isotopes by Cosmic Rays and Dating Techniques 273
produce Carbon 14, according to the process

14
N + n → 14 C + p
This new isotope, Carbon 14, is radioactive, with a half-life of 5730 years. It
can decay into nitrogen after emitting an electron. Despite its radioactive properties,
Carbon 14 retains all its chemical properties to combine with other atoms and form
molecules; in particular, it can combine with oxygen and produce the radioactive
carbon dioxide 14 CO2 , which spreads throughout the biosphere. The molecules of
14
CO2 can in fact be diffused through the air currents in the atmosphere, can be found
in the rain and eventually be absorbed by plants, which in turn are used as food by
humans and other animals. Carbon dioxide can also end up in the water of rivers and
oceans. In this regard, the time scale with which this Carbon 14 diffusion process
takes place in the atmosphere is important. It has been estimated that it takes about
7 years for it to spread from the stratosphere to the regions of the atmosphere near
the ground to reach living organisms or the water of rivers or oceans. This 7-year
time is very small compared to the half-life of 5730 years (Fig. 16.1).
The quantity of 14 C present on Earth, which is produced by the processes
described, with subsequent decay, sets the proportion of this isotope with respect
to the 12 C + 13 C isotopes to a very precise value in all living organisms, about 1
atom of 14 C every 1012 atoms of 12 C. When the organism (plant or animal) dies, it
no longer takes on additional 14 C from the outside, while the one contained within it
progressively decays, shrinking more and more over time. The proportion 14 C/(12 C
+ 13 C) therefore decreases over time.
Fig. 16.1 14 C production mechanism in the Earth’s atmosphere, induced by the interaction of
primary cosmic rays. Among the products of nuclear interaction with the nuclei of the atmosphere,
neutrons can in turn interact with a nucleus of 14 N producing 14 C
A measurement of this ratio, carried out on a small sample of the material to

be examined, can therefore provide the age of the sample, taking into account the
law of radioactive decay, which regulates the number of 14 C nuclei still present
after a certain time interval t. The measurement of this ratio can be made, as in the
original method introduced by Libby, by detecting the electrons emitted following
the 14 C decay process (and distinguishing them from the background produced by the
cosmic rays themselves or by decays produced by other radioactive natural material
placed in the proximity of the detector). Considering the activity of this isotope, it
can be estimated that for each gram of carbon present in the sample in equilibrium
conditions, 13.56 decays/min occur, a value already very low in itself, and which
progressively decreases when the age of the sample is a few thousand years old,
which makes the measurement difficult. For example, for a sample dating back to
3000 B.C. the activity would be reduced from 13.56 to a value of 7.39 per min per g.
The 14 C decays by beta emission, according to the scheme
14
C → 14 N + β − + υ
The emitted electrons have a continuous spectrum in energy, as occurs in beta

decay, since in the final state the energy is distributed between the residual nucleus,
the electron and the antineutrino, and the maximum energy of the electrons (end-
point) is only 156 keV, which makes it even more difficult to distinguish them from
the background of other natural decays. These factors limit the applicability of the
method, in the case of very ancient objects, since the measured values would be
indistinguishable from the background.
Another technique for the detection of 14 C, introduced in the 1970s, consists in
taking a small amount of the sample (even a few milligrams), and using it in the
source of an accelerator. The accelerated ions would then be deflected by a magnetic
field, based on their energy, charge and mass, thus being able to distinguish 14 C from
other ions.
Realistically, the maximum age that can be sampled with these methods, in the
best conditions, is a few tens of thousands of years, which still makes them suitable
for dating finds of historical interest. It must be said, however, that the results of
this technique can be affected by various factors, some due to natural causes, others
related to human-induced phenomena. The variation in the flux of cosmic rays over
the centuries or the solar cycle can induce variations in the amount of 14 C, slightly
distorting the dating estimates in case one wants to make precision measurements.
For example, it is known that in the period between 1650 and about 1730, when
sunspots were practically absent, the flux of low energy cosmic rays arriving on
Earth was greater, as the corresponding amount of 14 C found in the findings that date
back to those years. However, a significant cause of variation in the 14 C content in the
atmosphere also derives from human activities. The use of fossil fuels has produced
an increase in carbon content with a low 14 C content in the atmosphere, altering the
ratio by approximately 3%. Another major consequence on the relationship between
the different isotopes of carbon in the atmosphere was produced by the nuclear tests
carried out from the end of the Second World War up to the 1960s, which produced,
16.3 Cosmic Ray Dating Outside the Earth 275
in the following years, an amount of 14 C even double than the basic values. Currently,
after nuclear testing in the atmosphere ceased in 1960, this quantity is slowly returning
to its original value.
16.3 Cosmic Ray Dating Outside the Earth
Apart from the production of 14 C, which certainly represents the most relevant aspect
in terrestrial dating techniques using isotopes produced by cosmic rays, several other
isotopes of some interest in longer time scale dating are also produced as a conse-
quence of the arrival of primary cosmic rays on Earth and in the Solar System. On
Earth, for example, 10 Be (with a half-life of 1.4 × 106 years) or 26 Al (half-life about
7.4 × 105 years) are isotopes of interest for this type of analysis. The relationship
between these two isotopes, in particular, can be used to study their diffusion in
geological processes.
Dating techniques linked to the existence of cosmic rays have also been applied
to samples of the lunar soil. In the case of Moon, without an atmosphere, primary
cosmic rays can interact directly with the ground, producing isotopes on the surface
and up to a certain depth. The part of the ground close to the surface receives a greater
dose of radiation than that localized at a certain depth; however, due to the impact
of meteorites—microscopic or with a certain size—the lunar soil is continuously
modified, since in the impact of a meteorite fragments of soil are expelled, leaving
a certain area uncovered and covering instead other areas where the fragments are
deposited after having travelled a certain path. This is made possible by the lack of
an atmosphere and by a gravity smaller than on Earth.
Information on the time scale involved in these processes can also be extracted
through the damage effect produced by cosmic rays in certain crystalline materials.
The energy deposited in these crystals can be stored for very long times and subse-
quently released, in the form of light, when the crystal is heated. This is thermolumi-
nescence, a technique widely used in the laboratory for dating finds of historical and
archaeological interest, in particular ceramic artifacts. From samples of the lunar soil,
up to a depth of 2 m, carried out during the Apollo missions in the early 1970s, it was
possible to measure the exposure to cosmic rays of the different soil layers at various
depths, and therefore to understand how the lunar soil is periodically “remixed”, at
least in its most superficial layers. It has been estimated, for example, that the most
superficial layer (about 0.5 mm) is remixed on average every 10,000 years, while an
average time of one million years is required to move and redistribute one centimeter
of soil thickness. Modifications of the lunar soil below the depth of 10 cm, up to a
maximum depth of one meter, require time scales of one billion years.
In the case of thermoluminescence, the result obtained is the consequence of a
radiation dose absorbed over very long periods. Other useful information on objects
located outside the terrestrial environment can be extracted from the study of the indi-
vidual tracks generated by the passage of a primary particle. These tracks have been
observed, for example, in samples of material coming from comets, or in interplan-

etary dust, using etching techniques similar to those used in the case of nuclear
emulsions to detect the passage of particles. The very existence of these tracks
demonstrates, for example, the fact that these fragments were not exposed to very
high temperatures, above about 500 °C, during their journey, including re-entering
the atmosphere, otherwise the tracks originally present would have disappeared.
The presence of some isotopes of the noble gases in specimens coming from inter-
stellar dust also demonstrates their extraterrestrial character and their origin, due to
interactions of primary cosmic rays in the material.
The period of time during which meteorites were exposed to the flux of primary
cosmic rays indicates the period during which they existed as small objects (dimen-
sions of the order of one meter) in a free state or bound to the surface of a body of
larger dimensions. It therefore represents a sort of “age” of the meteorite itself and
can be estimated in some cases precisely on the basis of the interaction of the cosmic
rays in these bodies. This time interval, which often represents the time required for
the meteorite to reach the asteroid zone in the Solar System to the Earth, can provide
useful information about the mechanisms involved in operating this transport from
one region to another of the Solar System. Galactic cosmic rays, that is, of high
energy, mainly protons, can interact with these objects, up to depths of the order of
one meter or a few meters, according to their energy, and following this interaction
protons or neutrons can be emitted by the various elements present in the sample, by
processes called spallation.
Due to these interactions, a large number of isotopes, stable or radioactive, of
rare species in nature can be produced. They range from noble gas isotopes, such
as 3 He, 21 Ne, 38 Ar and 83 Kr to other elements, such as 10 Be, 26 Al, 36 Cl, and many
others. The concentration of radioactive isotopes can give us information on the
rate of interactions between cosmic rays and materials, while the accumulation of
stable isotopes can measure the time elapsed since the bombardment with cosmic
rays began, once the volume of the meteorite was exposed to their flux. As a result
of these investigations, it has been estimated, for example, that almost all meteorites
are older than one million years, which indicates that this is the minimum time
required for their orbit to intersect that of the Earth, according to simulations of their
trajectories. Although most meteorites show an age of some tens of millions of years,
some of them, especially those rich in iron, have been shown to be much older, up
to one billion years.
16.4 The Radiation Dose Produced by Cosmic Rays

on Earth and in the Solar System
Ionizing radiation is one of the components of the environment in which we live,

whether it is produced by natural or artificial causes. Examples of “natural” ionizing
radiation, that is present in the environment independently of man, are those due
16.4 The Radiation Dose Produced by Cosmic Rays on Earth and in the Solar … 277
to cosmic rays or to the natural radioactivity of rocks, while “artificial” sources of

radiation may be those related to diagnostics and therapy in the medical field, to
the functioning of particle accelerators or to radioactive elements, which, although
natural, are artificially produced, that is, as a result of human activities.
Cosmic rays, like other ionizing radiations, deposit a certain amount of radiation
dose in the materials with which they interact, as well as in the tissues of living organ-
isms. We will see in the next section the effect of such radiation dose in electronic
circuits that are now used in most human activities, while we will discuss here the
effect of the dose absorbed by people and due to radiation of cosmic origin.
Living beings coexist since their origin with a background level of natural radia-
tion, but it is important to minimize as far as possible the dose of radiation absorbed
by extra-natural causes. The overall effective dose absorbed by an organism in a
certain time interval, which depends on the energy deposited per unit of mass, as
well as on a biological damage factor for that specific radiation, can be expressed
in Sievert. However, it is important to evaluate the absorbed dose per unit of time,
which can conveniently be expressed in µSv/h (microSievert per hour). It has been
estimated that the average effective dose to which an organism is subject—because of
the cosmic radiation and of the terrestrial radioactivity, mainly due to Radon gas—
is about 2.4 mSv/year (milliSievert per year), equivalent to 0.27 µSv/h. Slightly
different values are reported depending on the territory. Current laws, sometimes
slightly different from country to country, then establish, in accordance with the
results of medicine, what are the maximum allowable doses for the different types
of workers.
The various factors that contribute to determining the average dose absorbed by a
person are slightly different depending on the environment, habits and other factors.
As an example, the following figure shows, through a pie diagram, the various causes
contributing to the radiation dose and the corresponding fraction to the overall dose,
in the case of the Italian population (Fig. 16.2).
This diagram shows that the effect of cosmic rays at sea level, as regards the dose
absorbed by an individual, amounts to just under 10% of the total average absorbed
dose. This dose is essentially due—again referring to sea level—to the secondary
radiation, since the primary component is essentially absorbed by the atmosphere.
In fact, the presence of the Earth’s atmosphere offers considerable protection against
these radiations. With the exception of particular periods of solar activity (in which
the low-energy components can make a notable contribution), the components that
essentially produce this dose of radiation are the galactic, high-energy components.
At altitudes significantly higher than sea level, the intensity of the secondary
radiation becomes higher, and consequently also the radiation dose. As an example,
the dose due to cosmic rays doubles for altitudes around 2500 m, while it even
becomes almost 5 times greater in the case of the Tibetan plateau, located at an
altitude of over 4000 m. Slight variations can be expected going from locations near
the equator to higher latitudes.
If such doses refer to populations that on average always live in the same territory,
the case of people who frequently travel by plane is different, in particular civilian
flight crews (pilots, on-board personnel) or military, who pass a consistent number of
Fig. 16.2 Percentage distribution of the various causes that contribute to the absorption of a certain
dose of radiation in human organisms, in the case of the Italian population
hours each year at high altitudes. The central part of each scheduled flight takes place
at cruising altitudes of around 10,000 m, with values ranging from 7000 to 12,000 m
depending on the type of aircraft and route, while the initial and final phases of the
flight, with a duration of 15–20 min each, take place between the initial (or final)
altitude and the cruise altitude. For the personnel, this can lead to an increase in the
total absorbed dose over the course of a year of 2–4 mSv, depending on the routes
followed and the total number of flight hours, which corresponds to a total dose of
2–3 times greater than that absorbed by the rest of the population.
For this reason, staff people are subject to periodic routine medical checks, which
also take this aspect into account. Small differences can be expected between flights
that take place mainly at low latitudes and those that take place at latitudes close
to those of the poles. In any case, these are still very low dose levels and statistical
studies have so far shown no association between these doses of radiation and an
increased incidence of cancer cases. A lower threshold is recommended in any case
for pregnant women, for whom it is suggested not to exceed a dose level of 1 mSv
throughout pregnancy, a dose that would be reached over a hundred flight hours
at high altitudes and latitudes. As for frequent passengers, significant doses could
accumulate only in the case of thousand flight hours per year. To give practical
examples, a London–Singapore–Melbourne flight would involve a dose of 42 µSv,
which means that it would take over 20 such flights in a year to accumulate a dose
of 1 mSv. These estimates lead to consider as very marginal the risk from cosmic
radiation in almost all cases.
The situation is completely different in the case of astronauts and space missions
that take place outside the Earth’s atmosphere, in the space surrounding the Earth or
in any mission to other bodies in the Solar System. In this case the radiation dose
16.4 The Radiation Dose Produced by Cosmic Rays on Earth and in the Solar … 279
can derive from one or more of these different components: the primary radiation
of galactic origin, composed mainly of protons and light nuclei of high energy, the
particles emitted into interplanetary space by the Sun during intense periods of solar
activity and finally the charged particles trapped around the Earth by the Earth’s
magnetic field. The first human missions in space flights began, as it is known, with
the flight of the cosmonaut Gagarin, in 1961. Since then, in more than 60 years, a few
hundred astronauts have spent more or less long periods in space, in orbit around
to the Earth at various altitudes, and some flying to the Moon. The hypothesis of
human missions on or around Mars has been and still is under discussion, in relation
to which the assessment of the possible doses of radiation absorbed and strategies to
improve radiation protection are particularly crucial.
As we have previously seen, the average dose that each individual receives while
living on Earth, due to the various causes of radiation, including cosmic radiation, is
about 2.7 mSv per year. This is a very low dose, and international standards require a
person working in contact with radiation sources to take annual doses not exceeding
50 mSv. In the case of space missions, the limits set, assessed in such a way as to
keep the risk of medical damage still low, are higher, in the order of 100 mSv.
What are the predictable radiation doses for space missions? In the case of the
lunar missions, the travel times were around ten days overall, with absorbed doses
from 2 to 12 mSv. A week-long orbital mission around the Earth, at an altitude of
400 km, would produce a dose of around 6 mSv. A stay on board the International
Space Station, which orbits at an altitude of 300–400 km, would give rise to doses
between 80 and 160 mSv, depending on the activity of the Sun in that period, therefore
already potentially exceeding the permitted limits. A possible mission to Mars, for
an estimated time of 3 years, could result in a dose of up to 1200 mSv. It is therefore
evident that the dose of radiation due to the presence of cosmic rays is one of the
limiting factors, if not the main one of any long-lasting human mission.
On the one hand, measures are possible to reduce the dose of radiation absorbed
in space missions. In the case of long stays on the surface of the Moon or Mars, the
possibility of choosing suitable areas where there is a weak magnetic field has been
investigated, that could reduce, by deflecting them, the intensity of the incoming
particles. The intensity of the magnetic field on these two bodies of the Solar System
has been carefully monitored at different latitudes and longitudes, but it does not
seem sufficient to provide an adequate shield from radiation. Recall that on Earth
this shielding effect is provided both by the presence of the magnetic field and by
the atmosphere, practically absent on the Moon and on Mars, even if on the latter
there is a weak atmosphere consisting mainly (95%) of CO2 , at a pressure of about
1% compared to the terrestrial one.
Any people who remain for long periods on the surface of these celestial bodies
should minimize the amount of time spent outdoors or on the surface, for example on
the occasion of explorations to be done at ground, and it is required instead to spend
a significant fraction of the time in underground environments or with an adequate
protection from radiation.
We do not address in this context the aspect of the type of damage caused by
radiation and the studies in this area, which are very active, with the participation of
physicists, medical doctors and biologists.
As for the aspects to consider about the possible strategies to be used to reduce the
absorbed dose and prevent possible radiation damage caused by cosmic rays during
space missions, the experimental monitoring of the dose in the different environments
in which astronauts are operating, and the theoretical evaluation of the effects of the
materials present are among the main activities carried out for this purpose.
An essential part is in fact the radiation dosimetry, which consists in monitoring,
characterizing and quantifying the radiation present in these environments. To this
end, a large series of dosimeters accompany astronauts during missions in space.
Detailed measurements on phantoms that incorporated radiation sensors arranged in
correspondence to the different organs were carried out in space, to extract not only
the average absorbed dose, but also its distribution within the body, corresponding
to the organs of interest. Dosimetric measurements are also carried out in the various
work environments of astronauts, particularly in the International Space Station, to
assess the dose in each environment and identify the areas with less exposure, based
on the structure of the station itself.
If on the one hand the first strategy to be used to reduce the effects of radiation
is to limit the time spent in space, particularly outside the spacecraft in the case of
extravehicular activity, on the other hand, solutions are also put into play to protect,
as far as possible, people aboard spacecrafts. They are based on various strategies for
shielding, adapted to the environments in which astronauts spend most of their time on
board, with appropriate layer thickness and choice of materials. In the International
Space Station many environments are shielded with polyethylene, a material rich
in hydrogen. It must be taken into account that the primary high-energy particles
can create secondary particles (neutrons and other charged particles) in the same
material used for the shielding, and that sometimes the secondary ones can create
more problems than the primary ones from the point of view of biological damage,
if the thickness of the screen is not sufficient to contain them. From this point of
view, light materials with high hydrogen and carbon content may be more suitable
than heavy materials such as lead, in which more secondary are generated. Tests and
calculations have shown that these materials are even more suitable than aluminum.
Another suitable material is water, although its oxygen content makes it slightly
heavier than polyethylene. The overall mass of the spacecraft and its components
is a determining factor, since a greater mass to be brought into orbit or sent over a
long distance requires larger thrusters and greater energies. It should be considered,
however, that a certain amount of water, as well as organic liquids, is in any case
on board a spacecraft, and that it can be housed in such a way as to contribute to
the shielding of people. In any case, the thickness of the shielding materials (5–
7 cm) provides a dose reduction of only 30–35%, while the remaining 65–70% is
able to cross the shield and therefore contribute to the absorbed dose. In the case of
permanence on the surface of other bodies, such as the Moon or Mars, the screens
can also be chosen using other materials, without necessarily transporting them from
the Earth. In addition to “absorbent” screens, other solutions based on the use of
16.5 Electronics and the Effect of Cosmic Radiations 281
magnetic fields to deflect incoming particles have also been studied; however, these
solutions in turn involve the use of heavy materials to create a suitable magnetic field.
Further strategies to reduce not the dose of radiation absorbed but at least mitigate
their effect, makes use of an appropriate diet and appropriate drugs, sometimes also
used in the terrestrial environment in case of exposure to high doses of radiation.
From what has been discussed, even briefly, it is clear that the problem caused by
cosmic radiations in any long-term stay in space is a problem that is not easy to solve
and at the moment represents one of the major challenges posed to space travels.
16.5 Electronics and the Effect of Cosmic Radiations
Electronic circuits, especially those based on semiconductors, use electrical charges

located in a certain area of the material as a means to define elementary binary
information (0, 1). Any noise source that is able to modify these charge distributions
can alter the information content. The noise source can come from the power line
itself, from electromagnetic radiations or even from sources of ionizing radiation,
including the cosmic radiation or any radioactive material placed near the circuit.
In a 16 kb memory chip, for example, introduced in 1977, the number of electrons
in the single cell was of the order of 106 , comparable to the number of electrons
created by the passage of an alpha particle along a path of a few microns, typical
cell dimensions. This justifies the non-negligible probability that the passage of an
ionizing radiation through the dimensions of the cell can induce a charge sufficient
to alter the content existing at that moment.
In the 1970s it became clear that ionizing particles could momentarily alter the
content of some bits in a memory chip, transforming “1” into “0” and vice versa
(Fig. 16.3).
Fig. 16.3 In a set of

memory cells, all initially set
to 0, one of the cells can
change its state, from 0 to 1,
following the passage of an
ionizing radiation (alpha
particles, cosmic rays, …)
This type of effect, called “soft fail” (or failure), or single event upset, even if
momentary, by altering the information contained in that bit, can produce a malfunc-
tioning of the entire circuit or produce an incorrect information. Soft fail effects
must be distinguished from permanent damage to the circuits due to the cumula-
tively absorbed radiation dose, effects which occur only with very high doses. The
alteration of the content of an elementary cell does not necessarily lead, however,
to the malfunctioning of the entire system, for example a computer that uses it. The
computer may not use the contents of that cell during the operation in progress,
or overwrite a new, correct information on that cell before reading the information
which was altered.
In the first DRAM (Dynamic Random Access Memory) circuits, introduced in
those years, the material of the ceramic package sometimes contained impurities,
with the presence of very small quantities of alpha radioactive material, which could
induce effects of this kind. The effect was then systematically studied, both to assess
the purity conditions which the materials had to meet to reduce these effects below
a certain value, and to understand how to improve immunity to these effects and
possibly diagnose and correct them. Interestingly, from an anecdotal point of view,
that effects of this kind in electronic components were imagined in a science fiction
story by Isaac Asimov in the 1950s (Caves of Steel, 1952), i.e., 25 years before they
were actually revealed, an event for which Asimov received recognition.
At sea level, the particles of cosmic radiation that are most likely to induce soft
errors are neutrons. Although neutral, they can interact with the nuclei of the material,
being captured and producing charged secondary particles that induce charges in the
chip. In the 1990s, it was estimated by IBM that soft errors of the order of one per
month could be induced in a 256 Mb memory of a PC due to cosmic rays. Of course,
the flux of cosmic rays depends on altitude, so at higher altitudes, the rate of these
errors also increases; this effect is enormously amplified aboard airplanes, which
travel at high altitudes. Conversely, under considerable shielding (for example in
underground laboratories) it has been verified that this rate is strongly reduced, even
if a small contribution remains, also due to the natural radioactive material present
in the rocks. The average rate of these possible errors also depends on the particular
period within the solar cycle [Ziegler1996a, Ziegler1996b].
Certain cases have been documented with certainty in which the transformation
of a bit into a memory location led to incorrect results or to the malfunctioning of
a system. Among these, for example, the case of the counting of votes during an
election, in which the system had assigned to one of the contenders a number of
votes higher than the actual one, equal to 4096 (212 ), due to a bit flip which had
triggered a bit from 0 to 1. It was also reported, on one occasion, the malfunctioning
of the automatic pilot system of an airliner, which due to this caused the aircraft to
momentarily lose altitude.
Proper chip design can help reduce the importance of these effects, through a
suitable design of geometry, semiconductor type, and choice of package materials,
such as avoiding the presence of 10 B and instead using materials enriched in 11 B,
which does not have this problem.
16.6 Muons and the Origin of Tomographic Techniques 283
It is important in any case, since the rate of these errors cannot be reduced to
zero, to carefully diagnose their presence and find solutions to correct the possible
errors that result from them. From a hardware point of view, the circuits can be
replicated—with consequent complexity and costs—in order to compare the result
obtained from multiple copies of the circuits. A strategy sometimes used, especially
in those situations where the reliability must be very high, such as in the applications
of air or space flights, is to have three copies of the same circuit and, in case of
discrepancy between the output of the three circuits, choose the value that two out
of three have. Software techniques for the correction of this type of errors use, at
the level of the compilation of the programs, a temporal replication of the program
instructions with verification of the results obtained from sequences of instructions
executed several times.
The diagnostics and possible correction of soft errors can be done, at least in some
cases, by introducing redundant data into the ordinary data flow, and by introducing
checks that evaluate and correct the error. For example, by introducing an additional
bit and establishing an appropriate parity of the bits, it can be verified whether some
bit flip has been introduced in the sequence of bits ordinarily assigned to the data
(see Appendix N).
The rate of these soft errors, as we have said, depends, for a given system, on
many environmental factors, linked to the location of the device and to the particular
operating conditions (time of the year, solar cycle, occurrence of catastrophic solar
events …). This explains why the estimates can vary widely between different studies.
The error rate is sometimes expressed as the number of errors per unit of time (Failure
in time, FIT = 1 error per 109 h of component operation), or as the average time
between two successive errors (Mean Time Between Failures, MTBF).
Quantitative evaluations of these effects by IBM in the 1990s led to estimates of
possible soft errors on the order of 1 per month per 256 Mb of RAM installed in a
PC, which corresponds to an error rate of 3.7 × 10−9 per bytes per month (1.5 ×
10−15 per byte per second).
Other evaluations report soft error rates in DRAM memories between 10 and 100
FIT per Mb. Based on this value, a system with 10 Gb memory should show such an
error between 1000 and 10,000 h.
16.6 Muons and the Origin of Tomographic Techniques
Although as early as the 1930s it seems that some measurements of the muon flux
in the London Underground were used to verify the thickness of the overlying rock,
the first documented attempts to use cosmic muons as a probe to explore the interior
of large solid structures probably date back to the mid-1950s [George1955], with
some measurements carried out inside a tunnel, above which the layers of solid rock
shielded a fraction of the muon component. Today, since a few years, an incredible
variety of studies related to the use of secondary cosmic muons have appeared in the
literature, with applications ranging from the exploration of volcanoes to archeology,
to the control of the content of potentially dangerous fissile materials. As this sector
has received a significant boost in very recent years, many projects are still under
development and further applications can be found by consulting the documenta-
tion available on the Web. In this context we will give in the following sections an
overview of the main methodologies in which muon detection has proved useful for
the exploration of the surrounding world. Further reviews discussing these issues can
be found for example in [Checchia2016, Bonomi2017, LaRocca2018, Bonechi2020].
It is useful to distinguish in this field between those applications in which the
absorption of muons is used to obtain information on the amount of material crossed
by the particles, as described in the case of the search for hidden chambers inside
the Egyptian pyramids, and other applications in which use is made of another
phenomenon to which muons, as well as other charged particles, are also subject, the
phenomenon of multiple scattering. This phenomenon, as we have seen, is particu-
larly relevant in the case of materials with a high atomic number and can therefore
detect the presence of these materials even within volumes that are difficult to access.
One of the studies that certainly marked the use of muons in these investiga-
tions is represented by the work of Alvarez and collaborators, published in 1970
[Alvarez1970], concerning the possibility of exploring one of the Egyptian pyra-
mids in search of hidden chambers inside. The proposal to carry out measurements
of this kind had already been advanced in 1965 by Luis Alvarez, an American
physicist of Spanish origin, who was to receive the Nobel Prize in 1968 for studies
related to particle physics. Furthermore, Alvarez’s name is linked—in addition to the
Manhattan project—to various scientific issues of an interdisciplinary nature, from
the hypothesis of the extinction of dinosaurs following the impact of a large mete-
orite on Earth (based on observations of the content of Iridium in some rocky layers)
to the study of ballistic information concerned with the murder of John Kennedy.
Alvarez’s proposal on muon experiments within a pyramid was sent—as the author
himself testifies—to a group of cosmic ray physicists and archaeologists, to evaluate
its feasibility, as well as to the political institutions of the countries involved. After
it was approved, the installation of the detector and equipment necessary to operate
it within the Pyramid of Chefren (Kafre) began in 1967.
Unlike other pyramids, even larger, such as the Pyramid of Cheops, which contains
many internal chambers, it seemed that in case of the Chefren pyramid, despite being
an enormous structure with sides equal to 215 m and an (original) height 143 m, only
a single internal structure, called Chamber of Belzoni, were present. The question
therefore arose whether this structure could also contain other rooms, which until
then had remained hidden from exploration or looting. Knowing the location of the
detector with respect to the pyramid structure, it would have been possible to evaluate,
along each direction, the actual thickness of material crossed and the consequent
absorption of the muon flux coming from that direction.
The detector used was based on a set of spark chambers, large area detectors with
digital reading, capable of tracing and reconstructing the trajectories of the particles.
During about two years of measurements, the detection apparatus had reconstructed
about 100 million muons coming mainly from a cone around the vertical, allowing to
16.6 Muons and the Origin of Tomographic Techniques 285
Fig. 16.4 Schematization of the Pyramid of Chefren, 215 m side and 143 m high. In the only known
chamber, the Berzoni chamber, located in a slightly displaced position with respect to the center of
the base, a detector capable of reconstructing the trajectories of muons crossing the solid material
of the structure can provide information on the presence of any hidden chambers. Experiment
performed by Alvarez and collaborators [Alvarez1970]
measure their flux as a function of direction, and to exclude the presence of additional
large chambers inside the pyramid (Fig. 16.4).
This experiment, although it gave a “negative” result, is of considerable impor-
tance in several respects. First of all, it opened the doors to a series of possible
investigations in this direction, which were actually carried out in the following
decades, and which in some cases led to archaeological discoveries of interest, such
as the evidence of further hidden chambers in the structure of the Large Pyramid, or
Pyramid of Khufu. In the latter case, complementary evidence was obtained from
independent measurements, performed with nuclear emulsions located in one of the
known chambers, with scintillators still located inside the same chamber and finally
from gas detectors located outside the pyramid [Morishima2017, Procureur2023].
Following the original Alvarez experiment, the use of this technique has been gener-
alized to apply it to the study of large natural structures, such as volcanoes or other
geological structures. Measures of this type are still in progress on different structures
and with slightly different purposes and are now a commonly accepted technique.
But, more generally, the awareness that cosmic muons, as “natural” sources of radia-
tion, could give information about the surrounding world, has led to numerous other
applications, as we shall see, not necessarily based on the absorption of muons in
a material solid, but on other possible phenomena characteristic of their interaction
with the media.
It must also be said that Alvarez’s experiment was often used as a model for
conducting educational studies or simulations of the phenomenon. A few years ago,
the “Pyramid Hunters” educational project was the winner of the 2016 edition of
the “CERN Beamline for Schools” competition, dedicated to educational projects
presented by school teams and to be conducted through the CERN facilities. In this
study, proposed by a Polish team [Gutowski2018], in addition to a modelling of the
geometry of the Chefren Pyramid, in order to evaluate the absorption of muons in

the different directions and compare it with the results of Alvarez and collaborators,
some measurements was organized to experimentally assess in detail the absorption
of muons in real pyramid blocks, using a beamline from the PS accelerator at CERN
and a particle identification system.
In the following section we will review in greater detail some of the specific
applications of cosmic muons in tomographic investigations.
16.7 Tomographic Techniques Based on Cosmic Muon

Absorption
The use of the technique based on the absorption of muons is particularly useful in the
case of relevant structures, which require very penetrating probes to be able to give
information on their interior. If the dimensions of the objects are of the order of one
meter, or less, the use of X-rays (or possibly neutrons) may be a reasonable choice.
The X-ray attenuation coefficients (energy 30 keV) are of the order of 30 cm2 /g in
the case of lead, and therefore even a lead thickness of several cm can be crossed
by a non-negligible fraction of these photons. However, for greater thicknesses, only
highly penetrating particles, such as high-energy cosmic muons, can penetrate inside.
These considerations are the basis of the absorption-based imaging technique,
which can also be applied to large solid structures, such as the summit of volcanoes.
Even in this case, the measurement of the muon flux coming from the different
orientations, especially if compared with the corresponding flux measured from the
“open sky”, that is from angular regions in which the muons do not cross any solid
material, can give quantitative information on the fraction of muons absorbed, and
therefore on the quantity of material crossed (density of the material integrated on the
length of the path, usually called opacity). In the simplest case in which a constant
density of the rock is assumed, this reduction in flux can highlight the presence of
cavities inside the volcano.
This technique, introduced in the mid-90s [Nagamine1995], has been success-
fully applied in recent years to the study of various volcanoes (active and inac-
tive) in various parts of the world [Tanaka2001, Tanaka2007a, Lesparre2010,
Ambrosi2011, Marteau2012, Ambrosino2014, Ambrosino2015a, Carbone2014,
Tanaka2014, Tanaka2016, LoPresti2018, Olah2018, Tioukov2019, LoPresti2020]
and is still under development and further applications.
From an experimental point of view, the use of this technique requires a muon
tracking detector (telescope), used in transmission mode (i.e., with the object to be
explored positioned between the open sky and the detector, see Fig. 16.5). The result
of measurements made with a setup of this kind provides two-dimensional density
maps, with spatial resolutions that depend on the tracking capabilities of the telescope
and on the amount of multiple scattering in the rock and materials surrounding the
detector. Many aspects of the detector’s performance, such as detection efficiency,
16.7 Tomographic Techniques Based on Cosmic Muon Absorption 287
Fig. 16.5 Schematic of the radiographic technique of the summit of a volcano, based on the absorp-
tion of cosmic muons. A muon tracking detector, hypothesized as a telescope consisting of three
detection planes, sensitive to the position, is used to measure the flux of muons as a function of the
arrival direction, obtaining information on the amount of solid material crossed by the muons along
the various orientations
resolution, stability, sensitive area, as well as portability and cost, ultimately deter-
mine the difference between various projects and contribute to the feasibility of the
measurements.
Although absorption tomography can only provide two-dimensional density
maps, in principle the combined use of multiple telescopes, which view the volcano
from different angles, coupled with numerical imaging algorithms, can also provide
a 3D map. Attempts of this kind are currently in the design or construction phase.
It must be remembered that the exploration of volcanoes by tomographic
techniques based on cosmic muons requires in any case long acquisition times
(weeks/months) to provide reliable images. It is therefore a technique that can give
information on the evolution of the internal structure of the volcano on this time scale,
to provide, in a complementary way to other volcanological exploration techniques,
useful information to predict in the medium term, possible changes and evolutions
of the volcano activity. At the moment it is possible to see the detail of the internal
structure with spatial resolutions of the order of ten meters, but this parameter is
constantly evolving, based on the improvement of the detection equipment.
The use of the absorption tomographic technique has not been limited to the study
of volcanoes, although these represent one of the most interesting challenges, but
has also been widely extended to the study of smaller structures, such as under-
ground cavities, caves, mines, tunnels, as well as other constructions of archaeolog-
ical interest [Caffau1997, Menchaca2011, Gomez2016, Saracino2017, Fuji2017]. In
many of these typical applications, the thickness of the rock that muons have to cross
can be a few tens of meters, which leads to a much smaller reduction in flux than the
typical crossing of the top of a volcano (hundreds of meters), with therefore shorter
measurement times.
16.8 Muon Tomography and Scattering from High-Z

Materials
Scattering muon tomography is based on the multiple scattering process that muons,
as charged particles, can undergo when passing through a material. The value of the
scattering angle, as we saw in a previous chapter, depends—with the same muon
energy—on the characteristics and thickness of the material crossed, in particular
on its atomic number. It is therefore understandable, at least in principle, that by
measuring the deviation that muons undergo when crossing a sample of material,
information on the main features of the sample can be obtained. However, the transi-
tion from a statement of principle to an established technique, capable of being used
under realistic conditions, required an enormous amount of work, and is still in a
development phase.
The first quantitative investigation using this technique to obtain a muog-
raphy of the hidden content inside a given volume dates back to the early 2000s
[Borozdin2003], using sets of drift chambers to reconstruct the trajectories of cosmic
muons above and below the volume to be examined, and therefore measure, in each
event, the deflection of the muon track. A schematic illustration of the technique is
shown in Fig. 16.6.
Position-sensitive tracking detectors are arranged above and below the volume to
be inspected, reconstructing two independent tracks that must be compared in each
event. Muons that pass through any material with a high atomic number, located
within the volume between the detectors, will be deflected, providing information
on the amount of scattering and the 3D position in which the scattering process took
place.
This technique has proved very promising since its inception, for several reasons.
First of all, as it is based on the flux of cosmic muons existing in nature, it does not
require artificially produced radiations, such as X-rays, usually used for scanning
small objects; furthermore, muons have a much greater penetrating power than X-
rays, and the technique can therefore also be used for volumes and objects larger than
those that can be explored with X-rays. Finally, as it is based on scattering and not
on absorption, each muon passing through the entire volume can contribute to the
overall information, that is, to the reconstruction of the tomographic image. Since
the scattering process depends on the atomic number Z of the element crossed, it
is possible to some extent to discriminate between the presence of elements with
low Z, intermediate Z and high Z, i.e., recognize the presence of elements with
high Z even in the presence of a background produced by scattering due to low or
intermediate Z materials. Even if the technique cannot distinguish between elements
with similar atomic numbers, such as uranium and lead, this is rarely a problem,
given the reasons with which this technique was proposed and the different fields of
application in which it proved useful.
It should be remembered, in fact, that one of the problems that gave rise to the use of
this technique is linked to the possibility that fissile material (Uranium or Plutonium)
can be transported illegally inside containers, mixed with other commercial goods,
16.8 Muon Tomography and Scattering from High-Z Materials 289
Fig. 16.6 Sketch of muon

tomography based on muon
scattering. The track of a
muon, reconstructed before
and after the scattering
process, provides
information on the deflection
of the track and on the
localization of the scattering
process. The statistical
distribution of a high number
of muons can provide a 3D
tomography of the interior of
a volume, in principle
detecting the presence of
materials with a high atomic
number
typically made up of lightweight materials. Systematic checking of each container

would take such a long time to be incompatible with the existing flow of goods.
For example, it is known that the number of containers used annually in the world
for the transport of goods is around 200 million, and of these only a small fraction
is controlled. Since the advent of containers as a standard means of transport on a
large scale, the need for inspection and control techniques that were fast and reliable
has been very strong. Muon tomography based on scattering has been suggested
as a viable way to achieve this aim, as it is capable of detecting in a reasonable
time the possible presence of small samples with a high atomic number (even cubic
decimetres) inside the entire container, that is, in volumes of about 50 m3 .
The use of this technique involves, as one can imagine from the previous figure,
the use of two muon tracking detectors with good performance, to be placed above
and below the volume to be inspected, in order to reconstruct the track of each
individual muon before and after the scattering process. In the simplest case, an
algorithm, called POCA (Point Of Closest Approach) can be used to determine the
3D coordinates of the scattering centre, as the midpoint of the segment of minimum
distance between the two skewed lines in space that represent the tracks of the muon
before and after scattering:
PPOCA = (a + b)/2
where a and b represent the vectors of the tracks reconstructed in the upper and lower
part, as indicated in the previous figure.
Considering a large number of individually tracked muons, which come from any
direction compatible with the geometrical acceptance of the detectors, it is possible
to reconstruct two- and three-dimensional density maps of the inspected volume. The
performance of such a tomograph can be evaluated in terms of its spatial and angular
resolution (which in turn depends on the characteristics of the tracking detectors),
of the overall detection efficiency, which allows to provide a response in a short
scan time, and the percentage of false positives, i.e., false reporting of potentially
dangerous areas within the volume.
In recent years, especially in the last decade, numerous prototypes of detectors for
muon tomography, specifically designed for container inspection have been designed
and built, both in small and full-scale dimensions [Schultz2004, Pesente2009,
Riggi2010, Gnanvo2011, Baesso2014, Mahon2013, Xing-Ming2014, Anghel2015,
Keizer2018, Riggi2018, Riggi2021b], making use of different detection technolo-
gies, from gas detectors to plastic scintillator strips with reading systems based on
silicon photomultipliers and WLS fibers. As an example, Fig. 16.7 shows a tomo-
graphic result obtained from a full-scale prototype measuring 7 m × 3 m × 3 m,
therefore capable of inspecting a volume equal to that of an entire 20, container
[Riggi2021b]. The figure shows the reconstructed XY density map, for a given Z
(vertical axis) section; the horizontal dimensions of this map are 300 × 300 cm,
corresponding to half of the overall volume. The other half of the volume, empty,
was taken in this case as a reference to evaluate the background, with respect to which
the excess density was extracted (difference between the left and right sides of the
volume). To this extent, a lead block of 4 dm3 size was placed within one of the two
half volumes between the two detectors. The algorithm used for the data analysis was
based on the POCA method, as previously described, i.e., reporting, for each event,
the coordinates (X, Y, Z) of the POCA minimum approach point between the upper
and lower tracks of the muon, with a cut on the minimum value of the scattering
angle.
In the approach called POCA, a single scattering centre is assumed. However,
there are more sophisticated algorithms [Schultz2007, Bandieramonte2015], with
better performances, to determine the distribution of scattering centres and provide
density maps of the explored volume.
Scattering-based muon tomography has also been used in recent years in numerous
other applications, other than being a technique for inspecting containers. In partic-
ular, it has been suggested that the material produced during the operation of
nuclear power plants and subsequently stored in sealed containers can be inspected
using cosmic muons, without the need to open these containers [Jonkmans2013,
Clarkson2014, Ambrosino2015b, Clarkson2015, Chatzidakis2016, Poulson2017].
Further applications of the same technique also refer to the integrity control of nuclear
reactors, especially in the event of possible accidents [Borozdin2012, Perry2013,
Fuji2013].
16.9 Imaging Techniques Based on Secondary Particle Production 291
Fig. 16.7 A density map obtained by placing a 4 dm3 lead block inside a volume equal to that of a
standard 20, container, obtained using a prototype portal for muon tomography [Riggi2021b]. The
red spot marks the presence of the lead block
Another possible application is the control of any “orphan” radioactive sources

that may be dispersed in the midst of scrap metal sent for their melting to steel mills
[Bonomi2014, Furlan2014]. Incidents of this kind are more frequent than one might
think, and can create serious contamination on the production line, so they must be
diagnosed before the scrap metal is melted.
16.9 Imaging Techniques Based on Secondary Particle

Production
While the tomographic techniques described above make use of the absorption
or scattering process of muons, it has recently been suggested that also the inter-
action of muons in matter with consequent production of secondary (gamma,
electrons) can give information on the characteristics of the material crossed
[Reidy1978, Terada2015, Knezevic2019]. This information can be related to the type
of secondary products, their energy distribution and their abundance. Muons, as we
have seen, interact with matter through ionization processes, e+ e− pair production,
bremsstrahlung and nuclear interaction. The relative importance of these processes
depends on the energy of the muons. At energies typical of muons at sea level
(some GeV) the nuclear interaction is irrelevant, and the predominant process is
that of ionization, while for high energies the effects of production of pairs and
bremsstrahlung start to become important, depending on the atomic number of the
material present in the sample.
Fig. 16.8 Cosmic muons,

tracked by a
position-sensitive detector,
can generate secondary
products (gamma, electrons)
in a material, which are
detected by a set of
scintillators placed around
the sample, and operating in
coincidence with the
tracking detector
An apparatus that in principle can implement this method consists of muon

tracking detectors, which define their trajectory, and a high-resolution Germanium
gamma detector [Bikit2016] or a set of scintillators placed around the sample to be
examined, so as to coincidentally detect the secondary products emitted following
the interaction of the muon [Mrdja2016, Galgoczi2020], as schematized in Fig. 16.8.
However, the potential of this method and its field of application still need to be
systematically explored.
16.10 Monitoring of the Stability of Buildings by Tracking

Cosmic Muons
Among the other possible applications that in recent years have attracted the interest
of some research groups, it is certainly worth mentioning the possibility of moni-
toring, through a precision tracking of muons, possible deformations on a long time
scale of buildings and civil structures.
In this case, it is a question of using two or more detectors, at least one of which is
large and with good tracking capabilities, i.e., capable of reconstructing the trajectory
of the detected muons, installing them in appropriate positions fixed to the structure,
in order to detect any shifts in time of one with respect to the other. Figure 16.9 shows
the diagram of a possible location, in the case of a building.
One of the detectors is placed on the ground or on the basement of the structure,
while the other, even small and without any tracking capability, is located at a certain
16.10 Monitoring of the Stability of Buildings by Tracking Cosmic Muons 293
Fig. 16.9 Two detectors, at least one of which with muon tracking capabilities, can be located in
different positions of a building, to evaluate long time scale sub-millimetre relative deformations
of the various parts of the structure
distance, anchored to that part of the structure whose possible displacement or defor-
mation over time is to be explored. The two detectors must operate in coincidence,
selecting those muons capable of passing through both detectors, even if separated by
floors, walls, or other solid structures. A continuous measurement over time allows
to extract, for example, the angular distribution of the muons detected in coincidence
by the two detectors, discriminating them from those coincidence events originating
from two independent muons, belonging to the same extensive atmospheric shower.
These, in fact, may constitute a continuous background in the angular distribution,
while the muons passing through both detectors will produce a well-defined peak
in the angular distribution. Any displacement or deformation of the structure, on
a long time scale (weeks or months) will produce a slight variation in the angular
distribution and the displacement of the centroid of this peak will give information
on the amount of deformation. In the case of very large distances between the two
detectors, which would make it impractical to pass the signals from each detector to
the coincidence circuit, it is also possible that each of the two detectors collect the
data independently, labelling the arrival of each event using the GPS information and
subsequently correlating off-line the collected data to search for coincidence events.
The use of this technique, for which some prototype measurements made in
recent years are available [Bodini2007, Donzella2014, Bonomi2019, Abbrescia2019,
Pinto2020b], is particularly useful compared to other techniques for monitoring the
stability of a building when the parts of the structure whose stability is to be moni-
tored are not in visual contact, in which case it would be possible to use laser-based
alignment systems or mechanical deformation monitoring systems. The precision
of the technique based on the reconstruction of cosmic muons depends on various
factors, including the spatial or angular resolution of the detectors used, the relative
distance between the detectors, the effects of multiple scattering in the crossing of
the materials interposed between the detectors and the time of overall measurement.
Given the precision that must be achieved, and the long measurement times, it is
important that the detection systems used are as stable as possible over time. It has
been estimated that with measurement times of the order of the month and adequate
detectors it is possible to achieve a precision such as to be able to highlight milli-
metric deformations of a large structure, which makes this technique suitable for
monitoring civil structures (buildings of historical interest, bridges, viaducts …) on
a long time scale.
16.11 Additional Possible Applications of Muon

Tomography
Further application examples of muon tomography techniques, in its various aspects,

have been proposed in recent years, both using absorption and scattering processes.
For example, tomographic images of water towers were obtained [Bouteille2016,
Jourde2016], as well as the amount of CO2 stored inside cavities [Kudryvtsev2012,
Klinger2015]. Examples of radiographs of industrial equipment obtained with muons
are reported in [Giboy2005, Durham2015]. Recently, the problem of controlling
reinforced concrete structures, in particular the deterioration of the internal metal
part, has also received new attention, based on the possibility of using muon scattering
tomography [Dobrowolska2020, Niederleithinger2021].
On a different side, it is worth mentioning the fact that cosmic ray-based tomog-
raphy has recently been proposed and discussed also in the context of explorations
of other bodies in the solar system [Tanaka2007b, Kedar2012, Kedar2013]. The
use of the cosmic radiation for the purposes of tomographic imaging outside the
Earth’s atmosphere is completely different, as the interaction of primary cosmic
rays produces different effects depending on the atmosphere present (or absent)
in the particular planet or satellite considered and the consistency of the soil. In
the case of celestial bodies totally devoid of atmosphere, such as the Moon, the
primary cosmic rays strike the ground without producing extensive atmospheric
showers. In other situations, such as Venus, the atmosphere is so dense that once
the showers are generated at the top of this atmosphere, the secondary products
are almost completely absorbed by the subsequent layers. The application of these
techniques should therefore be discussed in relation to the particular structure of the
celestial body considered.
Problems of this kind have been discussed above all in relation to the planet Mars,
in which a very thin atmosphere (approximately 1/100 of the Earth’s atmospheric
pressure) exists. Under these conditions, the proportion between protons (or primary
nuclei) and secondary component (pions, muons) is very different from that gener-
ated in the Earth’s atmosphere, and, in particular, depends on the inclination of the
particles. As an example, the flux of vertical muons is much less than that observed
on Earth, while the situation is very different for directions close to the horizontal. In
any case, there is a large overlap between primary protons and secondary products,
16.13 Use of Extensive Air Showers in Time Synchronization 295
which should be appropriately identified and discriminated. Examples of this kind

have been discussed [Tanaka2007b], also with possible applications to the prospec-
tion of Mount Olympus, the giant among the mountains existing in the solar system,
a structure 25 km high and with a base of about 600 km.
16.12 The Impact of Cosmic Rays on Cloud Formation
It has been hypothesized in recent years that the arrival of cosmic rays in the Earth’s
atmosphere may also have an impact in determining the formation of clouds and
in general in influencing the weather [Kirby2011]. When charged particles interact
with the atmosphere, they can ionize volatile compounds, causing them to condense
into droplets, around which clouds may begin to form. Although these elementary
processes are known, it is not entirely clear whether they can have a really important
impact on the climate, to the point of determining it and inducing climate changes,
in relation to the intensity of cosmic radiation over time.
For an understanding of these effects under controlled conditions, particle beam
experiments at CERN were started a few years ago, such as the Cosmic Leaving
Outdoor Droplets experiment [CLOUD], in which a cloud chamber filled with ultra-
pure air and defined chemical compounds, which are believed to give rise to cloud
formation, such as water vapor, ozone … is bombarded with beams of high-energy
protons, extracted from the SPS accelerator. The gaseous atmosphere of the chamber
is monitored following the bombardment with these ionizing particles, to analyse
its content and evaluate the effect of radiation. The first results [Kirby2011, Lehti-
palo2016], show some evidence that cosmic rays increase the formation of aerosols
in the atmosphere, although the activity is still ongoing and the issue is the subject
of debate, with varied opinions, some of the which deny that the importance of these
mechanisms is competitive with other known cloud formation mechanisms.
16.13 Use of Extensive Air Showers in Time

Synchronization
The problem of synchronizing systems operating in geographically distant locations

is very common in various applications of today’s technology. Usually, the best
possibility of synchronizing the operation of systems that operate remotely is that
which makes use of the GPS information available from the constellation of satellites
orbiting the Earth. Recent developments in GPS technology make it possible to
achieve very high resolutions, of the order of the nanosecond [Valat2020]. It should
be remembered that precisely the synchronization of distant events using the GPS
technique has allowed, as we have mentioned, the management of cosmic ray physics
experiments, making it possible to carry out coincidence measurements between
detectors located at great distances. This is the case of large surface arrays where the
average distance between the detectors is of the order of km, but also of experiments
in which evidence of correlated events on a planetary scale is sought, as is the case
with recent observations of gravitational waves in observatories located on different
continents.
However, the ability to use satellite GPS signals requires that the satellites can
be “seen” by an antenna, and that the area has adequate signal coverage. In many
situations this is not taken for granted, such as in certain mountain locations, or in
underground or submarine environments. The same difficulty in these environments
would also arise in using alternative technologies to GPS, such as radio signals
emitted by repeaters, or high intensity pulsed LED light signals. In cases where the
GPS signal is intermittent, a high-precision local clock can in principle be used,
which maintains synchronization for a certain period. However, the use of high
precision clocks (nanoseconds), such as those based on Cesium atomic clocks, is
enormously expensive if the system is to be replicated in multiple installations.
There are less accurate and cheaper local clock systems (OCXO: Oven-Controlled
Crystal Oscillator), with a limit drift of the order of microseconds per hour, which
could be used to maintain synchronization for a certain period in the absence of a
reliable GPS signal.
It has recently been suggested [Tanaka2022] that the detection of extensive air
showers could be a local synchronization event to be used to further improve the
accuracy of the local clock in the absence of a GPS signal. The muons of the atmo-
spheric showers, in fact, travelling at speeds close to that of light, and following
paths not too far from the primary axis, arrive within small time intervals, of the
order of tens of nanoseconds, as we have seen while discussing the time profile of a
shower. Each station, in addition to the local clock, should include a muon detector,
for example a scintillator, and should be able to communicate with the other stations
according to traditional technologies (Wi-Fi, optical systems …), even if not of high
precision. The simultaneous arrival—within a few tens of nanoseconds—of muons
in distant detectors would provide the synchronization signal to correct local clock
drift. It has been estimated that with an appropriate distribution of the detectors and
sensitive areas of the order of one square meter, synchronization between the various
stations could be maintained within a hundred nanoseconds, with very low false
synchronizations, depending on the amount of random coincidences between the
different detectors. Of course, synchronization is limited to an area corresponding to
the size of an extensive air shower, but it can be replicated by extending the stations,
to cover very large areas, or by exploiting the ability of smartphones to detect muons
via their CCD sensor, to create a virtual network extended to the whole planet Earth.
Appendix A
A Calculation of the Flux at the Top
of the Eiffel Tower Due to Soil Radioactivity
Considering the presence of gamma-emitter radioactive isotopes in the terrestrial soil,

we can try to evaluate, on the basis of the absorption coefficients of the gammas in
the air and in the soil, reported in Chap. 1, what could be the effect of this radioactive
material placed within a certain distance from the vertical corresponding to the centre
of the Eiffel Tower, and a given depth in the subsoil. In assessing the flux due to the
gammas coming from the ground, it must in fact be considered that at a certain
height h (for example h = 300 m for the Eiffel Tower) not only the gammas emitted
vertically but also those coming from the surrounding ground can contribute, within
a large area (Fig. A.1).
How far from the vertical of the Eiffel Tower and to what depth in the soil is the
contribution of gamma radioactivity still significant? We qualitatively expect that
the contribution of the material placed very deep underground will not contribute,
because of the average absorption coefficient in the rocks, estimated to be about
0.1 cm−1 (see Table 1.1). As regards the dependence on the radial distance with
respect to the vertical of the Tower, the contribution of the material close to the
ground surface is not negligible even at large distances, due to the small value of the
absorption coefficient in the air.
For a quantitative estimate we can numerically integrate the contribution due to
the flux produced by the existing material within a given circular annulus between R
and R + dR (where R is the radial distance with respect to the vertical passing through
the centre of the Tower) and within a depth with respect to the surface between p and
p + dp, assuming a uniform distribution of the radiation emitting centres distributed
in the soil.
Figure A.2 shows the contribution that could be expected if all the radioac-
tive material were distributed only on the surface of the ground (rather than also
underground), as a function of the radial distance from the vertical.
As can be seen from this figure, the largest contribution comes from an area
located at a certain distance, a few hundred meters, from the Tower, more than from
the one below the Tower itself, due to the linear growth of the area of the circular
© The Editor(s) (if applicable) and The Author(s), under exclusive license 297
to Springer Nature Switzerland AG 2023
https://doi.org/10.1007/978-3-031-24762-0
298 Appendix A: A Calculation of the Flux at the Top of the Eiffel Tower Due to Soil …
Fig. A.1 Gamma radiation emitted by isotopes distributed on the ground or at moderate depths
underground can reach great distances in the air. The contribution of this radiation present at the
top of the Eiffel Tower, 300 m high, may also come from areas far from the vertical passing through
the centre of the structure
Fig. A.2 Percentage contribution—for each meter of radial distance—to the total flux measured
at the top of the Eiffel Tower (h = 300 m), as a function of the radial distance with respect to
the vertical passing through the centre of the Tower, assuming that the radioactive material is
distributed exclusively at the ground level. For the absorption of gamma in the air, a coefficient
equal to 0.000044 cm−1 was assumed (Table 1.1)
annulus with the radial distance R. The contribution due to the flux originating from
a certain radial distance R is attenuated with
√ 2 a factor exp(−μd)/d2 , where μ is the
absorption coefficient due to air and d = h + R . It can realistically be assumed
2
that at radial distances of about 1 km this contribution becomes negligible. As for the
Appendix A: A Calculation of the Flux at the Top of the Eiffel Tower Due to Soil … 299
Fig. A.3 Percentage contribution to the total flux measured at the top of the Eiffel Tower (h =
300 m), as a function of the depth, integrating from 0 to 1000 m in the radial distance. For the
absorption of gamma in the soil, an absorption coefficient of 0.092 cm−1 has been assumed
contribution due to the soil layers below the surface, we can integrate up to a radial
distance of 1 km and evaluate the percentage contribution coming from the various
layers placed at a certain depth. This result is shown in Fig. A.3.
In this case it can be seen that the main contribution comes from a very thin
layer with respect to the soil, so much so that at depths greater than 30–40 cm the
contribution becomes negligible. However, assuming an integration limit of 1 m for
depth, it is possible to recalculate the contribution to the total flux as a function of
the radial distance, shown in Fig. A.4, to be compared with that shown in Fig.A.2,
obtained by assuming that all the contributions came from the soil surface. The
trend is qualitatively similar, with a maximum slightly shifted towards smaller radial
distances.
It can therefore be said by looking at these results that the effect on the ionization
of the air measurable at a certain height with respect to the ground comes from the
radioactive material distributed in the soil up to a depth of a few tens of centimeters
but over a very large area, up to radial distances comparable to or greater than the
height of the observation point itself, in this case about 1 km against the height of
300 m for the Eiffel Tower.
The overall result of the calculation, integrating both in depth, up to 1 m below
the ground, and in radial distance from the Eiffel Tower, up to 1 km, has already been
reported in Chap. 1 (Fig. 1.2), normalizing the vertical scale to the value obtained
experimentally by Wulf at ground level.
Readers are invited to implement such calculations by themselves, employing a
suitable programming tool and evaluating the precision of the numerical integration
procedure.
300 Appendix A: A Calculation of the Flux at the Top of the Eiffel Tower Due to Soil …
Fig. A.4 Percentage contribution—for each meter of radial distance—to the total flux measured at
the top of the Eiffel Tower (h = 300 m), as a function of the radial distance. For comparison with
Fig. A.2, this time the flux has been integrated along the depth, up to a maximum value of 1 m
Appendix B
The Absorption Coefficient in Water
and the Directionality of Cosmic Rays
In his third article of 1926 [Millikan1926c] Millikan evaluates the absorption coef-
ficient of cosmic rays in water, able to reproduce the measured data, taking into
account for the first time the assumption that they come from all directions (even if
with equal intensity, a hypothesis that will be corrected later). Indicating with I 0 the
intensity of this radiation outside the water, the intensity I measured at a depth H
will have a contribution dI given by
d I = 2π I0 sin θ dθ e−μH sec θ
and the overall intensity will be given by the following integral:
π/2
I = 2π I0 sin θ e−μH sec ϑ dϑ
0
If we denote x = sec θ, the previous integral can be written in the form
∞
1 −μH x
I = 2π I0 e dx
x2
1
Millikan uses a table of integrals of this type, published by Gold, to evaluate

whether a single absorption coefficient μ obtained at the top of the curve is capable
of reproducing the entire trend of the data, but realizes that no single coefficient is
able to provide a result that reproduces the overall trend. Rather, the data indicate
that this coefficient changes with depth, showing that radiations capable of reaching
greater depth are more penetrating. Previous observations, according to Millikan,
had too large uncertainties to point out deviations from a simple exponential law.
The previous integral can also be solved today using simple numerical methods,
once the value of the absorption coefficient has been selected.
https://doi.org/10.1007/978-3-031-24762-0
302 Appendix B: The Absorption Coefficient in Water and the Directionality of Cosmic Rays
Fig. B.1 Comparison between the experimental data obtained by Millikan [Millikan1926c] with
one of the electroscopes used under the waters of lakes Muir and Arrowheads and the absorption
curve calculated taking into account the (uniform) directionality of cosmic rays. The experimental
values are reported as a function of the thickness of equivalent water, assuming the thickness of the
atmosphere is 7 m. The absorption coefficient used is 0.003 cm−1 , which reproduces the first part
of the curve
Figure B.1 shows such a calculation made with the previous equation (blue line),
which takes into account the direction of the cosmic rays and therefore the different
amount of water crossed, depending on the orientation. The curve was obtained
by a numerical calculation of the integral in the previous equation, assuming an
absorption coefficient equal to 0.003 cm−1 , which reproduces well the experimental
values obtained by Millikan under the surface of the water (red symbols in the plot)
to small depths. The curve was normalized to the highest value in the dataset. As
can be seen from the comparison, the same curve does not reproduce the absorption
of cosmic rays at greater depths, suggesting that the absorption coefficient is greater
for radiations that have already passed through a certain thickness of water without
being absorbed
Even in this case the reader is invited to implement his/her own numerical method
to solve the integrals required in this calculation.
Appendix C
Geographic and Geomagnetic Latitude
Usually, the position of a place on Earth can be defined by its geographic coordinates
(latitude and longitude). Latitude, as it is known, is the coordinate that expresses how
much the position of the considered point is North (N) or South (S) of the equator,
and therefore is measured from an angle that can take values between 0 and 90° N for
the Northern hemisphere and between 0 and 90° S for the Southern hemisphere. The
geographical longitude instead expresses the East (E) or West (W) coordinate with
respect to the reference meridian, the one passing through Greenwich. The longitude
value will therefore assume values between 0 and 180° E or between 0 and 180° W. In
this context, the geographic North Pole and the geographic South Pole represent the
imaginary points, of the northern and southern hemisphere respectively, in which the
Earth’s rotation axis intersects the surface. As a first approximation, it was believed,
until the 19th century, that the position of the geographic poles was immutable.
However, around the end of the 1800s, accurate astronomical measurements (by the
American astronomer Chandler) made it possible to establish that the geographic
North Pole moves, describing a small circumference, around an “average” position,
with a period 435 days. This is due to the non-perfectly spherical shape of the Earth
and the distribution of its mass. The effect, however, is small enough, of the order of
ten meters, to have significant repercussions in determining the geographical position.
Another aspect, much more relevant, is the fact that the behaviour of the Earth as a
magnet is characterized by magnetic poles that do not coincide with the geographical
ones. As is well known, it was around 1600, with Gilbert, that the idea emerged that
the Earth was itself a huge magnet, with its magnetic poles and its lines of force
(Fig. C.1).
The North magnetic pole is the point where the Earth’s magnetic field points
exactly downward. Assuming that the Earth’s magnetic field can be described in
simplified form as that of an ideal magnet, we can define a further point, called the
geomagnetic north pole, which represents the pole of this simplified dipole model
of the Earth’s complex magnetic field. The position of the North magnetic pole and
the North geomagnetic pole are shown, as an example, as of 2017, with respect to
https://doi.org/10.1007/978-3-031-24762-0
304 Appendix C: Geographic and Geomagnetic Latitude
Fig. C.1 Schematic of the Earth as a huge magnet, with the North and South magnetic poles and
the field lines of force
Fig. C.2 Positions of the magnetic pole and the north geomagnetic pole in 2017. Source World
Data Center for Geomagnetism, Kyoto [WDC]
Appendix C: Geographic and Geomagnetic Latitude 305
the geographic North pole position, in Fig. C.2. As can be seen, these locations are
very far from the geographic North Pole.
The position of the North magnetic pole shifts over time, even reaching geographic
positions very far from the geographic North pole. For example, around 2001 its
position was located at about 81° N 110° W, while over the course of 20 years, in
2021, it progressively moved around 86° N 157° W, with a rate of about 55 km per
year. Fig. C.3 shows the “virtual path” of the position of the magnetic North pole
over time, and, in more detail, in the last decades.
Moreover, it should be remembered that during the Earth’s existence, the
polarity of the Earth’s magnetic field has reversed several times, on average every
450,000 years for the last 83 million years. The entire process of inversion of the
magnetic field is a phenomenon that seems to have occurred over a period of time
between 2000 and 12,000 years [Glatzmaier2015].
Finally, the magnetic North Pole and the magnetic South Pole are not exactly
opposite, since the Earth’s magnetic field is not exactly symmetrical. This means that
the line passing through the two magnetic poles, North and South, does not cross the
Earth through its centre. In recent decades, numerous highly detailed measurements
of the Earth’s magnetic field and its variations over time have been made.
On the basis of what has been outlined, one can then imagine a system of geomag-
netic coordinates, similar to that of geographic coordinates, in which the poles are
represented by the geomagnetic poles, and latitude and longitude are replaced by
geomagnetic latitude and longitude. A map in geomagnetic coordinates, evaluated
in 2020, is shown in Fig. C.4 [NOAA]. The reader may follow the evolution of these
displacement in the next years.
306 Appendix C: Geographic and Geomagnetic Latitude
Fig. C.3 North magnetic pole shift from 1600 onwards, particularly in recent decades, where
more frequent and accurate measurements have been obtained. Source Wikimedia Commons. The
observed positions of the poles are extracted from Newitt et al., Location of the North Magnetic
Pole in April 2007, Earth Planets Space 61 (2009) 703. Model predictions are extracted from the
National Geophysical Data Center. The geomagnetic pole path was created using GMT, CC BY
4.0, https://commons.wikimedia.org/w/index.php?curid=46888403
Appendix C: Geographic and Geomagnetic Latitude 307
Fig. C.4 Map in geomagnetic coordinates, showing the position of the geomagnetic poles in 2020
and the coordinate system with geomagnetic latitude (in blue) and longitude (in red). The lines corre-
sponding to the geomagnetic equator and the geomagnetic meridian of reference are represented in
green. Source National Center for Environmental Information (National Oceanic and Atmospheric
Administration, NOAA) [NOAA]
Appendix D
The Magnetic Rigidity of Particles
A magnetic field exerts on a charged particle with a charge q, in motion, a force,

perpendicular to the direction of its velocity v and of the magnetic field B (assumed
to be uniform within a certain region), expressed by the Lorentz force, by means of
the vector product:
F = qv × B
If the direction of the velocity is perpendicular to that of the magnetic field, the
action of the force produces a change in the direction of the velocity without altering
its modulus, resulting in a circular trajectory of radius R, which can be evaluated
from the relationship
qvB = m v2 /R
The radius will then be (Fig. D.1)
R = m v/qB
While the previous relationship holds under the approximation of small speeds
with respect to the speed of light, a similar relationship, correct from a relativistic
point of view, is the one that links the radius of the trajectory to the relativistic
momentum p:
R = p/q B
https://doi.org/10.1007/978-3-031-24762-0
310 Appendix D: The Magnetic Rigidity of Particles
Fig. D.1 A charged particle moving in a uniform magnetic field, with a direction perpendicular
to its speed (represented by the yellow points in the plane of the figure), is subject to the Lorentz
force, following a circular trajectory
Fig. D.2 Magnetic rigidity of protons and electrons as a function of their kinetic energy, from about
1 MeV to 10 GeV
The magnetic rigidity of a particle is then given by the product B R between the
intensity of the magnetic field and the radius of curvature of its trajectory:
B R = p/q
where p is the particle momentum.

Appendix D: The Magnetic Rigidity of Particles 311
Considering the relationship between momentum and total energy, E2 = p2 c2 +

(mc2 )2 , we can evaluate the magnetic rigidity of a particle as a function of its kinetic
energy K = E − mc2 , for specific categories of particles.
Figure D.2 shows as an example the magnetic rigidity curve, expressed in (gauss
cm), for electrons and protons of various energies, from about 1 MeV to 10 GeV.
For a given magnetic field, the p/q ratio is a measure of the angular deflection that
particle will be subject to as it moves through that magnetic field. The greater the
momentum of the particle, the less it will be curved.
If SI units are used, the product (BR) can be written in [Tesla × m] and the
following relation holds
(BR) [Tesla × m] = pc (MeV)/300
For example, a 6 GeV electron will have a value (BR) = 6000/300 = 20 [Tesla ×
m], and moving through a 2 T field, it will follow a circular path with radius R =
10 m.
The concept of magnetic rigidity is often used in evaluating the deflection of
charged particles using suitable magnets in the context of particle accelerators, but
it can also be used to understand the confinement properties of a charged particle of
the cosmic radiation moving in the existing magnetic field in a large region of space,
such as the interplanetary region of the Solar System, or the region corresponding
to the size of the Milky Way. As an example, the magnetic field existing within the
Solar System has an average value of about 6 × 10–9 T = 6 × 10–5 gauss = 60 μG.
On the other hand, typical values of the galactic magnetic field are of the order of
a few μG in the regions around the Sun, to increase up to a few tens of μG in the
innermost regions of the Milky Way.
To make a numerical estimate, we could consider protons of 3 GeV, which move
in the interplanetary magnetic field: in this case (BR) = 10, therefore R = 1.67 ×
109 m, to be compared with the Sun-Earth distance equal to 1.5 × 1011 m. Protons
of much greater energy could reach values of magnetic rigidity such that they cannot
be confined within the dimensions of the Solar System.
To perform additional calculations, one may consider different particles, such
as muons or pions, and check the validity of non-relativistic approximation at low
energy.
Appendix E
The Energy Loss of Charged Particles
and the Estimate of the Muon Mass
The average energy loss by ionization of a charged particle can be described by the
Bethe and Bloch relation [Leo1987, Knoll2000, Leroy2004, Grupen2008], which
allows to evaluate the specific energy loss, i.e., the energy loss per unit of path in
a material, based on the charge, mass and energy of the incident particle, together
with the properties of the material (atomic number, atomic weight and density), as
discussed in the chapter dedicated to the interaction of charged particles with matter.
The loss of energy is therefore different, in the same material, for the different
particles. As an example, the values of the specific energy loss (Stopping Power),
evaluated in (MeV cm2 /g) in the case of protons interacting with an absorber made
from Carbon, are shown in Fig. E.1 as a function of the proton momentum.
Fig. E.1 Specific energy loss of protons on a 12 C absorber, evaluated by the Bethe and Bloch
relationship and expressed in MeV cm2 /g, as a function of the momentum of protons
https://doi.org/10.1007/978-3-031-24762-0
314 Appendix E: The Energy Loss of Charged Particles and the Estimate of the Muon Mass
Fig. E.2 Specific muon energy loss, normalized to the value obtained at the minimum ionization
(MIP unit), as a function of the muon momentum. The two curves refer to two different hypotheses
about the mass of muons: 100 electronic masses (lower curve, in red) and 200 electronic masses
(upper curve, in blue). The black symbol shows the experimentally obtained value by Street and
Stevenson [Street1937]
It can be seen how the specific energy loss decreases as the incident energy
of the protons increases. At very high energies a minimum in the ionization loss is
reached, beyond which this quantity starts to increase slowly, due to relativistic effects
(Relativistic Rise). Particles with an energy close to that for which there is minimal
ionization are often called Minimum Ionizing Particles (MIP). This dependence of
the ionization density on their kinetic energy (or momentum) allows in principle to
“identify“ the particles, if one is able to measure both the momentum and the specific
energy loss at the same time. However, this technique can only be used for low values
of the incident energy, since at higher energies the curves corresponding to different
particles tend to approach each other making their separation difficult or impossible.
Particles having the same charge, but different masses, therefore ionize in a
different way. If we therefore consider muons, their ionization power depends on
the mass they possess. A measurement of the ionization density for muons having
a given momentum would allow in principle to determine their mass. The initial
measurements of Neddermayer and Anderson could not separate the different curves
corresponding to different hypotheses on the still unknown mass of the muons, as
the detected muons had a very high momentum. Subsequent measurements, made by
Street and Stevenson, optimized for low-energy muons, were able to separate these
curves.
Figure E.2 shows the trend of the specific energy loss, normalized to the minimum
ionization value, in the hypothesis of muons having a mass equal to 100 times the
mass of the electron (lower curve, in red) or 200 times the mass of the electron (upper
Appendix E: The Energy Loss of Charged Particles and the Estimate of the Muon Mass 315
curve, in blue). The experimentally obtained value (black symbol) is closer to the
curve corresponding to 200 electronic masses.
Implementing energy loss calculations, as described by the Bethe-Bloch equation
(Chap. 13), may result in a variety of activities, including the calculation of the energy
loss on different materials and compounds, the evaluation of the particle range and
the different behaviour of the energy loss at small and high incoming energy.
Appendix F
List of High-Altitude Observation Stations
in the Mid-1950s
The following table shows a (non-exhaustive) list of the main observation stations
built at high altitudes, totally or partially used for studies concerning cosmic radiation,
as shown in a Report prepared in the mid-1950s [Korff1952, Korff1982].
Station name Location Altitude Coordinates (Lat/Long)

Observatorio del Infiernillo Santiago, Chile 4320 m 33°.10, S/70°17, W
Observation Station Mendoza, Argentina 4000 m 33° N/69° W
Montezuma Solar Station Calama, Chile 9000 ft 22°40, S/68° 56, W
Laboratorio Fisica Cosmica Chacaltaya, La Paz, 5200 m 16° 19, S/68° 10, W
Bolivia
Instituto geofisico Huancayo, Peru 3350 m 12°02, S/75° 20, W
Morococha Junin, Peru 4540 m 11° 37, S/76°08, W
Timboroa Station Timboroa, Kenya 2880 m 0° 0, /35° 31, E
Kodaikanal Observatory Kodaikanal, India 2343 m 10° 13, N/77° 29, E
Teoloyucan Messico 2300 m 19° S/99° W
Mauna Loa Geophysical Hawai 13,453 ft 19° 26, N/155° 36, W
Observatory
Kole Kole Station Hawai 3067 m 20.8° N/156.3° W
Sacramento Peak New Mexico, USA 9200 ft 32°47, N/105° 49, W
Gulmarg Research Gulgmarg, India 9000 ft 34° 03, N/74° 24, E
Observatory
Capillo Peak Observatory Albuquerque, New 9200 ft 34° 41, N/106° 24, W
Mexico
Mount Fuji Weather Station Shizuoka, Japan 3772 m 35° 21, N/138°44, E
Cactus Peak Inyokern, California, 5415 ft 36° 04, N/117° 48, W
USA
Mt. Norikura Cosmic ray Neugawa-Mura, Japan 2840 m 36° 06, N/137° 33, W
Station
(continued)
https://doi.org/10.1007/978-3-031-24762-0
318 Appendix F: List of High-Altitude Observation Stations in the Mid-1950s
(continued)
Station name Location Altitude Coordinates (Lat/Long)
Lake Sabrina Bishop, USA 2765 ft 37.5° N/118° W
White Mountain Research Big Pine, USA 10,640 ft 37° 30, N/118° 10, W
Station
Tioga Pass Station Yosemite, USA 9941 ft 37° 55, N/119° 15, W
High Altitude Harvard Climax, USA 11,190 ft 39° 23, N/106° 12, W
Observatory
Mt. Evans Laboratory USA 14,150 ft 30° 35, N/105° 38, W
Echo Lake Laboratory USA 10,700 ft 39° 39, N/105° 35, W
Observatoire du Pic du Bagneres de Bigorre, 2857 m 42° 56, N/0° 8, E
Midi France
Laboratoire de l’Aiguille Chamonix, France 3600 m 43° N/7° E
du Midi
Puy de Dome Clermont Ferrand, 1450 m 45° 47, N/2° 58, E
France
Observatoire Vallot Chamonix, France 4353 m 45° 50, N/0° 27, E
Laboratorio Testa Grigia Val D’Aosta, Italy 3480 m 45° 56, N/7° 42, E
Jungfraujoch High Altitude Jungfraujoch, 3457 m 46° 22, N/7° 59, E
Station Switzerland
Laboratorio della Canazei, Italy 2030 m 46° 28, N/11° 53, E
Marmolada
Astrophysikalische Switzerland 2050 m 46° 40, N/0° 38, E
Observatorium
Sonnblick Observatorium Salzburg, Austria 3106 m 47° 03, N/12° 57, E
Observatorium Zugspitze Germany 2960 m 47° 25, N/10° 59, E
Hafelekar Sud Tirol, Austria 2297 m 47° 19, N/11° 23, E
Mt.Wrangell Observatory Alaska 14,000 ft 62° N/144° W
Estacion de altura Peron Mendoza, Argentina 3852 m 34° 10, S/69° 40, W
Observatorio Eva Peron Mendoza, Argentina 3100 m 32° 49, S/69° 10, W
Appendix G
An Estimate of the Particle Density
in an Extensive Air Shower
The average density of particles present in an extensive air shower (number of parti-
cles per square meter) depends on the overall energy of the shower and varies over
the region of the shower explored (whether it is close to the axis of the shower or at
a large distance from it).
A numerical estimate of the average density of particles can be obtained by making
use of a number of counters, of area S, placed at a certain distance, for example at the
vertices of a rectangle (Fig. G.1, left) or in a triangular configuration (Fig. G.1, right),
as reported in one of the papers by Cocconi and collaborators [Cocconi1944a].
The reasoning is based on the calculation of probabilities. If p(S, Δ) represents
the probability that an area counter S will not be hit by a shower having an average
particle density of Δ, the probability that it will be hit will be (1 − p). An increase dS
of this surface leads to a decrease in the probability p, equal to dp = − pSΔ. From
this we then obtain p = e−S Δ .
Fig. G.1 Geometric configurations used by Cocconi and collaborators [Cocconi1944a] for the
experimental estimation of the average density of particles in an extensive air shower
https://doi.org/10.1007/978-3-031-24762-0
320 Appendix G: An Estimate of the Particle Density in an Extensive Air Shower
The probability that a single counter is hit will therefore be (1 − p) = 1 − e−SΔ ,

while the probability that n counters are hit simultaneously will be given by
(1 − p)n = [1 − e−SΔ ]n
If we measure the coincidences Cn and Cn−1 between n counters and (n − 1)

counters, the ratio between these quantities will be equal to the probability ratio, i.e.
Cn /Cn−1 = [1 − e−SΔ ]n /[1 − e−SΔ ]n−1 = 1 − e−SΔ
hence
e−SΔ = 1 − Cn /Cn−1
and therefore, the average density Δ can be obtained:
Δ = −(1/S) ln(1 − Cn /Cn−1 )

Appendix H
The Relationship Between Altitude
and Atmospheric Depth
The Earth’s atmosphere plays an important role in understanding the phenomena

related to the study of the cosmic radiation. As a matter of fact, all possible interac-
tions of the primary particles take place in it, as well as the generation, propagation
and decay of the secondary ones.
The Earth’s atmosphere is a complex system, the conditions of which vary over
time. However, we can outline some of the basic properties of this system. We
can consider the atmosphere as a large volume of gas, which at sea level has a
density of about 1019 particles per cm3 . We know that the density, and therefore the
average number of particles per unit of volume, decreases with increasing altitude.
The composition of the atmosphere includes several gases, among which the most
abundant are Nitrogen (N2 ), with a percentage of 78.1%, Oxygen (O2 ) for 20.9%
and Ar, with a percentage of 0.9%. Small amounts of other substances, such as CO2 ,
make up the remaining 0.1%.
The atmospheric depth crossed is measured starting from the zone in which the
density is negligible, and gradually increases towards the sea level, corresponding to
a depth of about 1000 g/cm2 . It therefore represents the thickness of the atmosphere,
expressed in surface density units.
In a simplified schematization, we can assume an isothermal static model of the
atmosphere, according to which the density varies almost exponentially with altitude,
according to a trend of the type
n(h) = n(h 0 )e−h/ h s
where n(h) and n(h0 ) represent the densities at heights h and h0 respectively, and hS
is a parameter that for a perfect exponential trend should be constant, expressed in
cm from
h s = KT/Mg
https://doi.org/10.1007/978-3-031-24762-0
322 Appendix H: The Relationship Between Altitude and Atmospheric Depth
Fig. H.1 Atmospheric depth crossed (in g/cm2 ) as a function of the altitude with respect to sea
level (in km), according to a simplified atmosphere model
where K is the Boltzmann constant, T the absolute temperature (in Kelvin), M the
molecular weight (in g/mol) and g the Earth’s acceleration of gravity (in cm−1 s−2 ).
A quantitative representation of this ideal trend is shown in Fig. H.1.
In a model of a real atmosphere, hS varies slightly with the atmospheric depth,
from about 7 × 105 cm at the atmospheric depth of 10 g/cm2 (corresponding to an
altitude of about 30 km) to a value of about 8.2 × 105 cm at an atmospheric depth
of 900 g/cm2 , corresponding to about 1500 m of altitude. Quantitative graphs of the
relationship between altitude and atmospheric depth in the case of a real model of
the atmosphere are shown in [Grieder2001].
Appendix I
Gaisser-Hillas Parameterization
of the Longitudinal Profile of a Shower
The longitudinal profile of an extensive air shower (density of particles at various

atmospheric depths X) can be expressed according to different parameterizations.
One of the widely used functions was introduced by Gaisser and Hillas [Gaisser1977]:
X max
λ
X
X max
where X represents the atmospheric depth (expressed in g/cm2 ), Nmax is the number
of particles corresponding to the maximum Xmax , while λ represents a characteristic
depth (usually assumed to be 70 g/cm2 ). Figure I.1 shows this function, for a value
of λ assumed exactly equal to 70 g/cm2 , and for two different values of Xmax (200
and 400 g/cm2 ).
The next figure shows the effect of the λ parameter on the distribution, evaluated
with the same value of Xmax (400 g/cm2 ) and for two different values of λ (50 and
90 g/cm2 ).
The effect of these two parameters therefore determines the position of the
maximum in the profile of N(X) and the width of this distribution.
In some cases, it is possible to introduce an additional parameter X0 , to take into
account the fact that the first interaction does not occur at X = 0 [Pryke2001]. With
this modification, the previous equation can be written as:
X maxλ−X 0
X − X0
X max − X 0
The reader may use the previous equations to perform additional calculations as
a function of the two parameters (Xmax and λ).
https://doi.org/10.1007/978-3-031-24762-0
324 Appendix I: Gaisser-Hillas Parameterization of the Longitudinal Profile of a Shower
Fig. I.1 The Gaisser-Hillas distribution that parameterizes the behaviour of the longitudinal devel-
opment of a shower along the atmosphere, evaluated for a value of λ equal to 70 g/cm2 and for two
different values of the parameter Xmax (200 and 400 g/cm2 )
Fig. I.2 Gaisser-Hillas distribution which parameterizes the behaviour of the longitudinal devel-
opment of a shower along the atmosphere, evaluated for a value of Xmax equal to 400 g/cm2 and
for two different values of λ (50 and 90 g/cm2 )
Appendix J
The Thickness of Air Crossed by a Particle
in the Atmosphere
When a particle travels in the atmosphere, the effective thickness crossed depends on
the inclination of the track with respect to the vertical. For particles that move verti-
cally (perpendicular to the ground), the thickness crossed, in terms of surface density
(g/cm2 ) is given by the relationship between altitude and thickness, as described
in Appendix H. For inclined tracks, the thickness actually crossed is larger. If we
consider the approximation of a ’flat“ Earth, which is a good approximation for not
too large inclination angles (typically up to about 60 degrees), the actual thickness
(slant depth) is given by
X
X s = X sec ϑ =
cos ϑ
When the angle of inclination with respect to the vertical becomes very large, the
curvature of the Earth’s surface cannot be ignored, and the previous approximation
Fig. J.1 Effective thickness traversed by a particle moving from a height X, with an inclination
angle θ with respect to the vertical, in the approximation of a ’flat“ Earth, valid for not too high
inclination angles
https://doi.org/10.1007/978-3-031-24762-0
326 Appendix J: The Thickness of Air Crossed by a Particle in the Atmosphere
Fig. J.2 Effective thickness traversed by a particle moving from a height X, with an angle of
inclination θ with respect to the vertical, in the approximation of a ’curved“ Earth, based on a
parameterization of the Chapman function [Swider1967]
is no longer valid. This happens for example for the detection of almost horizontal
muons, or for detectors placed at very high altitudes. Taking into account the entire
thickness of the atmosphere crossed by particles moving horizontally, this is about
40 times larger than that crossed vertically.
In this case, a geometrical analysis of the situation leads to an expression of the
true crossed thickness which can be expressed through the Chapman functions, or
through approximations to these functions [Swider1967].
Figure J.2 shows the atmospheric depth crossed as a function of the angle of
inclination with respect to the vertical. While for a vertically moving particle, the
atmospheric depth is just over 1000 g/cm2 , it becomes about 10 times greater for
angles close to 85°
The interested reader may further investigate the use of the Chapman functions
or their approximate behaviour through a series of numerical calculations.
Appendix K
Evaluation of the Shower Direction
from the Relative Timing of Several Detectors
With three detectors placed horizontally and not aligned, it is possible, by measuring
the arrival time of the shower particles in each of the detectors, to determine with a
certain approximation the axis of the shower, and therefore the incoming direction of
the primary particle, using a technique of triangulation. With reference to Fig. K.1,
we can imagine having 3 detectors (0, 1 and 2), placed horizontally on a plane, in
the positions (x0 , y0 ), (x1 , y1 ), (x2 , y2 ), each of which detects one of the particles of
the shower, at the time T0 , T1 , T2 .
Assuming as a first approximation that the different detected particles travel at
the speed of light c, and that they move parallel, geometric considerations allow us
Fig. K.1 Use of 3 detectors arranged on a horizontal plane (not aligned) to measure the differences
in the arrival times of the particles in each of them and determine, based on geometric triangulation
considerations, the direction of the shower axis
https://doi.org/10.1007/978-3-031-24762-0
328 Appendix K: Evaluation of the Shower Direction from the Relative Timing of Several …
to determine the direction (θ, ϕ) of the axis in polar coordinates, using the following
set of equations:
−y0
d2 − d1 yy21 −y
cos ϑ sin ϕ = 0
−y0
(x2 − x0 ) − (x1 − x0 ) yy21 −y0
−x0
d2 − d1 xx21 −x
cos ϑ cos ϕ = 0
−x0
(y2 − y0 ) − (y1 − y0 ) xx21 −x 0
where d1 = c (T1 − T0 ), d2 = c(T2 − T0 ) and the two time-differences (T1 − T0 ),

(T2 − T0 ) are measured experimentally.
Based on the relative distances and the time resolution with which it is possible to
measure the differences in arrival times, a larger or smaller precision will be obtained
in the reconstruction of the angles θ and ϕ. Use of this technique has been also
exploited in educational experiments in cosmic ray physics [Aiola2012, Riggi2014].
In an array of several detectors, the measurement of the time differences between
all the possible pairs of detectors gives rise to an oversized set of equations, which
can be solved with minimization techniques to determine the optimal solution. For
example, referring to the work of Clark et al. [Clark1961], which used such a proce-
dure for managing the data obtained with an array of detectors, the direction of the
shower, defined by the director cosines l, m, was evaluated by minimizing the quantity
1 Σ i=N
χ2 = (cti + lxi + myi − ct0 )2
(N − 3) i=1
where the positions of the N detectors involved in the event to be reconstructed are
named (xi , yi ), i = 1, 2, …, N, the ti are the arrival times at the single detectors, and
t0 is the average time.
To further understand the limit and potential of this method, the reader may
investigate the role of the time resolution on the overall precision of the method.
Appendix L
Parameterizations of the Muon Spectrum at Sea
Level
The energy (or momentum) spectrum of secondary muons at sea level has been
measured in a variety of experiments in the past. In many circumstances it is useful
to have a parameterization of the dependence of the flux of secondary muons on
the energy and on the angle of inclination, in order to use simple relationships that
summarize different sets of measured data, and which can be used in further calcula-
tions, for example to model the expected cosmic ray flux in a simulation. One of the
classic parametrizations used is that due to Gaisser [Gaisser1990], already mentioned
in the text, which provides a dependency of the type:

dN −2.7 1 0.054
= 0.14E cos θ
− 1.1E cos θ
d EdΩ 1 + 1.1E
115 GeV
1 + 850 GeV
where E and θ represent respectively the energy and the angle of inclination of the
muons with respect to the vertical. An example of a spectrum obtained for vertical
muons on the basis of this parameterization has already been reported in Chap. 10.
Figure L.1 shows a comparison between the spectrum of vertical muons (on the left)
and that of muons with an inclination of 70° (on the right).
The Gaisser parameterization is valid for sufficiently high energies and for not too
large angles of inclination. Different parameterizations have been proposed in the
following years, some of which generalize that of Gaisser, so as to make it applicable
even in a range of smaller energies and at larger angles, where the effect of the Earth’s
curvature is not negligible. For example, one of these has been proposed by Guan
et al. [Guan2015]. In this we use the angle θ*, linked to θ by the relation
/
(cos θ )2 + p12 + p2 (cos θ ) p3 + p4 (cos θ ) p5
cos θ ∗ =
1 + p12 + p2 + p4
where the coefficients pi , determined by the best fit of the available experimental
data, have the following values:
https://doi.org/10.1007/978-3-031-24762-0
330 Appendix L: Parameterizations of the Muon Spectrum at Sea Level
Fig. L.1 Spectrum of vertical secondary muons (left) and inclined at 70° (right), multiplied by the
factor E2.7 , based on the parameterization of Gaisser [Gaisser1990]
Fig. L.2 Spectrum of vertical (left) and 70° inclined (right) secondary muons, multiplied by the
factor E2.7 , based on the parameterization of Guan et al. [Guan2015]
p1 p2 p3 p4 p5
0.102573 − 0.068287 0.958633 0.0407253 0.817285
The Gaisser relation is then modified as follows:

−2.7
dN 3.64
= 0.14 E 1 +
d EdΩ E(cos θ ∗ )1.29

1 0.054
× cos θ ∗
− 1.1E cos θ ∗
1 + 1.1E
115GeV
1 + 850GeV
If this parameterization is used, spectra similar to the previous ones are obtained
for high muon energies, but with a very different trend for the low energy part,
below some GeV, which reproduce the available experimental data. As an example,
Figure L.2 shows the corresponding spectra calculated according to this parame-
terization, for the vertical muons (on the left) and for those inclined at 70° (on the
right).
Appendix L: Parameterizations of the Muon Spectrum at Sea Level 331
Fig. L.3 Spectrum of secondary muons at angles of 0° (black symbols), 20° (red symbols), 40°
(blue symbols) and 60° (purple symbols), based on the parameterization of Guan et al. [Guan2015]
Figure L.3 shows the dN/dE spectra (therefore without the multiplicative factor
E2.7 ) for different angles, from 0 to 60 degrees, again using the parameterization of
Guan et al. [Guan2015]
Many other works, which report various theoretical calculations of the muon flux,
their energy spectrum, charge ratio and dependence on orientation or geomagnetic
latitude have been published after 1970. Among these, for example those discussed
by [RamanaMurthy1972, Gaisser1974, Honda1995, Bugaev1998].
As an exercise to the reader, the comparison between different parameterizations
of the muon spectrum may be further investigated, and approximate solution to the
problem in specific momentum or angular ranges may be found.
Appendix M
The Flux of Underground Muons
The relationship between the muon flux and depth is often parameterized on the basis
of semiempirical formulas, which take into account the available data, or a subset of
these data obtained under certain conditions, and use a series of parameters, usually
obtained from a fit of the experimental data. A frequently used relation has been
proposed by Miyake [Miyake1963], which expresses the vertical intensity I(X) as a
function of the depth X as
K
I (X ) = (X + a)−α e−β X
X+H
where the parameters a, K, H, α and β are obtained from a fit of the experimental
data and the depth X is expressed in m.w.e. (or hg/cm2 ). Using the following set of
parameters, the trend shown in Fig. M.1 (blue line) can be obtained.
a K H α β
11 174 400 1.53 8 × 10–4
In this relationship, the depth X (considered from the top of the atmosphere) is
expressed in hg/cm2 (or in m.w.e.), while the intensity is calculated in cm−2 s−1 sr−1 .
Alternatively, a slightly more elaborate formula, with a wider range of application
(10–104 hg/cm2 ) has been reported by Barbouti and Rastin [Barbouti1983]:
e−β X
I (X ) = K
(X + H )(X α + a)
with the parameters shown in the following table:
a K H α β
75 270.7 200 1.68 5.5 × 10–4
https://doi.org/10.1007/978-3-031-24762-0
334 Appendix M: The Flux of Underground Muons
Fig. M.1 Parameterizations of the vertical flux of underground muons as a function of depth,
expressed in hg/cm2 . Blue line: parameterization due to Myake [Myake1963]; red line: parameter-
ization due to Barbouti [Barbouti1983]
The resulting plot from this relationship is shown in the same figure by a red line.
The two parameterizations are quite similar especially in the intermediate region,
while some differences can be observed for smaller depth values.
Appendix N
Detection of Bit-Flip Errors Originated
by Cosmic Rays
The simplest technique to diagnose and possibly correct soft errors in a series of
binary information, such as those that could be introduced by the passage of cosmic
rays, by ionizing radiations, or simply by transmission disturbances in a line, is that
of parity check. This technique was introduced as early as the 1950s to evaluate
the correctness of a series of binary signals marked on a paper or magnetic tape. It
basically involves adding a bit (parity bit) to the other bits present in a given sequence,
making sure that the number of 0 s (or 1 s) present in the sequence is always even
(or odd, depending on the convention adopted). These bit sequences could simply
be those that make up a byte, or longer sequences.
In the case of a sequence of 8 bits (one byte), the most significant bit of the byte
can be used as the parity bit and the other 7 bits to define one of the possible 27 = 128
values (from 0 to 127), for example those used for the ASCII encoding of the various
symbols (letters, digits 0–9, special characters). The parity bit P is set by the system
based on the content of the other 7 bits and the convention adopted. In Fig. N.1, for
Fig. N.1 In a 7-bit sequence, an additional bit (parity bit) can be added to make sure that the number
of bits at 1 is always even or odd, depending on the chosen convention, and thus diagnose if one of
the bits of the sequence has changed its initial state
https://doi.org/10.1007/978-3-031-24762-0
336 Appendix N: Detection of Bit-Flip Errors Originated by Cosmic Rays
example, assuming a convention according to which the number of bits 1 (including

the parity bit) must be even, the 7-bit sequence “1001101” needs a parity bit set to 0
(so that the total number of bits 1 remains four) and therefore the total sequence of
8 bits would be “01001101”. Similarly, the 7-bit sequence “1101101”, in which the
number of bits 1 is five (odd), requires a parity bit set to 1, in order to bring the total
number of bits 1 to six (even). The complete sequence will therefore be “11101101”.
The check upon arrival of a sequence of bits transmitted by a transmitting circuit,
or present in a memory cell, is done by comparing the overall parity (number of
bits 1) with that expected according to the assumed convention. If it is different, it
can be diagnosed that a parity error has occurred, due to the fact that one of the
bits in the sequence has changed state (from 1 to 0 or viceversa). It is important to
note that diagnosing the error does not in itself lead to its correction, as we cannot
know which of the different bits in the sequence has changed its state. The system,
however, can be organized in such a way as to request a new sending of the same
sequence, neglecting the data received previously. Another important remark is that
the use of a parity bit can only diagnose errors in which only a single bit changes its
state, while it cannot detect situations in which two bits both change state, which is
generally much less probable, and which in certain cases can be further minimized
with a suitable geometric arrangement of the cells corresponding to bits forming part
of the same sequence.
Notes to the Bibliography
There is nothing that can replace the reading and study of an author’s original works.
If this is particularly true for literary works, we may say that even for a scientific
discipline a great help to understand the original ideas, the attempts to interpret
the observed phenomenology and the overall path towards the current status of that
discipline certainly derives from the knowledge of the texts in which those ideas and
those attempts were originally described.
In the study of the cosmic radiation, there is a very peculiar atmosphere when
reading the considerations about the first evidence of a cosmic radiation, the first
attempts to interpret—often overcome by subsequent events—the perspectives of
research, right from the written pages by the protagonists of this adventure. For
the most part the results and theoretical considerations in the scientific tradition are
communicated through short articles in specialized journals, through the proceedings
of conferences dedicated to specific topics, through review articles that report the
status of the knowledge on a given problem and summarize what it has been done in
that field, more rarely—only when the status of knowledge in that area has reached
a certain degree of maturity—even from textbooks.
Many of the authors who have made the history of cosmic ray physics have also
ventured into the non-trivial challenge of communicating their personal involvement,
the certainties and doubts they have reached, through works that are worth reading
for many decades after being published. These works, of a general nature, dedicated
to the history and evolution of cosmic ray physics in the first decades after their
discovery, are complementary to the more recently published specialist textbooks
on the subject. Without any completeness, we mention the following, from many of
which the author has benefited enormously:
• Pierre Auger, Rayon cosmiques, University of Paris Press, Paris, 1941.
• T.E. Cranshaw, Cosmic Rays, Clarendon Press, 1963.
• Vitaly L. Ginzbug and S.I. Syrovatskii, The origin of cosmic rays, English ed.
Pergamon Press Inc., New York, 1964.
• A. Michael Hillas, Cosmic Rays, Pergamon Press, Oxford, 1972.
https://doi.org/10.1007/978-3-031-24762-0
338 Notes to the Bibliography
• Lajos Janossy, Cosmic Rays, Clarendon Press, Oxford, 1950.

• Lajos Janossy, Cosmic Rays and Nuclear Physics, Pilot Press Ltd., 1948.
• Louis Leprince-Ringuet, Cosmic Rays, Prentice-Hall, New York, 1950.
• Robert Millikan, Cosmic Rays, three lectures, Cambridge University Press, 1939.
• Donald Joseph X. Montgomery, Cosmic Ray Physics, Princeton University Press,
Princeton, 1949.
• Martin A. Pomerantz, Cosmic rays, Van Nostrand Reinhold, 1971.
• Bruno Rossi, Cosmic Rays, McGraw-Hill, New York, 1964.
• Bruno Rossi, Moments in the life of a scientist, Cambridge University Press, New
York, 1990.
• Bruno Rossi, High Energy particles, Prentice Hall Inc., Englewood Cliffs, New
Jersey, 1959.
• Arne Eld Sandström, Cosmic ray physics, North-Holland, Amsterdam; Inter-
science (Wiley), New York, 1965.
• G.B. Zhdanov, Cosmic Rays, Lawrence and Wishart, 1959.
More recent texts, either introductory or more advanced, among the many
available in the literature are the following:
• Lev I. Dorman, Cosmic Rays: Variations and Space Explorations, American
Elsevier, 1974.
• Lev I. Dorman, Cosmic Rays in the Earth Atmosphere and Underground, Springer,
2004.
• Michael W. Friedlander, Cosmic rays, Harvard University Press, 1989.
• S. Flügge, Cosmic Rays I & II, Springer Nature, 2012.
• T.K. Gaisser, Cosmic Rays and Particle Physics, Cambridge University Press,
1988.
• Peter K.F. Grieder, Cosmic Rays at Earth, Researcher’s Reference Manual and
Data Book, Elsevier Science, Amsterdam, 2001.
• Peter K.F. Grieder, Extensive Air Showers, High-Energy Phenomena and Astro-
physical Aspects. A tutorial, reference manual and data book, Volumes 1 & 2,
Springer, Berlin, Heidelberg, 2010.
• T.K. Gaisser and A.A. Watson, The Highest Energy Cosmic Rays, Cambridge
University Press, 2004.
• D.V. Skobel’tsyn (Ed.), Cosmic rays and nuclear interactions at high energies,
Springer, 2014.
• T.S. Stanev, Cosmic Ray Physics, John Wiley and Sons, 1998.
A guide to the bibliography concerned with general and historical works on
Cosmic Ray Physics, updated in the mid-1960s, was published in the American
Journal of Physics in 1967 (J.R. Winckler and D.J. Hofmann, Resource Letter CR-1
on Cosmic Rays, American Journal of Physics 35 (1967) 2), and contains informa-
tion on the original articles that marked the main milestones in the history of this
discipline in its first 50 years, together with a list of review articles on some aspects
of this discipline and general books on the subject, either informative or by a more
in-depth nature.
Notes to the Bibliography 339
There is also a large number of articles, or textbooks, on the historical and educa-
tional aspects of cosmic ray physics, some of which were published just in the years
that marked the centenary of their discovery. We cite, without any exhaustive aim,
only a few, of interest to what is discussed in this text:
• Luisa Bonolis, International scientific cooperation during the 1930s. Bruno Rossi
and the development of the status of cosmic rays into a branch of physics, Annals
of Science 71 (2014) 355.
• Jonathan F. Ormes (Editor), Centenary Symposium 2012: Discovery of Cosmic
Rays, Denver, June 26–28, 2012, AIP Conference Proceedings 1516 (2013) 1–301,
reporting the proceedings of the Symposium organized in 2012 on the occasion
of the 100th anniversary of the discovery of cosmic rays.
• Johannes Knapp, Michael Walter and Christian Spiering (Editors), Centenary of
cosmic ray discovery, Astroparticle Physics 53 (2014) 1–190, which includes a
collection of papers mainly of historical nature, on the occasion of the 100th
anniversary of the discovery of cosmic rays.
• Mario Bertolotti, Celestial messengers. Cosmic rays, the story of a scientific
adventure, Springer-Verlag, Berlin, Heidelberg, 2013.
• P. Carlson, A Century of Cosmic Rays, Physics Today 65 (2012) 30.
• Roger Clay and Bruce Dawson, Cosmic bullets, High energy particles in
astrophysics, Basic Books, 1999.
• Alessandro De Angelis, L’enigma dei raggi cosmici, Springer, Milano, 2012.
• Lev I. Dorman and Irina V. Dorman, Cosmic Ray History, Nova Science Publisher
Inc., 2014.
• Lev I. Dorman and Irina V. Dorman, How cosmic rays were discovered and why
they received this misnomer, Advances in Space Research 53 (2014) 1388.
• M.W. Friedlander, A Thin Cosmic Rain: Particles from Outer Space, Harvard
University Press, 2002.
• M.W. Friedlander, Physics: A Century of Cosmic Rays, Nature 483 (2012) 400.
• Karl-Heinz Kampert and Alan A.Watson, Extensive air showers and ultra-high-
energy cosmic rays: a historical review, European Physical Journal H 37 (2012)
359.
• L. Marafatto, In search of the cosmic rays, Aracne Editrice, 2016.
• Y. Sekido and H. Elliott (Eds.), Early History of Cosmic Ray Studies, D. Reidel
Publishing Company, Dordrecht, 1985.
• William F.G. Swann, History of cosmic rays, American Journal of Physics 29
(1961) 811.
• M. Walther and A.W. Wolfendale, Early history of cosmic ray physics, European
Physical Journal H 37 (2012) 323.
• A.W. Wolfendale, History of British contributions to the astrophysical aspects of
cosmic rays and gamma rays since the second world war, Quarterly Journal of
the Royal Astronomical Society 29 (1988) 27.
• Q. Xu and M.L. Brown, The early history of cosmic ray research, American
Journal of Physics 55 (1987) 23.
340 Notes to the Bibliography
Among the periodic conferences dedicated to cosmic ray physics, it is certainly

worth mentioning the series of the International Conference on Cosmic Ray Physics
(ICRC), whose proceedings have been digitized and made accessible through the
website: https://adsabs.harvard.edu/Proceedings/proceedings/ICRC.
The specific bibliographic references, cited throughout the text in square brackets,
for example [Aab2017a], with the name of the first (or only) author and the year
of publication, are shown below, following the alphabetical order. Even if it is a
rather long list, it was preferred to group them together at the end of the text, rather
than distributing them within each chapter, which would perhaps have allowed a
more direct consultation, however at the expense of the possibility of comparing
references by the same author but cited in different contexts. In the case of biblio-
graphic indications pointing to Web sites, initials in capital letters have been used, for
example [NIST]. For each reference, the name of the author was generally indicated,
or authors, if limited in number, or the wording “et al.” after the name of the first
author, in the case of a large number of co-authors. In the case of articles published in
scientific Journals, the name of the Journal, the Volume, the initial page and the Year
of publication are shown below, while in the case of books, reference is made to the
Publishing House and the place and year of publication. In a few cases, especially
for historical articles, some of the previous indications may be missing, where it
has not been possible to find all the details. With regard to the accessibility of this
material, it should be remembered that many articles considered to be of historical
interest are available on public servers, even if originally published in journals for
which credentials or subscriptions are required. In recent years, the editorial policy
of Open Access publishing has been introduced in several Journals, offering many
of the relevant scientific articles freely available. Students and researchers should
be able to access most of the bibliographic material through the accounts of the
Institutions they belong to.
Bibliography
[Aab2017a] A. Aab et al. (The Pierre Auger Collaboration), Observation of a large-scale anisotropy
in the arrival directions of cosmic rays above 8 × 1018 eV. Science 357, 1266 (2017)
[Aab2017b] A. Aab et al. (The Pierre Auger Collaboration), Inferences on mass composition and
tests of hadronic interactions from 0.3 to 100 EeV using the water-Cherenkov detectors of the
Pierre Auger Observatory. Phys. Rev. D96, 122003 (2017)
[Aartsen2016] M.G. Aartsen et al., Anisotropy in cosmic-ray arrival directions in the southern
hemisphere based on six years of data from the Icecube detector. Astrophys. J. 826, 220 (2016)
[Abbasi2010] R. Abbasi et al., Measurement of the anisotropy of cosmic-ray arrival directions with
Icecube. Astrophys. J. Lett. 718, L194 (2010)
[Abbrescia2011] M. Abbrescia et al. (The EEE Collaboration), Observation of the February 2011
Forbush decrease by the EEE telescopes. Eur. Phys. J. Plus 126, 61 (2011)
[Abbrescia2013] M. Abbrescia et al. (The EEE Collaboration), Time Correlation measurements
from extensive air showers detected by the EEE telescopes. Eur. Phys. J. Plus 128, 148 (2013)
[Abbrescia2015] M. Abbrescia et al. (The EEE Collaboration), Looking at the sub-TeV sky with
cosmic muons detected in the EEE MRPC telescopes. Eur. Phys. J. Plus 130, 187 (2015)
[Abbrescia2016] M. Abbrescia et al. (The EEE Collaboration), A study of upward going particles
with the Extreme Energy Events telescopes. Nucl. Instrum. Methods Phys. Res. A816, 142 (2016)
[Abbrescia2018] M. Abbrescia et al. (The EEE Collaboration), Search for long distance correlations
between extensive air showers detected by the EEE network. Eur. Phys. J. Plus 133, 34 (2018)
[Abbrescia2019] M. Abbrescia et al. (The EEE Collaboration), The EEE MRPC telescopes as
tracking tools to monitor building stability with cosmic muons. J. Instrum. 14, P06035 (2019)
[Abbrescia2020] M. Abbrescia et al. (The EEE Collaboration), New high precision measurements
of the cosmic charged particlerate beyond the Arctic Circle with the PolarquEEEst experiment.
Eur. Phys. J. C80, 665 (2020)
[Abbrescia2021] M. Abbrescia (The EEE Collaboration), Search for multi-coincidence cosmic ray
events over large distances with the EEE MRPC telescopes. J 4, 838 (2021)
[Abbrescia2022] M. Abbrescia et al. (The EEE Collaboration), Detection of Rayleigh-Lamb waves
generated by the 2022 Hunga-Tonga volcanic eruption with the POLA detectors at Ny-Alesund.
Sci. Rep. 12, 19978 (2022)
[Abbrescia2023] M. Abbrescia et al. (The EEE Collaboration), Measurement of the cosmic charged
particle rate at sea level in the latitude range 35°–82° N. Eur. Phys. J. C (2023)
[Abdo2009] A.A. Abdo et al., The large scale cosmic-ray anisotropy as observed with Milagro.
Astrophys. J. 698, 2121 (2009)
[Abdollahi2013] S. Abdollahi et al., Study of atmospheric muons using a cosmic ray telescope. J.
Phys. G: Nucl. Part. Phys. 40, 025202 (2013)
https://doi.org/10.1007/978-3-031-24762-0
342 Bibliography
[Abeysekara2019] A.U. Abeysekara et al., All-sky measurement of the anisotropy of cosmic rays
at 10 TeV and mapping of the local interstellar magnetic field. Astrophys. J. 871, 96 (2019)
[Abreu2010] P. Abreu et al. (The Pierre Auger Collaboration), Update on the correlation of the
highest energy cosmic rays with nearby extragalactic matter. Astropart. Phys. 34, 314 (2010)
[Aglietta1996] M. Aglietta et al., A measurement of the solar and sidereal cosmic-ray anisotropy
at E~1014 eV. Astrophys. J. 470, 501 (1996)
[Aguilar2016] M. Aguilar et al. A(MS Collaboration), Antiproton flux, antiproton-to-proton ratio,
and properties of elementary particle fluxes in primary cosmic rays measured with the Alpha
Magnetic Spectrometer on the International Space Station. Phys. Rev. Lett. 117, 091103 (2016)
[Aguilar2021] M. Aguilar et al., (The AMS Collaboration), The Alpha Magnetic Spectrometer
(AMS) on the International Space Station: Part II - Results from the First Seven Years. Phys. Rep.
894, 1 (2021)
[Aiola2012] S.Aiola, P.La Rocca, F.Riggi and S.Riggi, Detection of extensive cosmic air showers
by small scintillation detectors with wavelength-shifting fibers. Eur. J. Phys. 33, 1207 (2012)
[Aizu1959] H. Aizu et al., Heavy nuclei in the primary cosmic radiation at Prince Albert, Canada.
I. Carbon, Nitrogen and Oxygen. Phys. Rev. 116, 436 (1959)
[Aizu1961] H. Aizu et al., Heavy nuclei in the primary cosmic radiation at Prince Albert, Canada.
II. Phys. Rev. 121, 1206 (1961)
[Alexeev1973] E.N. Alexeev et al., The intensity of the μ->e decays at different depths underground,
in Proceedings of the 13th International Cosmic Ray Conference, Denver, Colorado, vol. 3 (1973),
p. 1936
[Alfvén1949] H. Alfvén, On the solar origin of the cosmic radiation. Phys. Rev. 75, 1732 (1949)
[Alfvén1959] H. Alfvén, Momentum spectrum of cosmic radiation. Tellus 11, 106 (1959)
[Allan1966] H.R. Allan, J.K. Jones, Radio pulses from extensive air showers. Nature 212, 129
(1966)
[Allan1971] H.R. Allan, Radio emission from extensive showers. Prog. Elem. Part. Cosmic Ray
Phys. 10, 171 (1971)
[Allkofer1964] O.C. Allkofer, J. Trumper, The muon spectrum at 3000 m altitude. Z. Naturforsch.
A19, 1304 (1964)
[Allkofer1965] O.C. Allkofer, E. Kraft, Das Spektrum der Muonen, der Protonen und der Elektronen
in 5200 m Höhe. Nuovo Cimento 39, 1051 (1965)
[Allkofer1967] O.C. Allkofer, R.D.Andresen, Nuovo Cimento 51B, 329 (1967)
[Allkofer1970] O.C. Allkofer et al., Proceedings of the VI Inter-American Seminar on Cosmic Rays,
vol. 4 (1970), p. 930
[Allkofer1971] O.C. Allkofer, K. Carstensen, W.D. Dau, The absolute cosmic ray muon spectrum
at sea level. Phys. Lett. B 36, 425 (1971)
[Allkofer1979] O.C. Allkofer et al., The muon charge ratio up to 7 TeV measured with the DEIS
spectrometer, in Proceedings of the 16th International Cosmic Ray Conference, Kyoto (Japan),
vol. 10 (1979), p. 56
[Aloisio2001] A. Aloisio et al., The ARGO-YBJ experiment in Tibet. Nuovo Cimento 24C, 739
(2001)
[Alvarez1933] L. Alvarez, H.A. Compton, A positively charged component of cosmic rays. Phys.
Rev. 43, 835 (1933)
[Alvarez1970] L.W. Alvarez et al., Search for hidden chambers in the pyramids. Science 167, 832
(1970)
[Ambrosi2011] G.Ambrosi et al., The MU-RAY project: volcano radiography with cosmic ray
muons. Nucl. Instrum. Methods Phys. Res. A628, 120 (2011)
[Ambrosio1999] M. Ambrosio et al., Measurement of the energy spectrum of underground muons
at Gran Sasso with a transition radiation detector. Astropart. Phys. 10, 11 (1999)
[Ambrosino2014] F. Ambrosino et al., The MU-RAY project: detector technology and first data
from Mt. Vesuvius. J. Instrum. 9, C02029 (2014)
Bibliography 343
[Ambrosino2015a] F. Ambrosino et al., Joint measurement of the atmospheric muon flux through
the Puy de Dôme volcano with plastic scintillators and Resistive Plate Chambers detectors. J.
Geophys. Res. Solid Earth 120, 7290 (2015)
[Ambrosino2015b] F. Ambrosino et al., Assessing the feasibility of interrogating nuclear waste
storage silos using cosmic-ray muons. J. Instrum. 10, T06005 (2015)
[Amenomori2004] M. Amenomori et al., Observation by an air-shower array in Tibet of the multi-
TeV cosmic-ray anisotropy due to terrestrial orbital motion around the Sun. Phys. Rev. Lett. 93,
061101 (2004)
[Amenomori2006] M. Amenomori et al., Anisotropy and corotation of galactic cosmic rays. Science
314, 439 (2006)
[Amenomori2010] M. Amenomori et al., On temporal variations of the multi-TeV cosmic ray
anisotropy using the Tibet III air shower array. Astrophys. J. 711, 119 (2010)
[Amenomori2017] M. Amenomori et al., Northern sky galactic cosmic ray anisotropy between 10
and 1000 TeV with the Tibet air shower array. Astrophys. J. 836, 153 (2017)
[Anchordoqui2019] L.A. Anchordoqui, Ultra-high-energy cosmic rays. Phys. Rep. 801, 1 (2019)
[Anderson1932a] C.D. Anderson, Energies of cosmic-ray particles. Phys. Rev. 41, 405 (1932)
[Anderson1932b] C.D. Anderson, The apparent existence of easily deflectable positives. Science
76, 238 (1932)
[Anderson1933a] C.D. Anderson, The positive electron. Phys. Rev. 43, 491 (1933)
[Anderson1933b] C.D. Anderson, S.H. Neddermayer, Positrons from gamma rays. Phys. Rev. 43,
1034 (1933)
[Anderson1936] C.D. Anderson, S.H. Neddermayer, Cloud chamber observations of cosmic rays
at 4300 meters elevation and near sea-level. Phys. Rev. 40, 263 (1936)
[Anghel2015] V. Anghel et al., A plastic scintillator-based muon tomography system with an
integrated muon spectrometer. Nucl. Instrum. Methods Phys. Res. A798, 12 (2015)
[Antoni2003] T. Antoni et al., The cosmic ray experiment KASCADE. Nucl. Instrum. Methods
Phys. Res. A513, 490 (2003)
[Antonov1957] Yu.N. Antonov et al., Structure of the periphery of extensive atmospheric air
showers. Soviet Physics JETP 5, 172 (1957)
[Antonov1960] R.A. Antonov et al., Air showers at altitudes 9–12 km, in Proceedings of the 6th
International Cosmic Ray Conference, Moscow, vol. 2 (1960), p. 96
[Antonov1964a] R.A. Antonov et al., Altitude variation of vertical extensive air showers in the
upper part of the atmosphere. Sov. Phys. JETP 18, 1279 (1964)
[Antonov 1964b] R.A. Antonov et al., Formation of high energy gamma quanta in extensive air
showers with energy 1014 –1015 eV in the upper third of the atmosphere. Sov. Phys. JETP 19, 20
(1964)
[Antonov1971] R.A. Antonov et al., Altitude variation of EAS in the upper atmosphere, in
Proceedings of the 12th International Cosmic Ray Conference, Tasmania, vol. 6 (1971), p. 2194
[Antonov1975] R.A. Antonov et al., Altitude dependence of EAS in the upper atmosphere and the
μ-meson shower intensity near sea level, in Proceedings of the 14th International Cosmic Ray
Conference, Munich, vol. 8 (1975), p. 2714
[Antonov1983] R.A. Antonov et al., Altitude variation of EAS and particle lateral distribution func-
tion in the upper atmosphere, in Proceedings of the 18th International Cosmic Ray Conference,
Bangalore, vol. 6 (1983), p. 19
[Apel2010] W.D. Apel et al., The KASCADE-Grande experiment. Nucl. Instrum. Methods Phys.
Res. A620, 202 (2010)
[Apparao1967] K.M.V. Apparao, Upper limits on the abundance of antiprotons in the low energy
galactic cosmic radiation. Nature 215, 727 (1967)
[Armstrong1973] T.W. Armstrong et al., Calculations of neutron flux spectra induced in the Earth’s
atmosphere by galactic cosmic rays. J. Geophys. Res. 78, 2715 (1973)
[Arneodo2019] F. Arneodo et al., Measurement of cosmic muons angular distribution in Abu Dhabi
at sea level. Nucl. Instrum. Methods Phys. Res. A936, 242 (2019)
344 Bibliography
[Ashton1960] F. Ashton et al., in Proceedings of the 6th International Cosmic Ray Conference,
Moscow, vol. 1 (1959), p. 302
[Askaryan1962] G.A. Askaryan, Excess negative charge of an electron-photon shower and its
coherent radio emission. Sov. Phys. JETP 14, 441 (1962)
[Auger1934] P. Auger, P. Ehrenfest, Corpuscules ultrapénétrants du rayonnement cosmique.
Comptes Rendus 199, 1609 (1934)
[Auger1938a] P. Auger, R. Maze, T. Grivet-Meyer, Grandes gerbes cosmique atmospherique
contenantes des corpuscules ultrapénétrants. Comptes Rendus 206, 1721 (1938)
[Auger1938b] P. Auger, R. Maze, Les grandes gerbes cosmiques de l’atmosphère. Comptes Rendus
207, 228 (1938)
[Auger1939a] P. Auger, R. Maze, P. Ehrenfest Jr., A. Fréon, Les grandes gerbes de rayons cosmiques.
J. Phys. Radium 10, 39 (1939)
[Auger1939b] P. Auger, R. Maze, Robley, Comptes Rendus 208 (1939) 1641
[Auger1939c] P. Auger et al., Extensive cosmic-ray showers. Rev. Mod. Phys. 11, 288 (1939)
[Auger1940] P. Auger, J. Daudin, Diurnal variations of extensive showers. Phys. Rev. 61, 95 (1940)
[Auger1942] P. Auger, J. Daudin, Absorption of extensive atmospheric showers of cosmic rays in
air and lead. Phys. Rev. 61, 91 (1942)
[Auger1945] P. Auger, J. Dadin, Les grandes gerbes de l’air. J. Phys. Radium 6, 233 (1945)
[Auger1949] R. Maze, A. Fréon, J. Daudin, P. Auger, Extensive and penetrating atmospheric
showers. Rev. Mod. Phys. 21, 914 (1949)
[AUGER_OBS] Web site of the Pierre Auger Observatory: www.auger.com
[Axford1977] W.I. Axford, E. Leer, G. Skadron, The acceleration of cosmic rays by shock waves,
in Proceedings of the 15th International Cosmic Ray Conference, Plovdiv (Bulgaria), vol. 11
(1977), p. 132
[Babson1990] J. Babson et al., Cosmic ray muons in the deep ocean. Phys. Rev. D 42, 3613 (1990)
[Baesso2014] P. Baesso et al., Toward a RPC-based muon tomography system for cargo containers.
J. Instrum. 9, C10041 (2014)
[Bagge1950] E. Bagge, Eine Deutung der Expansion des Kosmos. Z. Phys. 128, 239 (1950)
[Bahmanabadi2005] M. Bahmanabadi et al., Experimental studies of positive and negative
atmospheric muons with a cosmic ray telescope. Astropart. Phys. 24, 183 (2005)
[Bahmanabadi2019] M. Bahmanabadi, A method for determining the angular distribution of atmo-
spheric muons using a cosmic ray telescope. Nucl. Instrum. Methods Phys. Res. A916, 1
(2019)
[Bandieramonte2015] M. Bandieramonte et al., Clustering analysis for muon tomography data
elaboration in the muon portal project. J. Phys: Conf. Ser. 608, 012046 (2015)
[Baradzei1959] L.T. Baradzei et al., Momentum spectrum of particles of the hard component of
cosmic rays at an altitude of 9 km. Sov. Phys. JETP 36, 1151 (1959)
[Barber1949] W.C. Barber, East-West asymmetry and latitude effect of cosmic rays at altitudes up
to 33000 feet. Phys. Rev. 75, 590 (1949)
[Barbouti1983] A.I. Barbouti, B.C. Rastin, A study of the absolute intensity of muons at sea level
and under various thicknesses of absorber. J. Phys. G9, 1577 (1983)
[Barkas1963] W.H. Barkas, Nuclear research emulsions, techniques and theory, in Pure and Applied
Physics, A Series of Monographs and Textbooks, vol. 15 (Academic, New York, London, 1963)
[Barnóthy1939] J. Barnóthy, M. Forró, Cosmic ray particles at great depth. Phys. Rev. 55, 870
(1939)
[Barradas2010] F. Barradas-Solad, P. Alameda-Melendez, Bringing particle physics to life: build
your own cloud chamber. Sci. Sch. 14, 36 (2010)
[Barrett1952] P.H. Barrett et al., Interpretation of cosmic ray measurements far underground. Rev.
Mod. Phys. 24, 133 (1952)
[Bartelt2018] V.K. Cirkel-Bartelt, A Quest for Truth? Heroic Patterns in Early Cosmic Ray Physics.
https://doi.org/10.6094/helden.heroes.heros./2018/HS/06
[Bassi1953] P. Bassi, G. Clark, B. Rossi, Distribution of arrival times of air shower particles. Phys.
Rev. 92, 441 (1953)
Bibliography 345
[Batignani2011] G. Batignani et al., L’esperimento di Pacini sull’origine dei raggi cosmici. G. Fis.
42, 1 (2011)
[Bazilevskaya2014] G.A. Bazilevskaya, Skobeltsyn and the early years of cosmic particle physics
in the Soviet Union. Astropart. Phys. 53, 61 (2014)
[Beedle1970] R.E. Beedle, W.R. Webber, Measurement of cosmic ray electrons in the energy range
4 MeV to 6 BeV at 2 g/cm2 atmospheric depth at Ft. Churchill. Can. J. Phys. 46, S1014 (1968)
[Bell1948] P.R. Bell, The use of anthracene as a scintillation counter. Phys. Rev. 73, 1405 (1948)
[Bell1978] R. Bell, The acceleration of cosmic rays in shock front. Mon. Not. Roy. Astron. Soc.
182, 147 (1978)
[Bellotti1999] R. Bellotti et al., Balloon measurements of cosmic ray muon spectra in the atmosphere
along with those of primary protons and helium nuclei over midlatitude. Phys. Rev. D 60, 052002
(1999)
[Belov2005] A. Belov et al., A study of the ground level enhancement of 23 February 1956. Adv.
Space Res. 35, 697 (2005)
[Belov2009] A.V. Belov, Forbush effects and their connection with solar, interplanetary and
geomagnetic phenomena. Proc. IAU Symp. 257, 439 (2008)
[Bennett1931] R.D. Bennett, J.C. Stearns, A.H. Compton, The constancy of cosmic rays. Phys. Rev.
38, 1566 (1931)
[Bennett1932a] R.D. Bennett, J.C. Stearns, H. Compton, Diurnal variation of cosmic rays. Phys.
Rev. 41, 119 (1932)
[Bennett1932b] R.D. Bennett et al., Intensity of cosmic ray ionization in Western North America.
Phys. Rev. 42, 446 (1932)
[Běhounek1929] F. Běhounek, Atmospheric-electric researches made in 1928 during the Nobile
Arctic Expedition in collaboration with Professor A. Pontremoli (Milan) and Professor F.
Malmgren (Uppsala). Terr. Magn. Atmos. Electr. 34, 173 (1929)
[Bergeson1975] H.E. Bergeson et al., The Fly’s Eye, a novel technique for sensing extensive air
showers, in Proceedings of the 14th International Cosmic Ray Conference, Munich, vol. 8 (1975),
p. 3059
[Bergwitz1910] K. Bergwitz, Die Gammastrahlung des Erdkörpers und ihr Anteil an der spontanen
Ionisierung der Atmosphäre. Jahresbericht des Braunschweiger Vereins für Naturwissenschaft
10, 196 (1910)
[Bethe1934] H.A. Bethe, W. Heitler, On the stopping of fast particles and on the creation of positive
electrons. Proc. Roy. Soc. 146, 83 (1934)
[Bethe1953] H.A. Bethe, Molière’s theory of multiple scattering. Phys. Rev. 89, 1256 (1953)
[Beuermann1968] K.P. Beuermann, G. Wibberenz, Can. J. Phys. 46, S1034 (1968)
[Bhat1978] P.N. Bhat, P.V. Murthy, Angular distribution of low-energy cosmic ray muons
underground. J. Phys. G: Nucl. Phys. 4, 453 (1978)
[Bhattacharyya1974] D.P. Bhattacharyya, Absolute sea-level integral muon spectra at zenith angles
45 °W and 60 °W near the geomagnetic equator in the momentum range (0.4÷3) GeV/c. Nuovo
Cimento B24, 78 (1974)
[Biehl1949] A.T. Biehl et al., Cosmic-ray experiments at high altitudes over a wide range of latitudes.
Phys. Rev. 76, 914 (1949)
[Bierman1949] L. Bierman, E. Bagge, Über die Entstehung der solaren Komponente der Ultra-
strahlung. Z. Naturforsch. 4a, 303 (1949)
[Bikit2016] I. Bikit et al., Novel approach to imaging by cosmic ray muons. Europhys. Lett. 113,
58001 (2016)
[Bjorklund1950] R. Bjorklund et al., High energy photons from proton-nucleon collisions. Phys.
Rev. 77, 213 (1950)
[Blackett1932] P.M.S. Blackett, G. Occhialini, Photography of penetrating corpuscular radiation.
Nature 130, 363 (1932)
[Blackett1933a] P.M.S. Blackett, G. Occhialini, Some photographs of the tracks of penetrating
radiation. Proc. Roy. Soc. A139, 699 (1933)
[Blackett1933b] P.M.S. Blackett, The positive electron. Nature 132, 917 (1933)
346 Bibliography
[Blackett1948] P.M.S. Blackett, A possible contribution to the light of the night sky from the
Cherenkov radiation emitted by cosmic rays, in Physical Society Gassiot Committee Report: The
Emission Spectra of the Night Sky and Aurorae, p. 34
[Blanco2006a] F. Blanco et al., Geiger counters offer powerful way to teach detection methods.
Phys. Educ. 41, 204 (2006)
[Blanco2006b] F. Blanco et al., Educational cosmic ray experiments with Geiger counters. Nuovo
Cimento C 29, 381 (2006)
[Blanco2008a] F. Blanco, P. La Rocca, F. Riggi, Educational experiments with cosmic rays, in
Science Education in Focus, ed. by M.V. Thomase (Nova Publishers, New York, 2008). ISBN:
1-60021-949-7
[Blanco2008b] F. Blanco, P. La Rocca, F. Riggi, S. Riggi, Timing the random and anomalous arrival
of particles in a Geiger counter with GPS devices. Eur. J. Phys. 29, 355 (2008)
[Blanco2008c] F. Blanco et al., Cosmic ray measurements by scintillators with metal resistor
semiconductor avalanche photo diodes. Phys. Educ. 43, 536 (2008)
[Blanco2009] F. Blanco, P. La Rocca, F. Riggi, Cosmic ray physics from sea level to aircraft cruise
altitudes. Eur. J. Phys. 30, 685 (2009)
[Blandford1978] R.D. Blandford, J.P. Ostriker, Particle acceleration by astrophysical shocks.
Astrophys. J. 221, L29 (1978)
[Blandford1987] R. Blandford, D. Eichler, Particle acceleration at astrophysical shocks: a theory
of cosmic ray origin. Phys. Rep. 154, 1 (1987)
[Blau1925] M. Blau, Über die photographische Wirkung natürlicher H-Strahlen. Proc. Vienna Acad.
Sci. Sect. IIa 134, 427 (1925)
[Blau1932] M. Blau, H. Wambacher, Photographic detection of protons liberated by neutrons. Proc.
Vienna Acad. Sci. Sect. II 141, 617 (1932)
[Blau1937] M. Blau, H. Wambacher, Disintegration processes by cosmic rays with the simultaneous
emission of several heavy particles. Nature 140, 585 (1937)
[Blokh1977] Y.L. Blokh, L.I. Dorman, I.Y. Libin, Studies of the angular distribution and of the alti-
tude dependence of the cosmic-ray muon and general components using scintillation telescopes.
Nuovo Cimento B37, 198 (1977)
[Blumer2009] J. Blumer et al., Cosmic rays from the knee to the highest energies. Prog. Part. Nucl.
Phys. 63, 293 (2009)
[Bodini2007] I. Bodini et al., Cosmic ray detection based measurement systems. Meas. Sci. Technol.
18, 3537 (2007)
[Bogomolov1971] E.A. Bogomolov et al., Proceedings of the 12th International Cosmic Ray
Conference, Hobart (Australia), vol. 5 (1971), p. 1730
[Bogomolov1979] E.A. Bogomolov et al., Proceedings of the 16th International Cosmic Ray
Conference, Kyoto (Japan), vol. 1 (1979), p. 330
[Bonechi2020] L. Bonechi et al., Atmospheric muons as an imaging tool. Rev. Phys. 5, 100038
(2020)
[Bonolis2011a] L. Bonolis, Bruno Rossi and the racial laws of fascist Italy. Phys. Perspect. 13, 58
(2011)
[Bonolis2011b] L. Bonolis, Walther Bothe and Bruno Rossi: the birth and development of
coincidence methods in cosmic-ray physics. Am. J. Phys. 79, 1133 (2011)
[Bonolis2014] L. Bonolis, From cosmic ray physics to cosmic ray astronomy: Bruno Rossi and the
opening of new windows on the universe. Astropart. Phys. 53, 67 (2014)
[Bonomi2014] G. Bonomi et al., Muon tomography as a tool to detect radioactive source shielding
in scrap metal containers. Int. J. Mod. Phys.: Conf. Ser. 27, 1460157 (2014)
[Bonomi2017] G. Bonomi, Progress in muon tomography, in Proceedings of Science POS (EPS-
HEP2017), p. 609
[Bonomi2019] G. Bonomi et al., Cosmic ray tracking to monitor the stability of historical buildings:
a feasibility study. Meas. Sci. Technol. 30, 045901 (2019)
[Borozdin2003] K.N. Borozdin et al., Radiographic imaging with cosmic ray muons. Nature 422,
277 (2003)
Bibliography 347
[Borozdin2012] K.N. Borozdin et al., Cosmic ray radiography of the damaged cores of the
Fukushima reactors. Phys. Rev. Lett. 109, 152501 (2012)
[Bothe1929a] W. Bothe, W. Kolhörster, Das Wesen der Höhenstrahlung. Z. Phys. 56, 751 (1929)
[Bothe1929b] W. Bothe, W. Kolhörster, The nature of the penetrating radiation. Nature 123, 638
(1929)
[Bothe1930a] W. Bothe, W. Kolhörster, Berl. Ber. 450 (1930)
[Bothe1930b] W. Bothe, H. Becker, Künstliche Erregung von Kern–γ–Strahlen. Z. Phys. 66, 289
(1930)
[Bothe1937] W. Bothe et al., Phys. Z. 38, 964 (1937)
[Bothe1954] Walther Bothe Nobel Lecture, see the Web site: https://www.nobelprize.org/prizes/
physics/1954/bothe/lecture/
[Bouteille2016] S. Bouteille et al., A Micromegas-based telescope for muon tomography: The
WatTo experiment. Nucl. Instrum. Methods Phys. Res. A834, 223 (2016)
[Boys1889] C.V. Boys, Quartz as an insulator. Philos. Mag. 28, 14 (1889)
[Bradt1948] H.L. Bradt, B. Peters, Investigation of the primary cosmic radiation with nuclear
photographic emulsions. Phys. Rev. 74, 1828 (1948)
[Bradt1950] H.L. Bradt, B. Peters, The heavy nuclei of the primary cosmic radiation. Phys. Rev.
77, 54 (1950)
[Broadbent1950] D. Broadbent et al., The density spectrum and structure of extensive cosmic ray
showers at sea level. Proc. Phys. Soc. A63, 864 (1950)
[Brode1936] R.B. Brode et al., The heavy particle component of the cosmic radiation. Phys. Rev.
51, 581 (1936)
[Brooke1964] G. Brooke, A.W. Wolfendale, The momentum spectrum of cosmic ray protons near
sea level in the momentum range 0.6–150 GeV/c. Proc. Phys. Soc. 83, 843 (1964)
[Broser1947a] I. Broser, H. Kallmann, Uber die Anregung von Leuchtstoffen durch schnelle
Korpuskularteilchen I. Zeitschrift fur Naturforschung 2a, 439 (1947)
[Broser1947b] I. Broser, H. Kallmann, Ueber den Elementarprozess der Lichtanregung in
Leuchtstoffen durch Alphateilchen, Schnelle Elektronen und Gammaquanten. Zeitschrift fur
Naturforschung 2a, 642 (1947)
[Brown1949] R. Brown et al., Observations with electron-sensitive plates exposed to cosmic
radiation. Nature 163, 47 (1949)
[Brownlee1968] R.G. Brownlee et al., Design of an array to record air showers of energy up to 1021
eV. Can. J. Phys. 46, S259 (1968)
[Bugaev1998] E.V. Bugaev et al., Atmospheric muon flux at sea level, underground, and underwater.
Phys. Rev. D 58, 054001 (1998)
[Bunner1967] N.A. Bunner, Cosmic Ray Detection by Atmospheric Fluorescence, Ph.D. Thesis,
Cornell University (1967)
[Caffau1997] E. Caffau et al., Underground cosmic ray measurement for morphological recon-
struction of the Grotta Gigante natural cave. Nucl. Instrum. Methods Phys. Res. A385, 480
(1997)
[CALTECH] https://www.caltech.edu/about/news/caltech-to-remove-the-names-of-robert-a-mil
likan-and-five-other-eugenics-proponents
[Carbone2014] D. Carbone et al., An experiment of muon radiography at Mt. Etna (Italy). Geophys.
J. Int. 196, 633 (2014)
[Carlson1941] J.F. Carlson, M. Schein, On the production of mesotrons. Phys. Rev. 59, 840 (1941)
[Carlson2011] P. Carlson, A. De Angelis, Nationalism and internationalism in science: the case of
the discovery of cosmic rays. Eur. Phys. J. H35, 309 (2011)
[Castagnoli1997] C. Castagnoli et al., Multiple interactions of muons in the NUSEX detector and
muon energy spectrum deep underground. Astropart. Phys. 6, 187 (1997)
[Cerenkov1937] P. Cherenkov, Visible radiation produced by electrons moving in a medium with
velocities exceeding that of light. Phys. Rev. 52, 378 (1937)
[CERN-Courier1966] A.J. Herz, W.O. Lock, CERN Courier, May 1966, https://cerncourier.com/a/
nuclear-emulsions/
348 Bibliography
[Chadwick1932a] J. Chadwick, Possible existence of a neutron. Nature 129, 312 (1932)

[Chadwick1932b] J. Chadwick, The existence of a neutron. Proc. Roy. Soc. 136, 692 (1932)
[Chamberlain1955] O. Chamberlain et al., Observation of antiprotons. Phys. Rev. 100, 947 (1955)
[Chandrasekhar1953] S. Chandrasekhar, E. Fermi, Problems of gravitational stability in the presence
of magnetic field. Astrophys. J. 118, 113 (1953)
[Chatzidakis2016] S. Chatzidakis et al., Interaction of cosmic ray muons with spent nuclear fuel
dry casks and determination of lower detection limit. Nucl. Instrum. Methods Phys. Res. A828,
37 (2016)
[Checchia2016] P. Checchia, Review of possible applications of cosmic muon tomography. J.
Instrum. 11, C12072 (2016)
[Chiba1992] N. Chiba et al., Akeno Giant Air Shower Array (AGASA) covering 100 km2 area.
Nucl. Instrum. Methods Phys. Res. A311, 338 (1992)
[Clark1957a] G.W. Clark et al., Preparation of large plastic scintillators. Rev. Sci. Instrum. 28, 433
(1957)
[Clark1957b] G.W. Clark et al., An experiment on air showers produced by high-energy cosmic
rays. Nature 180, 353 (1957)
[Clark1961] G.W. Clark et al., Cosmic-ray air showers at sea level. Phys. Rev. 122, 637 (1961)
[Clarkson2014] A. Clarkson et al., The design and performance of a scintillating-fibre tracker for
the cosmic-ray muon tomography of legacy nuclear waste containers. Nucl. Instrum. Methods
Phys. Res. A745, 138 (2014)
[Clarkson2015] A. Clarkson et al., Characterising encapsulated nuclear waste using cosmic-ray
muon tomography. J. Instrum. 10, P03020 (2015)
[Clay1927] J. Clay, Penetrating radiation. Proc. Acad. Sci. Amst. 30, 1115 (1927)
[Clay1928] J. Clay, Penetrating radiation II. Proc. Acad. Sci. Amst. 31, 1091 (1928)
[Clay1939] J. Clay, The absolute value of cosmic-ray ionization at sea level in different gases. Rev.
Mod. Phys. 11, 128 (1939)
[Clay1984] R.W. Clay et al., The Buckland Park Air Shower Array, in Proceedings of the 19th
International Cosmic Ray Conference, La Jolla (USA), vol. 3 (1984), p. 414
[CLIMATE] https://www.climate4you.com/Sun.htm
[CLOUD] https://home.cern/science/experiments/cloud
[Cocconi1941] G. Cocconi, On the protonic nature of the cosmic radiation. Phys. Rev. 60, 532
(1941)
[Cocconi1944a] G. Cocconi, A. Loverdo, V. Tongiorgi, Lo spettro di densità degli sciami estesi
nell’aria. Nuovo Cimento 2, 14 (1944)
[Cocconi1944b] G. Cocconi, A. Loverdo, V. Tongiorgi, Sulla costituzione degli sciami estesi
nell’aria. Nuovo Cimento 2, 28 (1944)
[Cocconi1946a] G. Cocconi, A. Loverdo, V. Tongiorgi, Experimental and theoretical evaluation of
the density spectrum of extensive cosmic-ray showers. Phys. Rev. 70, 846 (1946)
[Cocconi1946b] G. Cocconi, A. Loverdo, V. Tongiorgi, Penetrating particles in air showers. Phys.
Rev. 70, 852 (1946)
[Cocconi1946c] G. Cocconi, V. Tongiorgi, On the fine structure of the zenithal curves of the cosmic
radiation. Phys. Rev. 70, 850 (1946)
[Cocconi1949a] G. Cocconi, Results and problems concerning the extensive air showers. Rev. Mod.
Phys. 21, 26 (1949)
[Cocconi1949b] G. Cocconi, Some properties of the cosmic ray ionizing particles that generate
penetrating showers. Phys. Rev. 75, 1074 (1949)
[Cocconi1949c] G. Cocconi, Meson background in penetrating shower experiments. Phys. Rev. 76,
984 (1949)
[Cocconi1949d] G. Cocconi, V. Cocconi Tongiorgi, The density spectrum of extensive air showers
of cosmic rays. Phys. Rev. 75, 1058 (1949)
[Cocconi1949e] G. Cocconi, V. Cocconi Tongiorgi, Angular distribution and multiplicity of neutrons
associated with local cosmic-ray showers. Phys. Rev. 76, 318 (1949)
Bibliography 349
[Cocconi1949f] G. Cocconi, V. Cocconi Tongiorgi, K. Greisen, The lateral structure of cosmic-ray

air showers. Phys. Rev. 76, 1020 (1949)
[Cocconi1956] G. Cocconi, Intergalactic space and cosmic rays. Nuovo Cimento 3, 1433 (1956)
[Colgate1960] S.A. Colgate, M.H. Johnson, Hydrodynamic origin of cosmic rays. Phys. Rev. Lett.
5, 235 (1960)
[Coltman1947] J.W. Coltman, F.H. Marshall, Photomultiplier radiation detector. Nucleonics 1, 58
(1947)
[Compton1923] A.H. Compton, A quantum theory of the scattering of X-rays by light elements.
Phys. Rev. 21, 483 (1923)
[Compton1931] A.H. Compton, R.D. Bennett, J.C. Stearns, Ionization as a function of pressure and
temperature. Phys. Rev. 38, 1565 (1931)
[Compton1932a] A.H. Compton, R.D. Bennett, J.C. Stearns, Ionization by penetrating radiation as
a function of pressure and temperature. Phys. Rev. 39, 873 (1932)
[Compton1932b] A.H. Compton, Variation of cosmic rays with latitude. Phys. Rev. 41, 111 (1932)
[Compton1932c] A.H. Compton, Progress of cosmic-ray survey. Phys. Rev. 41, 681 (1932)
[Compton1933] A.H. Compton, A geographic study of cosmic rays. Phys. Rev. 43, 387 (1933)
[Compton1934] A.H. Compton, H. Bethe, Composition of cosmic rays. Nature 134, 734 (1934)
[Compton1935] A.H. Compton, I.A. Getting, An apparent effect of galactic rotation on the intensity
of cosmic rays. Phys. Rev. 47, 817 (1935)
[Compton1936a] A.H. Compton, Cosmic rays as electrical particles. Phys. Rev. 50, 1119 (1936)
[Compton1936b] A.H. Compton, Recent developments in cosmic rays. Rev. Sci. Instrum. 7, 71
(1936)
[Copeland2020] K. Copeland, CARI-7 Documentation: Geomagnetic Cutoff Rigidity Calculations
and Tables for 1965–2010, DOT/FAA/AM-19/4 Office of Aerospace Medicine, Washington, DC
[Corlin1927] A. Corlin, Über den kosmischen Ursprung der durchdringenden Höhenstrahlung.
Astron. Nachr. 9, 137 (1927)
[Corlin1928] A. Corlin, The highly penetrating cosmic rays. Nature 121, 322 (1928)
[Corlin1930] A. Corlin, Lund Medd. 121, 39 (1930)
[Cosyns1940] M.G.E. Cosyins, Barometric coefficient of extensive air showers. Nature 145, 668
(1940)
[Coulomb1785] C. de Coulomb, Troisième mémoire sur l’électricité et le magnétisme. De la quantité
d’électricité qu’un corps isolé perd dans le temps donné, soit par le contact de l’air plus ou moins
humide, soit le lond des soutiens plus ou moins idio-électriques, avec 1 planche. Mém. Acad.
Roy. Sci. 88, 612 (1785)
[Cowan1948] E.W. Cowan, Evidence for the existence of a low mass mesotron. Science 108, 534
(1948)
[Critchfield1951] C.L. Critchfield, E.P. Ney, S. Oleska, Soft radiation at balloon altitudes. Phys.
Rev. 85, 461 (1951)
[Crookes1878] W. Crookes, On Electrical Insulation in high vacua. Proc. R. Soc. Lond. 28, 347
(1879)
[Crookes1903] W. Crookes, Certain properties of the emanations of radium. Chem. News 87, 241
(1903)
[Crookes1972] J.N. Crookes, B.C. Rastin, An investigation of the absolute intensity of muons at
sea level. Nucl. Phys. B 39, 493 (1972)
[Curie1898] M. Curie, Rayons émis par les composés de l’uranium et du thorium. Comptes Rendus
126, 1101 (1898)
[Curie1899] M. Curie, Les rayons de Becquerel et le polonium. Rev. Gen. Sci. Pures Appl. 10, 41
(1899)
[Curie1930] I. Curie, Sur le rayonnement γ nucléaire excité dans le glucinium et dans le lithium par
les rayons αdu polonium. Compte Rendus Séances Academie Science Paris 193, 1412 (1930)
[Curie1932a] I. Curie, F. Joliot, Émission de protons à grande vitesse par les substances hydrogénées
sous l’influence des rayons γ très pénétrants. Compte Rendus Séances Academie Science Paris
194, 273 (1932)
350 Bibliography
[Curie1932b] I. Curie, F. Joliot, Effet d’absorption de rayons γde très haute fréquence par projection
de noyaux légers. Compte Rendus Séances Academie Science Paris 194, 708 (1932)
[Curie1932c] I. Curie, F. Joliot, Sur la nature du rayonnement pénétrant excité dans les noyaux
légers par les particules α. Compte Rendus Séances Academie Science Paris 194, 1229 (1932)
[Curie1933] I. Curie, F. Joliot, Sur l’origin des electron positifs. Compte Rendus Séances Academie
Science Paris 196, 1581 (1933)
[Curran1944] S.C. Curran, W.R. Backer, A photoelectric alpha particle detector, U. S. Atomic
Energy Commission Report MDDC 1296, 17 Nov 1944
[Curtiss1928] L.F. Curtiss, On the action of the Geiger counter. Phys. Rev. 31, 1060 (1928)
[Daggs1952] W. Daggs, W. Parr, A. Krebs, Tests on scintillation G-M tubes. Nucleonics 10, 54
(1952)
[Daion1959] M.I. Daion, L.I. Potapov, Underground muon spectrum at ~40 m water equivalent.
Sov. Phys. JETP 36, 488 (1959)
[Dauvillier1934] M.A. Dauvillier, Researches on cosmic rays at Scoresby Sund during the polar
year. J. Phys. Radium 5, 640 (1934)
[Davis1956] L. Davis, Modified Fermi mechanism for the acceleration of cosmic rays. Phys. Rev.
101, 351 (1956)
[Davitaev1970] L.N. Davitaev et al., Intensity measurements of cosmic rays in water at large depths,
in Proceedings of the 11th International Conference on Cosmic Rays, Budapest, Suppl. Acta Phys.
Acad. Sci. Hung. 29, 53 (1970)
[DeAngelis2010] A. De Angelis, Domenico Pacini, uncredited pioneer of the discovery of cosmic
rays. La Rivista del Nuovo Cimento 33, 713 (2010)
[DeAngelis2012] A. De Angelis, L’enigma dei raggi cosmici (Springer, Milano, 2012)
[DeAngelis2014] A. De Angelis, Atmospheric ionization and cosmic rays: studies and measure-
ments before 1912. Astropart. Phys. 53, 19 (2014)
[DeAngelis2018a] V.F. Hess, Measurements of the Penetrating Radiation During Seven Balloon
Flights (Translated and commented by A. De Angelis, C.A.B. Schultz), arXiv:1808.02927v2 10
Aug 2018
[DeAngelis2018b] A. De Angelis, M. Mallamaci, Gamma-ray astrophysics. Eur. Phys. J. Plus 133,
324 (2018)
[Degrange2013] B. Degrange, G. Fontaine, P. Fleury, Tracking Louis Leprince-Ringuet’s contribu-
tion to cosmic ray physics. Phys. Today 66, 8 (2013)
[DeShong1964] J.A. De Shong Jr., R.H. Hildebrand, P. Meyer, Ratio of electrons to positrons in
the primary cosmic radiation. Phys. Rev. Lett. 12, 3 (1964)
[Deutschmann1947] M. Deutschmann, Untersuchung der kosmischen Strahlenschauer mit Hilfe
einer großen Wilson-Kammer. Z. Naturforsch. 2a, 61 (1947)
[Dirac1928] P.A.M. Dirac, The quantum theory of the electron. Proc. Roy. Soc. 117, 610 (1928)
[Dmietreva2006] A.N. Dmietreva et al., Measurement of integrated muon intensity at large zenith
angles. Phys. At. Nucl. 69, 865 (2006)
[Dobrowolska2020] M. Dobrowolska et al., Towards an application of muon scattering tomography
as a technique for detecting rebars in concrete. Smart Mater. Struct. 29, 055015 (2020)
[Donzella2014] A. Donzella et al., Historical building stability monitoring by means of a cosmic
ray tracking system. Nuovo Cimento C 37, 223 (2014)
[Dorman1974] L.I. Dorman, Cosmic Rays, Variations and Space Explorations (North Holland
Publishing Company, Amsterdam, 1974)
[Drury1983] L. O’C Drury, An introduction to the theory of diffusive shock acceleration of energetic
particles in tenuous plasmas. Rep. Prog. Phys. 46, 973 (1983)
[Drury2012] L. O’C Drury, Origin of cosmic rays. Astropart. Phys. 39–40, 52 (2012)
[Durham2015] J.M. Durham et al., Tests of cosmic ray radiography for power industry applications.
AIP Adv. 5, 067111 (2015)
[Earl1961] J.A. Earl, Cloud-chamber observations of primary cosmic ray electrons. Phys. Rev. Lett.
6, 125 (1961)
[Ebert1901] H. Ebert, Über Elektricitätszerstreuung in grösseren Höhen. Ann. Phys. 310, 718 (1901)
Bibliography 351
[EEE] Web site of the EEE Collaboration: https://eee.centrofermi.it/

[Elster1899] J. Elster, H. Geitel, Uber einen apparat zur messung der Elektricitatszerstreuung in der
luft. Phys. Z. 1, 11 (1899)
[Elster1900a] J. Elster, H. Geitel, Über Elektricitätszerstreuung in der Luft. Ann. Phys. 307, 425
(1900)
[Elster1900b] J. Elster, H. Geitel, Weitere Versuche ber die Elektrizitätszerstreuung in
abgeschlossenen Luftmengen. Phys. Z. 2, 560 (1900)
[Elster1901] J. Elster, H. Geitel, Uber die radioaktivitat der im erdboden henthaltenen luft. Phys.
Z. 3, 574 (1901)
[ESPER] https://www.esperimentanda.com/come-costruire-una-camera-a-nebbia-a-diffusione-
con-ghiaccio-secco-per-rivelare-particelle-subatomiche/
[Euler1940] H. Euler, H. Wergeland, Über die ausgedehnten Luftschauer der kosmischen Strahlung.
Astrophisica Norvegica 3, 165 (1940)
[Eve1906] A.S. Eve, On the radioactive matter in the earth and the atmosphere. Philos. Mag. Lond.
12, 189 (1906)
[Eve1907] A.S. Eve, The ionization of the atmosphere over the ocean. Philos. Mag. Lond. 13, 248
(1907)
[Fadhel2021] K.F. Fadhel et al., Estimating the lateral distribution of high energy cosmic ray particles
by depending on Nishimura-Kamata-Greisen function. J. Phys. G: Conf. Series 1879, 032089
(2021)
[Famoso2005] B. Famoso, P. La Rocca, F. Riggi, An educational study of the cosmic ray barometric
effect with a Geiger counter. Phys. Educ. 40, 461 (2005)
[Fan1951] C.Y. Fan, On Fermi’s theory of the origin of the cosmic radiation. Phys. Rev. 82, 211
(1951)
[Fanselow1969] J.L. Fanselow et al., Charge composition and energy spectrum of primary cosmic-
ray electrons. Astrophys. J. 158, 771 (1969)
[Fermi1949] E. Fermi, On the origin of cosmic radiation. Phys. Rev. 75, 1169 (1949)
[Fermi1954] E. Fermi, Galactic magnetic fields and the origin of the cosmic radiation. Astrophys.
J. 119, 1 (1954)
[Fick2014] D. Fick, D. Hoffman, Werner Kolhoster (1887–1945): the German pioneer of cosmic
ray physics. Astropart. Phys. 53, 50 (2014)
[Fichtel1978] C.E. Fichtel, G.A. Simpson, D.J. Thomson, Diffuse gamma radiation. Astrophys. J.
222, 833 (1978)
[Forbush1937] S.E. Forbush, On the effects in cosmic-ray intensity observed during the recent
magnetic storm. Phys. Rev. 51, 1108 (1937)
[Forbush1938] S.E. Forbush, On world-wide changes in cosmic-ray intensity. Phys. Rev. 54, 975
(1938)
[Forbush1938] F. von Forstner et al., Using Forbush decreases to derive the transit time of ICMEs
propagating from 1 AU to Mars. J. Geophys. Res. Space Phys. 123, 39 (2018)
[Foster1949] J.S. Foster, Arthur Stewart Eve, 1862–1948. Obituary Note Fellows Roy. Soc. 6, 396
(1949)
[Fradkin1956] M.I. Fradkin, On the problem of antiprotons in the primary stream of cosmic rays.
Sov. Phys. JETP 2, 87 (1956)
[Freier1948a] P. Freier, E.J. Lofgren, E.P. Ney, F. Oppenheimer, H.L. Bradt, B. Peters, Evidence
for heavy nuclei in the primary cosmic radiation. Phys. Rev. 74, 213 (1948)
[Freier1948b] P. Freier, E.J. Lofgren, E.P. Ney, F. Oppenheimer, The heavy component of the
primary cosmic radiation. Phys. Rev. 74, 1818 (1948)
[Fretter1948] W.B. Fretter, Penetrating showers in lead. Phys. Rev. 73, 41 (1948)
[Fricke2017] R.G.A. Fricke, K. Schlegel, Julius Elster and Hans Geitel, Dioscuri of physics and
pioneer investigators in atmospheric electricity. Hist. Geo- Space-Sci. 8, 1 (2017)
[Fuji2013] H. Fuji et al., Performance of a remotely located muon radiography system to identify
the inner structure of a nuclear plant. Prog. Theor. Exp. Phys. 7, 073C01 (2013)
352 Bibliography
[Fuji2017] H. Fuji et al., Detection of on-surface objects with an underground radiography detector
system using cosmic ray muons. Prog. Theor. Exp. Phys. 5, 053C01 (2017)
[Fulks1974] G. Fulks, P. Meyer, Cosmic ray electrons in the atmosphere. J. Geophys. (Z. Geophys.)
40, 751 (1974)
[Furlan2014] M. Furlan et al., Application of muon tomography to detect radioactive sources hidden
in scrap metal containers. IEEE Trans. Nucl. Sci. 61, 2204 (2014)
[Fyodorov1986] V.M. Fyodorov, Muon registration under water in the ocean with a Cherenkov
detector. Nucl. Instrum. Methods Phys. Res. A248, 221 (1986)
[Gaisser1974] T.K. Gaisser, Calculation of Muon yields, response functions, and sea level integral
energy spectrum using recent accelerator data and Feynman scaling. J. Geophys. Res. 79, 2281
(1974)
[Gaisser1977] T.K. Gaisser, A.M. Hillas, Reliability of the method of constant intensity cuts
for reconstructing the average development of vertical showers, in Proceedings of the 15th
International Cosmic Ray Conference, Plovdiv, vol. 8 (1977),p. 353
[Gaisser1990] T.K. Gaisser, Cosmic Rays and Particle Physics (Cambridge University Press, 1990)
[Galbraith1953] W. Galbraith, J.V. Jelley, Light pulses from the night sky associated with cosmic
rays. Nature 171, 349 (1953)
[Galgoczi2020] G. Galgoczi et al., Imaging by muons and their induced secondary particles: a novel
technique. J. Instrum. 15, C06014 (2020)
[Geiger1909] H. Geiger, E. Marsden, On a diffuse reflection of the α-particles. Proc. Roy. Soc. A82,
495 (1909)
[Geiger1928a] H. Geiger, W. Müller, Elektronenzählrohr zur Messung schwächster Aktivitäten.
Naturwissenschaften 16, 617 (1928)
[Geiger1928b] H. Geiger, W. Müller, Das Elektronenzählrohr. Phys. Z. 29, 839 (1928)
[Geiger1929a] H. Geiger, W. Müller, Technische Bemerkungen zum Elektronenzählrohr. Phys. Z.
30, 489 (1929)
[Geiger1929b] H. Geiger, W. Müller, Demonstration des Elektronenzählrohrs. Phys. Z. 30, 523
(1929)
[Geitel1901] H. Geitel, Über die Elektrizitätszerstreuung in abgeschlossenen Luftmengen. Phys. Z.
2, 116 (1901)
[George1955] E.P. George, Cosmic rays measure overburden of tunnel. Commonw. Eng. 455 (1955)
[George1956] E.P. George, G.S. Shrikantia, Observation of the energy spectrum of the cosmic
radiation below ground. Nucl. Phys. 1, 54 (1956)
[Gerontidou2021] M. Gerontidou et al., World grid of cosmic ray vertical cut-off rigidity for the
last decade. Adv. Space Res. 67, 2231 (2021)
[Giacconi1962] R. Giacconi et al., Evidence for X-rays from sources outside the solar system. Phys.
Rev. Lett. 9, 439 (1962)
[Gibbs1988] G.K. Gibbs et al., The Chicago Air Shower Array (CASA). Nucl. Instrum. Methods
Phys. Res. A264, 67 (1988)
[Giboy2005] W.B. Giboy et al., Industrial thickness gauging with cosmic-ray muons. Radiat. Phys.
Chem. 74, 454 (2005)
[Ginzburg1964] V.L. Ginzburg, S.I. Syrovatskii, The Origin of Cosmic Rays (Pergamon Press,
Oxford, 1964)
[Glatzmaier2015] G.A. Glatzmaier, R.S. Coe, Magnetic polarity reversals in the core, Treatise on
Geophysics (Elsevier, 2015)
[Gockel1908] A. Gockel, Th. Wulf, Beobachtungen über die Radioaktivität der Atmosphäre im
Hochgebirge. Phys. Z. 9, 907 (1908)
[Gockel1910] A. Gockel, Luftelektrische Beobachtungen bei einer Ballonfahrt. Phys. Z. 11, 280
(1910)
[Gockel1911] A. Gockel, Messungen der durchdringenden Strahlung bei Ballonfahrten. Phys. Z.
12, 595 (1911)
[Gockel1915] A. Gockel, Beiträge zur Kenntnis der in der Atmosphäre vorhandenen durchdrin-
genden Strahlung. Phys. Z. 16, 345 (1915)
Bibliography 353
[Gockel1923] A. Gockel, Ionization unserer Atmosphäre und Sonnentätigkeit. Phys. Z. 24, 500
(1923)
[Goldader2010] J.D. Goldader, S. Choi, An inexpensive cosmic ray detector for the classroom.
Phys. Teach. 48, 594 (2010)
[Golden1978] R.L. Golden et al., A magnetic spectrometer for cosmic ray studies. Nucl. Instr. Meth.
148, 747 (1978)
[Golden1979a] R.L. Golden et al., Results of a recent attempt to observe antiprotons in the cosmic
rays, in Proceedings of the 16th International Cosmic Ray Conference, Kyoto (Japan), vol. 12
(1979), p. 76
[Golden1979b] R.L. Golden et al., Evidence for the existence of cosmic ray antiprotons. Phys. Rev.
Lett. 43, 1196 (1979)
[Gomez2016] H. Gomez et al., Studies in muon tomography for archeological internal structure
scanning. J. Phys: Conf. Ser. 718, 052016 (2016)
[Grant1931] K. Grant, Observations on the penetrating radiation in the Antarctic. Nature 127, 924
(1931)
[Gnanvo2011] K. Gnanvo et al., Imaging of high-Z material for nuclear contraband detection with
a minimal prototype of a muon tomography station based on GEM detectors. Nucl. Instrum.
Methods Phys. Res. A652, 16 (2011)
[GRAND] Web site of the GRAND Collaboration: https://www3.nd.edu/~grand/
[GRAPES-3] Web site of the GRAPES-3 Collaboration: https://grapes-3.tifr.res.in/
[Gray1980] P.C. Gray, S.I. Akasofu, L.C. Lee, A model of the heliospheric magnetic field
configuration. Planet. Space Sci. 28, 609 (1980)
[Greisen1942] K.I. Greisen, The intensities of the hard and soft components of cosmic rays as
functions of altitude and zenith angle. Phys. Rev. 61, 212 (1942)
[Greisen1948] K.I. Greisen, On neutral mesons in cosmic rays. Phys. Rev. 73, 521 (1948)
[Greisen1966] K. Greisen, End to the cosmic-ray spectrum? Phys. Rev. Lett. 16, 748 (1966)
[Grieder1979] P.K.F. Grieder, Average development and properties of the hadronic, muonic and
electromagnetic components in showers of 104 to 108 GeV derived from an all-component-
model calculation, in Proceedings of the 16th International Cosmic Ray Conference, Kyoto, vol.
9 (1979), p. 178
[Grieder2001] P.K.F. Grieder, Cosmic Rays at Earth (Elsevier, Amsterdam, 2001)
[Grieder2009] P.K.F. Grieder, Extensive Air Showers (Elsevier, Amsterdam, 2009)
[Grigorov1961] N.L. Grigorov et al., Artificial Earth Satellites 10, 96 (1961)
[Groom2001] D.E. Groom, N.V. Mokhov, S.I. Striganov, Muon stopping-power and range tables,
10 MeV–100 TeV. At. Data Nucl. Data Tables 78, 183 (2001)
[Grupen2008] C. Grupen, B. Shwartz, Particle Detectors (Cambridge University Press, 2008)
[Guan2015] M. Guan et al., A parametrization of the cosmic ray muon flux at sea level, arXiv:1509.
06176v1 [hep-ex], 21 Sept 2015
[Guillian2007] S. Guillian et al., Observation of the anisotropy of 10 TeV primary cosmic ray nuclei
flux with the Super-Kamiokande-I detector. Phys. Rev. D75, 062003 (2007)
[Guo2018] J. Guo et al., Measurements of Forbush decreases at Mars: both by MSL on ground and
by MAVEN in orbit. Astron. Astrophys. 611, A79 (2018)
[Gutowski2018] B. Gutowski et al., The secret chambers in the Chefren pyramid. Phys. Educ. 53,
045011 (2018)
[Hardcastle2010] M.J. Hardcastle, Which radio galaxies can make the highest energy cosmic rays?
Mon. Not. Roy. Astron. Soc. 405, 2801 (2010)
[Haskin1959] D.M. Haskin et al., Proceedings of the 6th International Cosmic Ray Conference,
Moscow, vol. 3 (1959), p. 123
[Heck1998] D. Heck et al., CORSIKA: A Monte Carlo Code to Simulate Extensive Air Showers,
Report Forschungszentrum Karlsruhe FZKA-6019 (1998)
[Heisenberg1936] W. Heisenberg, On the theory of showers in cosmic radiation. Z. Phys. 101, 533
(1936)
354 Bibliography
[Heisenberg1938] W. Heisenberg, The absorption of the penetrating components of cosmic

radiation. Ann. Phys. 33, 596 (1938)
[Herz1966] A.J. Herz, W.O. Lock, The particle detectors. I. Nuclear emulsions. CERN Courier 6,
83 (1966)
[Hess1911] V.F. Hess, On the absorption of gamma-radiation in the atmosphere. Phys. Z. 12, 998
(1911)
[Hess1912] V.F. Hess, Uber Beobachtungen der durchdringenden Strahlung bei sieben Freiballon-
fahrten. Phys. Z. 13, 1084 (1912)
[Hess1913] V.F. Hess, The origins of penetrating radiation. Phys. Z. 14, 612 (1913)
[Hess1917] V.F. Hess, M. Kofler, Ganzjährige Beobachtungen der durchdringenden Strahlung auf
dem Obir (2044m). Phys. Z. 18, 585 (1917)
[Hess1926] V.F. Hess, Uber den Ursprung der Höhenstrahlung. Phys. Z. 27, 159 (1926)
[Hess1932] V.F. Hess, The cosmic ray observatory on the Hafelekar (2300 m) near Innsbruck
(Austria) and its first results. Terr. Magn. Atmos. Electr. 37, 399 (1932)
[Hess1937] V.F. Hess, A. Demmelmair, Cosmic rays and the aurora of January 25–26. Nature 140,
316 (1937)
[Hess1938] V.F. Hess, R. Steinmaurer, A. Demmelmair, Meteorologische und solare Einflüsse auf
die Intensität der Ultrastrahlung. Nature 141, 686 (1938)
[Hewitt1978] J.E. Hewitt et al., Ames collaborative study of cosmic ray neutrons. Health Phys. 34,
375 (1978)
[Higashi1966] S. Higashi et al., Cosmic ray intensities under sea-water at depths down to 1400 m.
Nuovo Cimento 43A, 334 (1966)
[Hilberry1941] N. Hilberry, Extensive cosmic-ray showers and the energy distribution of primary
cosmic rays. Phys. Rev. 60, 1 (1941)
[HISPARC] Web site of the HISPARC Collaboration: https://www.hisparc.nl/en/
[Hodson1952] A.L. Hodson, The altitude variation of penetrating showers. Proc. Phys. Soc. A65,
702 (1952)
[Hodson1953a] A.L. Hodson, Some aspects of the altitude variation of extensive air showers. Proc.
Phys. Soc. A66, 49 (1953)
[Hodson1953b] A.L. Hodson, Penetrating particles in extensive air showers. Proc. Phys. Soc. A66,
65 (1953)
[Hofstadter1948] R. Hofstadter, Alkali halide scintillation counters. Phys. Rev. 74, 100 (1948)
[Honda1995] M. Honda et al., Calculation of the flux of atmospheric neutrinos. Phys. Rev. D 52,
4985 (1995)
[Horandel2013] J.R. Horandel, Early cosmic ray work published in German. AIP Conference
Proceedings, vol. 1516 (2013), p. 52
[Hsiung1934] D.S. Hsiung, A coincidence test of the corpuscular hypothesis of cosmic rays. Phys.
Rev. 46, 653 (1934)
[Ichimura1993] M. Ichimura et al., Observation of heavy cosmic-ray primaries over the wide energy
range from ∼100 GeV/particle to ∼100 TeV/particle: is the celebrated ‘“knee”’ actually so
prominent? Phys. Rev. D 48, 1949 (1993)
[Ivanov2009] A.A. Ivanov et al., Measuring extensive air showers with Cherenkov light detectors
of the Yakutsk array: the energy spectrum of cosmic rays. New J. Phys. 11, 065008 (2009)
[IZMIRAN] Web site of the Moscow Neutron Monitor: http://cr0.izmiran.ru/mosc/
[JACEE1999] M.L. Cherry (JACEE Collaboration), Where is the bend in the cosmic ray proton
spectrum?, in Proceedings of the 26th International Cosmic Ray Conference, Salt Lake City,
USA, vol. 3 (1999), p. 187
[Janeselli1968] R. Janeselli, Doppia oscillazione diurna della pressione atmosferica e del gradiente
verticale del campo elettromagnetico. Ann. Geophys. 21, 305 (1968)
[Janossy1949] L. Janossy, C.B.A. McCusker, The nature of penetrating particles in air showers.
Nature 163, 182 (1949)
[Jelley1955] J.V. Jelley, Cherenkov radiation and its application. Br. J. Appl. Phys. 6, 227 (1955)
Bibliography 355
[Jelley1965] J.V. Jelley et al., Radio pulses from extensive cosmic ray air showers. Nature 205, 327
(1965)
[Jelley1981] J.V. Jelley, The atmospheric Cherenkov technique in γ-ray astronomy: the early days.
Philos. Trans. Roy. Soc. Lond. A301, 611 (1981)
[Johnson1933a] T.H. Johnson, J. Street, A circuit for recording multiply-coincident discharges of
Geiger-Müller counters. J. Franklin Inst. 215, 239 (1933)
[Johnson1933b] T.H. Johnson, The azimuthal asymmetry of the cosmic radiation. Phys. Rev. 43,
834 (1933)
[Johnson1934] T.H. Johnson, Coincidence counter studies of the corpuscular component of the
cosmic radiation. Phys. Rev. 45, 569 (1934)
[Johnson1935] T.H. Johnson, Progress of the directional survey of the cosmic ray intensities and
its application to the analysis of the primary cosmic radiation. Phys. Rev. 48, 287 (1935)
[Johnson1938] T.H. Johnson, Cosmic ray intensity and geomagnetic effects. Rev. Mod. Phys. 10,
193 (1938)
[Johnson1941] T.H. Johnson, The East-West asymmetry of the cosmic radiation in high latitudes
and the excess of positive mesotrons. Phys. Rev. 59, 11 (1941)
[Johnston1967] M. Johnston (ed.) The Cosmos of Arthur Holly Compton (Knopf, 1967)
[Joliot1931] F. Joliot, Sur l’excitation des rayons γ nucléaires du bore par les particules α. Énergie
quantique du rayonnement γdu polonium. Compte Rendus Séances Academie Science Paris 193,
1415 (1931)
[Jones1965] J.K. Jones, The measurement of the time variation of small showers, in Proceedings
of the 9th International Conference on Cosmic Rays, London, vol. 2 (1965), p. 787
[Jones1991] F.C. Jones, D.C. Ellison, The plasma physics of shock acceleration. Space Sci. Rev.
58, 259 (1991)
[Jones1994] F.C. Jones, A theoretical review of diffusive shock acceleration. Astrophys. J. 90, 561
(1994)
[Jonkmans2013] G. Jonkmans et al., Nuclear waste imaging and spent fuel verification by muon
tomography. Ann. Nucl. Energy 53, 267 (2013)
[Jordan1949] W.H. Jordan, P.R. Bell, Scintillation counters. Nucleonics 5, 30 (1949)
[Jourde2016] K. Jourde et al., Monitoring temporal opacity fluctuations of large structures with
muon tomography: a calibration experiment using a water tower tank. Sci. Rep. 6, 23054 (2016)
[Judge1965] R.J.R. Judge, W.F. Nash, Measurement of the muon flux at various zenith angles.
Nuovo Cimento 35, 999 (1965)
[Kahlert2011] H. Kahlert, H. Krenn, L. Wilmes, Echophysics, the first European Center for the
History of Physics in Poellau, Austria. Europhys. News 42, 28 (2011)
[Kamata1958] K. Kamata, J. Nishimura, The lateral and the angular structure functions of electron
showers. Prog. Theor. Phys. Suppl. 6, 93 (1958)
[Kampert2012] K. Kampert, A. Watson, Extensive air showers and ultra-high-energy cosmic rays:
a historical review. Eur. Phys. J. H37, 359 (2012)
[Karmakar1973] N. Karmakar et al., Measurements of absolute intensities of cosmic-ray muons in
the vertical and greatly inclined directions at geomagnetic latitudes 16 °N. Nuovo Cimento B17,
173 (1973)
[Kedar2012] S. Kedar et al., Low cost, low power, passive muon telescope for interrogating Martian
subsurface. Concepts Approach Mars Expl. 1679, 4150 (2012)
[Kedar2013] S. Kedar et al., Muon radiography for exploration of Mars geology. Geosci. Instrum.
Methods Data Syst. 2, 157 (2013)
[Keil1960] E. Keil, E. Zeitler, W. Zinn, Zur Einfach- und Mehrfachstreuung geladener Teilchen. Z.
Naturforsch. A15, 1031 (1960)
[Keizer2018] F. Keizer et al., A compact, high resolution tracker for cosmic ray muon scattering
tomography using semiconductor sensors. J. Instrum. 13, P10028 (2018)
[Kinoshita1910] S. Kinoshita, The photographic action of the α-particles emitted from radioactive
substance. Proc. Roy. Soc. A83, 432 (1910)
356 Bibliography
[Kinoshita1915] S. Kinoshita, H. Ikeuti, The tracks of alpha particles in sensitive photographic

films. Philos. Mag. 29, 420 (1915)
[Kiraly2013] P. Kiraly, Two centenaries: the discovery of cosmic rays and the birth of Lajos Janossy.
J. Phys: Conf. Ser. 409, 012001 (2013)
[Kirby2011] J. Kirby et al., Role of sulphuric acid, ammonia and galactic cosmic rays in atmospheric
aerosol nucleation. Nature 476, 429 (2011)
[Klinger2015] J. Klinger et al., Simulation of muon radiography for monitoring CO2 stored in a
geological reservoir. Int. J. Greenh. Gas Control 42, 644 (2015)
[Knezevic2019] J. Knezevic et al., Simple coincidence technique for cosmic-ray intensity
exploration via low energy photon detection. Appl. Radiat. Isot. 151, 157 (2019)
[Knoll2000] G.F. Knoll, Radiation Detection and Measurements (Wiley, New York, 2000)
[Kocharian1956] N.M. Kocharian et al., Meson component of the cosmic radiation at an altitude of
3200 m above sea level. Sov. Phys. JETP 3, 350 (1956)
[Kocharian1959] N.M. Kocharian et al., Energy spectra and nuclear interactions of cosmic ray
particles. Sov. Phys. JETP 8, 933 (1959)
[Kolhörster1913a] W. Kolhörster, Uber eine Neukonstruktion des Apparates zur Messung der durch-
dringenden Strahlung nach Wulf und die damit bisher gewonnen Ergebnisse. Phys. Z. 14, 1066
(1913)
[Kolhörster1913b] W. Kolhörster, Messungen der durchdringenden strahlung im freiballon in
grösseren höhen. Phys. Z. 14, 1153 (1913)
[Kolhörster1914] W. Kolhörster, Messungen der durchdringenden Strahlungen bis in Höhen von
9300 m. Verh. Dtsch. Phys. Ges. 16, 719 (1914)
[Kolhörster1926] W. Kolhörster, Bemerkungen zu der Arbeit von R. A. Millikan: Kurzwellige
Strahlen kosmischen Ursprungs. Ann. Phys. 14, 621 (1926)
[Kolhörster1933] W. Kolhörster, The hardest cosmic rays and the electric charge of the Earth. Nature
132, 407 (1933)
[Kolhörster1934] W. Kolhörster, Cosmic rays under 600 m of water. Nature 133, 419 (1934)
[Kolhörster1938] W. Kolhörster et al., Gekoppelte Höhenstrahlen. Naturwissenschaften 26, 576
(1938)
[Korff1952] S. Korff, The world’s high altitude laboratories. Phys. Today 5, 28 (1952)
[Korff1982] S. Korff, High altitude observatories for cosmic rays and other purposes, in
[Sekido1982]
[Korff2013] S. Korff, How the Geiger counter started to crackle: electrical counting methods in
early radioactivity research. Ann. Phys. A88–A92, 525 (2013)
[Kuhn1962] T.S. Kuhn, The Structure of Scientific Revolutions (University of Chicago Press, 1962)
[Kraushaar1962] W.L. Kraushaar, G.W. Clark, Search for primary cosmic gamma rays with the
satellite Explorer XI. Phys. Rev. Lett. 8, 106 (1962)
[Kraybill1948] H.I. Kraybill, Altitude dependence of high energy atmospheric showers. Phys. Rev.
73, 632 (1948)
[Kraybill1949] H.I. Kraybill, Extensive air showers at high altitudes. Phys. Rev. 76, 1092 (1949)
[Kraybill1954a] H.I. Kraybill, Density of extensive air showers at airplane altitudes. Phys. Rev. 93,
1360 (1954)
[Kraybill1954b] H.I. Kraybill, The altitude and angular dependence of cosmic ray showers. Phys.
Rev. 93, 1362 (1954)
[Krebs1941] A. Krebs, Ein Demonstrationsversuch zur Emanationsdiffusion. Ann. Phys. 431, 330
(1941)
[Krebs1955] A.T. Krebs, Early history of the scintillation counter. Science 122, 17 (1955)
[Kudryvtsev2012] A. Kudryvtsev et al., Monitoring subsurface CO2 injection and security of storage
using muon tomography. Int. J. Greenh. Gas Control 11, 21 (2012)
[Kulikov1958] G.V. Kulikov, G.B. Khristiansen, On the size spectrum of extensive air showers.
Sov. Phys. JETP 35, 441 (1959)
[Kuzmin1968] V.A. Kuzmin, G.T. Zatsepin, On cosmic ray interactions with photons. Can. J. Phys.
46, S617 (1968)
Bibliography 357
[LaRocca2005] P. La Rocca, F. Riggi, Analysis of Neutron and Muon Counting During a Forbush
Decrease. Report INFN/AE-05/2, 4 April 2005
[LaRocca2018] P. La Rocca, D. Lo Presti, F. Riggi, Cosmic ray muons as penetrating probes to
explore the world around us, in Cosmic Rays, ed. by Z. Szadkowski (InTechOpen, 2018). https://
doi.org/10.5772/intechopen.75426
[LaRocca2019] P. La Rocca et al. (The EEE Collaboration), Search for coincident air showers over
large-scale distances with the EEE network. Nucl. Phys. B306–308, 175 (2019)
[LaRocca2020a] P. La Rocca et al. (The EEE Collaboration), Scientific and educational aspects of
the EEE Project. J. Phys. G: Conf. Ser. 1561, 012012 (2020)
[LaRocca2020b] P. La Rocca et al. (The EEE Collaboration), Search for long distance time correla-
tions between cosmic air showers with the MRPC telescopes of the EEE network, in Proceedings
of Science POS (EPS-HEP2019), p. 051
[LaRocca2020c] P. La Rocca et al., First commissioning measurements with a reconfigurable cosmic
ray mini-array. J. Instrum. 15, T02003 (2020)
[Lattes1947a] C.M.G. Lattes, G.P.S. Occhialini, C.F. Powell, Observations of the tracks of slow
mesons in photographic emulsions, Part I. Nature 160, 453 (1947)
[Lattes1947b] C.M.G. Lattes, G.P.S. Occhialini, C.F. Powell, Observations of the tracks of slow
mesons in photographic emulsions, Part II. Nature 160, 486 (1947)
[Lawrence1991] M.A. Lawrence, R.J.O. Reid, A.A. Watson, The cosmic ray energy spectrum above
4 × 1017 eV as measured by the Haverah Park array. J. Phys. G17, 733 (1991)
[Lehtipalo2016] K. Lehtipalo et al., The effect of acid-base clustering and ions on the growth of
atmospheric nano-particles. Nat. Commun. 7, 11594 (2016)
[Lemaitre1933] G. Lemaitre, S. Vallarta, On Compton’s latitude effect of cosmic radiation. Phys.
Rev. 43, 87 (1933)
[Lemaitre1936a] G. Lemaitre, S. Vallarta, On the geomagnetic analysis of cosmic radiation. Phys.
Rev. 49, 719 (1936)
[Lemaitre1936b] G. Lemaitre, S. Vallarta, On the allowed cone of cosmic radiation. Phys. Rev. 50,
493 (1936)
[Leo1987] W.R. Leo, Techniques for Nuclear and Particle Physics Experiments (Springer,
Berlin/Heidelberg/New York, 1987)
[Leone2011] M. Leone, Particles that take photographs of themselves: the emergence of the triggered
cloud chamber technique in early 1930s cosmic ray physics. Am. J. Phys. 79, 454 (2011)
[Leroy2004] C. Leroy, P.G. Rancoita, Principles of Radiation Interaction in Matter and Detection
(World Scientific Publishing Company, 2004)
[Lesparre2010] N. Lesparre et al., Geophysical muon imaging: feasibility and limits. Geophys. J.
Int. 183, 1348 (2010)
[Lewis1948] H.V. Lewis, J.B. Oppenheimer, S.A. Wouthuysen, The multiple production of mesons.
Phys. Rev. 73, 127 (1948)
[Libby1932] W.F. Libby, Simple amplifier for Geiger-Müller counter. Phys. Rev. 42, 440 (1932)
[Libby1949a] W.F. Libby et al., Age determination by radiocarbon content: world-wide assay of
natural radiocarbon. Science 109, 227 (1949)
[Libby1949b] J.R. Arnold, W.F. Libby, Age determination by radiocarbon content: checks with
samples of know ages. Science 110, 678 (1949)
[Linke1904] F. Linke, Luftelektrische Messungen bei zwölf Ballonfahrten, Abhandlungen der
Gesellschaft der Wissenschaften in Göttingen. Mathematisch-Physikalische Klasse 3, 1 (1904)
[Linsley1961] J. Linsley, L. Scarsi, B. Rossi, Extremely energetic cosmic ray event. Phys. Rev. Lett.
6, 485 (1961)
[Linsley1963] J. Linsley, Evidence for a primary cosmic ray particle with energy 1020 eV. Phys.
Rev. Lett. 10, 146 (1963)
[Linsley1977] J. Linsley, in Proceedings of the 15th International Conference on Cosmic Rays,
Plovdiv, vol. 8 (1977), p. 207
[Longair2014] M. Longair, C. T. R. Wilson and the cloud chamber. Astropart. Phys. 53, 55 (2014)
358 Bibliography
[Lorenz2012] E. Lorenz, R. Wagner, Very high energy gamma ray astronomy. Eur. Phys. J. H37,
459 (2012)
[LoPresti2018] D. Lo Presti et al., The MEV project: design and testing of a new high-resolution
telescope for muography of Etna Volcano. Nucl. Instrum. Methods Phys. Res. A904, 195 (2018)
[LoPresti2020] D. Lo Presti et al., Muographic monitoring of the volcano-tectonic evolution of
Mount Etna. Sci. Rep. 10, 11351 (2020)
[Lovell1939] A.C.B. Lovell, J.G. Wilson, Investigation of cosmic ray showers of atmospheric origin
using two cloud chambers. Nature 144, 863 (1939)
[Lowder1991] D.M. Lowder et al., Observation of muons using the polar ice cap as a Cherenkov
detector. Nature 144, 863 (1939)
[Maghrabi2021] A. Maghrabi et al., Charged particle detector-related activities of the KACST
radiation laboratory. J. Radiat. Res. Appl. Sci. 14, 111 (2021)
[Mahon2013] D. Mahon et al., A prototype scintillating-fibre tracker for the cosmic-ray muon
tomography of legacy nuclear waste containers. Nucl. Instrum. Methods Phys. Res. A732, 408
(2013)
[Malkov2001] M.A. Malkov, L. O’C Drury, Nonlinear theory of diffusive acceleration of particles
by shock waves. Rep. Prog. Phys. 64, 429 (2001)
[Mandeville1950a] C.E. Mandeville, H.O. Albrecht, Detection of scintillations from crystals with
a photosensitive Geiger-Mueller counter. Phys. Rev. 79, 1010 (1950)
[Mandeville1950b] C.E. Mandeville, H.O. Albrecht, Detection of beta-induced scintillations from
crystals with a photosensitive Geiger-Mueller counter. Phys. Rev. 80, 117 (1950)
[Mandeville1950c] C.E. Mandeville, H.O. Albrecht, Detection of gamma-induced scintillations
from crystals with a photosensitive Geiger-Mueller counter. Phys. Rev. 81, 164 (1950)
[Mandeville1950d] C.E. Mandeville, H.O. Albrecht, Crystals for the scintillation Geiger counter.
Phys. Rev. 81, 163 (1950)
[Mandeville1950e] C.E. Mandeville, M.V. Scherb, Photosensitive Geiger counters: their applica-
tions. Nucleonics 7, 34 (1950)
[Marshall1947a] F.H. Marshall et al., The photomultiplier X-ray detector. Rev. Sci. Instrum. 18,
504 (1947)
[Marshall1947b] F.H. Marshall, J.W. Coltman, The photomultiplier radiation detector. Phys. Rev.
72, 528 (1947)
[Marteau2012] J. Marteau et al., Muon tomography applied to geosciences and volcanology. Nucl.
Instrum. Methods Phys. Res. A695, 23 (2012)
[Maurin2014] D. Maurin et al., A database of charged cosmic rays. Astron. Astrophys. A32, 569
(2014)
[Mayer1913] F. Mayer, Über die Zerstreuung der α-Strahlen. Ann. Phys. 41, 931 (1913)
[Mayneord1950] W.V. Mayneord, E.H. Belcher, Scintillation counting and its medical applications.
Br. Radiol. Suppl. 2–3, 259 (1950)
[Maze1938] R. Maze, Etude d’un appareil a grand pouvoir de resolution pour rayons cosmiques. J.
Phys. Radium 9, 162 (1938)
[Maze1948] R. Maze et al., High altitude measurements on extensive showers. Phys. Rev. 73, 418
(1948)
[McLennan1903a] J.C. McLennan, E.F. Burton, Some experiments on the electrical conductivity
of atmospheric air. Phys. Rev. 16, 184 (1903)
[McLennan1903b] J.C. MacLennan, E.F. Burton, Phys. Z. 4, 553 (1903)
[Meitner1933] L. Meitner, K. Philipp, Die bei neutronenanregung auftretenden elektronenbahnen.
[Menchaca2011] A. Menchaca-Rocha, Searching for cavities in the Teotihuacan Pyramid of the Sun
using cosmic muon experiments and instrumentation, in Proceedings of the 32nd International
Cosmic Ray Conference, Beijing (China), vol. 4 (2011), p. 325
[Mendonca2019] R.R.S. Mendonca et al., Analysis of Cosmic Rays atmospheric effects and their
relationships to cutoff rigidity and zenith angle using global muon detector network data. J.
Geophys. Res. Space Phys. 134, 9791 (2019)
Bibliography 359
[Menzel1948] D.H. Menzel, W.W. Salisbury, The origin of cosmic rays. Nucleonics 2, 67 (1948)
[Meyer1961] P. Meyer, R. Vogt, Electrons in the primary cosmic radiation. Phys. Rev. Lett. 6, 193
(1961)
[Meyer1970] B.S. Meyer et al., Cosmic ray muon intensity deep underground versus depth. Phys.
Rev. D 1, 2229 (1970)
[Michl1912] W. Michl, Akad Wiss. Wien 121, 1431 (1912)
[Millikan1923] R.A. Millikan, I.S. Bowen, Penetrating radiation at high altitudes. Phys. Rev. 22,
198 (1923)
[Millikan1925] R.A. Millikan, High frequency rays of cosmic origin. Nature 116, 823 (1925)
[Millikan1926a] R.A. Millikan, I.S. Bowen, High frequency rays of cosmic origin. I. Sounding
balloon observations at extreme altitude. Phys. Rev. 27, 353 (1926)
[Millikan1926b] R.A. Millikan, R.M. Otis, High frequency rays of cosmic origin. II. Mountain peak
and airplane observations. Phys. Rev. 27, 645 (1926)
[Millikan1926c] R.A. Millikan, G.H. Cameron, High frequency rays of cosmic origin. III.
Measurements in snow-fed lakes at high altitudes. Phys. Rev, 28, 851 (1926)
[Millikan1928a] R.A. Millikan, G.H. Cameron, High altitude tests on the geographical, directional
and spectral distribution of cosmic rays. Phys. Rev. 31, 163 (1928)
[Millikan1928b] R.A. Millikan, G.H. Cameron, The origin of the cosmic rays. Phys. Rev. 32, 533
(1928)
[Millikan1928c] R.A. Millikan, G.H. Cameron, New precision in cosmic ray measurements,
yielding extension of spectrum and indication of bands. Phys. Rev. 31, 921 (1928)
[Millikan1930a] R.A. Millikan, History of research in cosmic rays. Nature 126, 14 (1930)
[Millikan1930b] R.A. Millikan, On the question of the constancy of the cosmic radiation and the
relation of these rays to meteorology. Phys. Rev. 36, 1595 (1930)
[Millikan1934] I.S. Bowen, R.A. Millikan, H.V. Neher, A very high altitude survey of the effect
of latitude upon cosmic-ray intensities and an attempt at a general interpretation of cosmic-ray
phenomena. Phys. Rev. 46, 641 (1934)
[Millikan1935a] R.A. Millikan, Electrons, Protons, Photons, Neutrons and Cosmic Rays (Chicago,
1935)
[Millikan1935b] R.A. Millikan, H.V. Neher, The equatorial longitude effect in cosmic rays. Phys.
Rev. 47, 205 (1935)
[Millikan1936] R.A. Millikan, H.V. Neher, A precision world survey of sea level cosmic ray
intensities. Phys. Rev. 50, 15 (1936)
[Millikan1938a] I.S. Bowen, R.A. Millikan, H.V. Neher, New evidence as to the nature of the
incoming cosmic rays, their absorbability in the atmosphere, and the secondary character of the
penetrating rays found in such abundance at sea level and below. Phys. Rev. 53, 217 (1938)
[Millikan1938b] I.S. Bowen, R.A. Millikan, H.V. Neher, New light on the nature and origin of the
incoming cosmic rays. Phys. Rev. 53, 855 (1938)
[Miyake1963] S. Miyake, Empirical formula for range spectrum of cosmic ray μ-mesons at sea
level. J. Phys. Soc. Jpn. 18, 1093 (1963)
[Miyake1973] S. Miyake, Underground muon angular distribution, in Proceedings of the 13th
International Conference on Cosmic Ray Physics, Denver (USA), vol. 5 (1973), p. 3638
[Mizutani1979] K. Mizutani et al., Energy spectrum of cosmic rays deep underground, in
Proceedings of the 16th International Cosmic Ray Conference, Tokyo, vol. 10 (1979), p. 40
[Mohanty2016] P.K. Mohanty et al., Transient weakening of Earth’s magnetic shield probed by a
cosmic ray burst. Phys. Rev. Lett. 117, 171101 (2016)
[Montgomery1936] C.G. Montgomery, D.D. Montgomery, W.E. Ramsey, W.F.G. Swann, A search
for protons in the primary cosmic-ray beam. Phys. Rev. 50, 403 (1936)
[Morales2019] J.A. Morales-Soto et al., The lateral distribution function of cosmic-ray induced air
showers studied with the HAWC observatory. Proceedings of the 36th International Cosmic Ray
Conference, Madison (USA), POS (ICRC2019), p. 358
[Morishima2013] K. Morishima et al., Development of an automated nuclear emulsion analyzing
system. Radiat. Meas. 50, 237 (2013)
360 Bibliography
[Morishima2017] K. Morishima et al., Discovery of a big void in Khufu’s pyramid by observation

of cosmic-ray muons. Nature 552, 386 (2017)
[Morrison1954] P. Morrison, S. Olbert, B. Rossi, The origin of cosmic rays. Phys. Rev. 94, 440
(1954)
[Morton1949] G.A. Morton, J.A. Mitchell, Performance of 931-A type multiplier as a scintillation
counter. Nucleonics 4, 6 (1949)
[MPI] https://www.mpi-hd.mpg.de/hfm/CosmicRay/CosmicRaySites.html
[Mrdja2016] D. Mrdja et al., First cosmic ray images of bones and soft tissues. Europhys. Lett. 116,
48003 (2016)
[Murdoch1960] H.S. Murdoch et al., The Sydney cosmic ray spectrometer, in Proceedings of the
6th International Cosmic Ray Conference, Moscow, vol. 1 (1960), p. 304
[Miyazaki1949] Y. Miyazaki, Cosmic rays at a great depth. Phys. Rev. 76, 1733 (1949)
[Myssowsky1925] L. Myssowsky, L. Tuwim, Versuche über die Absorption der Höhenstrahlung
im Wasser. Z. Phys. 35, 299 (1926)
[Myssowsky1926] L. Myssowsky, L. Tuwim, Unregelmäßige Intensitätsschwankungen der Höhen-
strahlung in geringer Seehöhe. Z. Phys. 39, 146 (1926)
[Myssowsky1927] L. Myssowsky, L. Tuwim, Absorptionskurve der Höhenstrhlung im Wasser. Z.
Phys. 44, 369 (1927)
[Myssowsky1928] L. Myssowsky, L. Tuwim, Absorption in Blei, sekundäre Strahlen und Wellen-
länge der Höhenstrahlung. Z. Phys. 50, 273 (1928)
[Nagamine1995] K. Nagamine et al., Method of probing inner structure of geophysical substance
with the horizontal cosmic ray muons and possible application to volcanic eruption prediction.
[NALTA] Web site of the NALTA Collaboration: http://neutrino.phys.washington.edu/~walta/
NALTA.html
[Nash1956] W.F. Nash, A.J. Pointon, The momentum spectrum of μ-mesons at a depth of 40 m
water equivalent. Proc. Phys. Soc. A69, 725 (1956)
[Neddermayer1937] S.H. Neddermayer, C.D. Anderson, Note on the nature of cosmic ray particles.
Phys. Rev. 51, 884 (1937)
[Neddermayer1938] S.H. Neddermayer, C.D. Anderson, Cosmic ray particles of intermediate mass.
Phys. Rev. 54, 88 (1938)
[Neher1940] H.V. Neher, W.H. Pickering, An attempt to measure the latitude effect of extensive
cosmic ray showers. Phys. Rev. 58, 665 (1940)
[Nesvizhevsky2017] V. Nesvizhevsky, J. Villain, The discovery of the neutron and its consequences
(1930–1940). C. R. Phys. 18, 592 (2017)
[Niederleithinger2021] E. Niederleithinger et al., Muon tomography of the interior of a reinforced
concrete block: first experimental proof of concept. J. Nondestr. Eval. 40, 65 (2021)
[Nishimura2013] J. Nishimura, The birth of cosmic ray work in Japan. AIP Conf. Proc. 1516, 25
(2013)
[Nishina1937] Y. Nishina, M. Takeuchi, T. Ichimiya, On the nature of cosmic-ray particles. Phys.
Rev. 52, 1198 (1937)
[Nishina1941] Y. Nishina et al., Cosmic rays at a depth equivalent to 1400 meters of water. Phys.
Rev. 59, 401 (1941)
[NIST] https://www.nist.gov/pml/x-ray-and-gamma-ray-data
[NOAA] National Oceanic and Atmospheric Administration (NOAA) Web site: ngdc.noaa.gov
[NRIC] National Radiation Instrument Catalog, http://www.national-radiation-instrument-catalog.
com/
[Ochi2003] N. Ochi et al., The status and future prospect of the LAAS project, in Proceedings of the
28th International Conference on Cosmic Ray Physics, Tsukuba (Japan), vol. 2 (2003), p. 1005
[Olah2018] L. Olah et al., High-definition and low noise muography of the Sakurajima volcano
with gaseous tracking detectors. Sci. Rep. 8, 1 (2018)
[Oppenheimer1933] J.R. Oppehneimer, M.S. Plesset, On the production of the positive electron.
Phys. Rev. 43, 53 (1933)
Bibliography 361
[ORAU] Oak Ridge Associated Universities Museum of Radiation and Radioactivity, https://www.
orau.org/health-physics-museum/index.html
[Otis1923a] R.M. Otis, The variation of penetrating radiation with altitude. Phys. Rev. 22, 198
(1923)
[Otis1923b] R.M. Otis, The penetrating radiation on Mt. Whitney. Phys. Rev. 22, 199 (1923)
[Otis1924] R.M. Otis, R.A. Millikan, The source of penetrating radiation found in the earth’s
atmosphere. Phys. Rev. 23, 778 (1924)
[OULU] Web site of the Oulu Neutron Monitor: http://cosmicrays.oulu.fi/
[Pacini1910] D. Pacini, Penetrating radiation on the sea. Le Radium 8, 307 (1910)
[Pacini1912] D. Pacini, Penetrating radiation at the surface of and in water. Nuovo Cimento 6, 93
(1912)
[Pal2012] S. Pal et al., Measurement of the integrated flux of cosmic ray muons at sea level using
the INO-ICAL prototype detector. J. Cosmol. Astropart. Phys. 07, 033 (2012)
[PDG] The Review of Particle Physics, R.L. Workman et al. (Particle Data Group), Prog. Theor.
Exp. Phys. 083C01 (2022). PDG Web Site: pdg.lbl.gov
[Penzias1965] A.A. Penzias, R.W. Wilson, A measurement of the excess antenna temperature at
4080 Mc/s. Astrophys J 142, 419 (1965)
[Perry2013] J. Perry et al., Imaging a nuclear reactor using cosmic ray muons. J. Appl. Phys. 113,
184909 (2013)
[Peruzzi2007] G. Peruzzi, S. Talas, The Italian contributions to cosmic ray physics from Bruno
Rossi to the G-stack. A new window into the inexhaustible wealth of nature. La Rivista del
[Pesente2009] S. Pesente et al., First results on material identification and imaging with a large-
volume muon tomography prototype. Nucl. Instrum. Methods Phys. Res. A604, 738 (2009)
[Pethurai2017] S. Pethuraj et al., Measurement of cosmic muon angular distribution and vertical
integrated flux by 2m × 2 m RPC stack at IICHEP-Madhurai. J. Cosmol. Astropart. Phys. 9, 021
(2017)
[Piccard1932] A. Piccard, E. Stahel, P. Kipfer, Messung der Ultrastrahlung in 16000 m Höhe.
[Pinto2020a] C. Pinto et al., A low cost reconfigurable mini-array facility for (under) graduate
studies in cosmic ray physics. Proc. Sci. (POS-HEPS2019) 438 (2020)
[Pinto2020b] C. Pinto et al. (The EEE Collaboration), Measurements with cosmic muons to monitor
the stability of a civil building on a long time-scale. J. Instrum. 15, C03058 (2020)
[Pfotzer1936] G. Pfotzer, Dreifachkoinzidenzen der Ultrastrahlung aus vertikaler Richtung in der
StratosphÄre. Z. Phys. 102, 23 (1966)
[Pfotzer1972] G. Pfotzer, History of the use of balloons in scientific experiments. Space Sci. Rev.
13, 199 (1972)
[Pfotzer1982] G. Pfotzer, Early evolution of coincidence counting, a fundamental method in cosmic
ray physics, in [Sekido1982]
[Poluianov2017] S.V. Poluianov et al., GLE and sub-GLE redefinition in the light of high-altitude
polar neutron monitors. Sol. Phys. 292, 176 (2017)
[Porter1958] N.A. Porter et al., Observations on extensive air showers VII. The lateral distribution
of energy in the electron-photon component. Philos. Mag. 3, 826 (1958)
[Poulson2017] D. Poulson et al., Cosmic ray muon computed tomography of spent nuclear fuel in
dry storage casks. Nucl. Instrum. Methods Phys. Res. A842, 48 (2017)
[Powell1950] C.F. Powell, Mesons. Rep. Prog. Phys. 13, 350 (1950)
[Powell1959] C.F. Powell, P.H. Fowler, D.H. Perkins, The Study of Elementary Particles by the
Photographic Method (Pergamon Press, New York, 1959)
[Pringle1950] R.W. Pringle, The scintillation counter. Nature 166, 11 (1950)
[Procureur2023] S. Procureur et al., Precise characterization of a corridor-shaped structure in
Khufu’s Pyramid by observation of cosmic-ray muons. Nature Comm. 14, 1144 (2023)
[Pryke2001] C.L. Pryke, A comparative study of the depth of maximum of simulated air shower
longitudinal profiles. Astropart. Phys. 14, 319 (2001)
362 Bibliography
[Pyle1999] R. Pyle et al., The use of 3 He tubes in a neutron monitor latitude survey, in Proceedings
of the 26th International Cosmic Ray Conference, Salt Lake City (USA), vol. 7 (1999), p. 386
[RamanaMurthy1972] P.V. Ramana Murthy, A. Subramanian, Test of scale invariance in pion
production at high energies using cosmic ray primary nucleon and sea level muon intensities.
Phys. Lett. B39, 646 (1972)
[Rasetti1941] F. Rasetti, Disintegration of slow mesotrons. Phys. Rev. 60, 198 (1941)
[Ravel2013] O. Ravel, Early cosmic ray research in France. AIP Conf. Proc. 1516, 67 (2013)
[Regener1932a] E. Regener, Über das Spektrum der Ultrastrahlung. Z. Phys. 74, 433 (1932)
[Regener1932b] E. Regener, Intensity of cosmic radiation in the high atmosphere. Nature 130, 364
(1932)
[Regener1932c] E. Regener, Messung der Ultrastrahlung in der Stratosphäre. Naturwissenschaften
20, 695 (1932)
[Regener1933] E. Regener, Die Absorptionskurve der Ultrastrahlung und ihre Deutung. Phys. Z.
34, 306 (1933)
[Reidy1978] J.J. Reidy et al., Use of muonic X rays for nondestructive analysis of bulk samples for
low Z constituents. Anal. Chem. 50, 40 (1978)
[Reinganum1911] M. Reinganum, Streuung und photographische Wirkung der α-Strahlen. Phys.
Z. 12, 1076 (1911)
[Riadigos2020] I. Riadigos et al., Atmospheric temperature effects in secondary cosmic rays
observed with a 2 m2 ground based tPRC detector. Earth Space Sci. 7, 1131 (2020)
[Richtmeyer1949] R.D. Richtmeyer, E. Teller, On the origin of cosmic rays. Phys. Rev. 75, 1729
(1949)
[Riggi2010] S. Riggi et al., GEANT4 simulation of plastic scintillator strips with embedded optical
fibers for a prototype of tomographic system. Nucl. Instrum. Methods Phys. Res. A624, 583
(2010)
[Riggi2014] F. Riggi et al., An extensive air shower trigger station for the Muon Portal detector.
[Riggi2017] F. Riggi et al. (The EEE Collaboration), Time and orientation long-distance correlations
between extensive air showers detected by the MRPC telescopes of the EEE Project. Il Nuovo
Cimento C40, 196 (2017)
[Riggi2018] F. Riggi et al., The Muon Portal project: commissioning of the full detector and first
results. Nucl. Instrum. Methods Phys. Res. A912, 16 (2018)
[Riggi2020] F. Riggi et al., Investigation of the cosmic ray angular distribution and the East-West
effect near the top of Etna volcano with the MEV telescope. Eur. Phys. J. Plus 135, 280 (2020)
[Riggi2021a] F. Riggi et al., A modular telescope facility to investigate the cosmic ray decoherence
curve. J. Instrum. 16, T08006 (2021)
[Riggi2021b] F. Riggi et al., Multiparametric approach to the assessment of muon tomographic
results for the inspection of a full-scale container. Eur. Phys. J. Plus 136, 139 (2021)
[Rogers1984] I.W. Rogers, M. Tristam, The absolute depth-intensity curve for cosmic ray muons
underwater and the integral sea-level momentum spectrum in the range 1–100 GeV/c. J. Phys.
G: Nucl. Part. Phys. 10, 983 (1984)
[Rossi1930a] B. Rossi, Method of registering multiple simultaneous impulses of several Geiger
counters. Nature 125, 636 (1930)
[Rossi1930b] B. Rossi, On the magnetic deflection of cosmic rays. Phys. Rev. 36, 606 (1930)
[Rossi1931a] B. Rossi, Magnetic experiments on the cosmic rays. Nature 128, 300 (1931)
[Rossi1931b] B. Rossi, Measurements on the absorption of the penetrating corpuscular rays coming
from inclined directions. Nature 128, 408 (1931)
[Rossi1931c] B. Rossi, Ricerche sull’azione del campo magnetico terrestre sopra i corpuscoli della
radiazione penetrante. Nuovo Cimento 8, 85 (1931)
[Rossi1932a] B. Rossi, Absorptionsmessungen der durchdringenden Korpuskularstrahlung in einem
Meter Blei. Naturwissenschaften 20, 65 (1932)
[Rossi1932b] B. Rossi, Nachweis einer Sekundärstrahlung der durchdringenden Korpusku-
larstrahlung. Phys. Z. 33, 304 (1932)
Bibliography 363
[Rossi1932c] B. Rossi, Ricerche sulla radiazione secondaria della radiazione corpuscolare pene-
trante. Ric. Sci. 3, 234 (1932)
[Rossi1932d] B. Rossi, La curva di assorbimento della radiazione corpuscolare penetrante. Ric.
Sci. 3, 435 (1932)
[Rossi1933a] B. Rossi, Über die Eigenschaften der durchdringenden Korpuskularstrahlung in
Meeresniveau. Z. Phys. 82, 151 (1933)
[Rossi1933b] B. Rossi, I risultati della missione scientifica in Eritrea per lo studio dei raggi cosmici.
Ric. Sci. 4, 365 (1933)
[Rossi1934a] B. Rossi, Directional measurements on the cosmic rays near the geomagnetic equator.
Phys. Rev. 45, 212 (1934)
[Rossi1934b] B. Rossi, Misure della distribuzione angolare di intensità della radiazione penetrante
all’Asmara. La Ricerca Scientifica Supplement 1, 579 (1934)
[Rossi1941] B. Rossi, D.B. Hall, Variation of the rate of decay of mesotrons with momentum. Phys.
Rev. 59, 223 (1941)
[Rossi1942] B. Rossi, N. Nereson, Experimental determination of the disintegration curve of
mesotrons. Phys. Rev. 62, 417 (1942)
[Rossi1943] N. Nereson, B. Rossi, Further measurements on the disintegration curve of mesotrons.
Phys. Rev. 64, 199 (1943)
[Rossi1964] B. Rossi, Cosmic Rays (McGraw-Hill, New York, 1964)
[Rossi1981] B. Rossi, Early days in cosmic rays. Phys. Today 34, 34 (1981)
[Rossi1990] B. Rossi, Moments in the Life of a Scientist (Cambridge University Press, New York,
1990)
[Rutherford1903] E. Rutherford, H.L. Cooke, A penetrating radiation from the earth’s surface. Phys.
Rev. 16, 183 (1903)
[Rutherford1908] E. Rutherford, H. Geiger, An electrical method of counting the number of α
particles from radioactive substances. Proc. Roy. Soc. (Lond.) A81, 141 (1908)
[Ryan1979] J.M. Ryan et al., Atmospheric gamma ray angle and energy distribution from sea level
to 3.5 g/cm2 and 2 to 25 MeV. J. Geophys. Res. 84, 5279 (1979)
[Rypdal1997] K. Rypdal, T. Brundtland, The Birkeland Terrella experiments and their importance
for the modern synergy of laboratory and space plasma physics. J. Phys. 7, 113 (1997)
[Sandström1965] A.E. Sandström, Cosmic ray physics (North Holland, Amsterdam, 1965)
[Sahni1915] R.R. Sahni, The photographic action of alpha, beta and gamma rays. Philos. Mag. 29,
836 (1915)
[Saracino2017] G. Saracino et al., Imaging of underground cavities with cosmic ray muons from
observations at Mt. Echia (Naples). Sci. Rep. 7, 1181 (2017)
[Saraiva2020] J.P. Saraiva et al., The TRISTAN detector—2018–2019 latitude survey of cosmic
rays. J. Instrum. 15, C09024 (2020)
[Scarsi2005] L. Scarsi, Bruno Rossi and the school of Arcetri. Int. J. Mod. Phys. A20, 6539 (2005)
[Schein1941] M. Schein et al., The nature of the primary cosmic radiation and the origin of the
mesotron. Phys. Rev. 59, 615 (1941)
[Schmeiser1938] K. Schmeiser, W. Bothe, Die harten Ultrastrahlschauer. Ann. Phys. 424, 161
(1938)
[Schönfelder1980] V. Schönfelder et al., The vertical component of 1–20 MeV gamma rays at
balloon altitudes. Astrophys. J. 240, 350 (1980)
[Schopper1967] E. Schopper et al., Handbuch der Physik. Kosmische Strahlung XLVI(2), 372
(1967)
[Schröder2017] F.G. Schröder, Radio detection of cosmic ray air showers and high energy neutrinos.
Prog. Part. Nucl. Phys. 93, 1 (2017)
[Schultz2004] L.J. Schultz et al., Image reconstruction and material Z discrimination via cosmic
ray muon radiography. Nucl. Instrum. Methods Phys. Res. A519, 687 (2004)
[Schultz2007] L.J. Schultz et al., Statistical reconstruction for cosmic ray muon tomography. IEEE
Trans. Image Process. 16, 1985 (2007)
364 Bibliography
[Schuster2014] P.M. Schuster, The scientific life of Victor Franz Hess. Astropart. Phys. 53, 33
(2014)
[Sekido1982] Y. Sekido, H. Elliott (eds.), Early History of Cosmic Study (D. Reidel Publishing
Company, 1982)
[Setter1939] G. Setter, H. Wambacher, Z. Phys. 40, 702 (1939)
[Shamos1966] M.H. Shamos et al., A new measurement of the intensity of cosmic ray ionization
at sea level. J. Geophys. Res. 71, 4651 (1966)
[Shapiro1941] M. Shapiro, Tracks of nuclear particles in photographic emulsions. Rev. Mod. Phys.
13, 58 (1941)
[Shapiro1970] M.M. Shapiro, R. Silberberg, Heavy cosmic ray nuclei. Annu. Rev. Nucl. Sci. 20,
323 (1970)
[Sheldon1963] W.R. Sheldon, N.M. Duller, Intensity of muons underground at large zenith angles.
[Sime2012] R.L. Sime, Marietta Blau in the history of cosmic rays. Phys. Today 65, 8 (2012)
[Sime2013] R.L. Sime, Marietta Blau, Pioneer of photographic nuclear emulsions and particle
physics. Phys. Perspect. 15, 3 (2013)
[Skobeltsyn1927] D.V. Skobeltsyn, Die Intensitätsverteilung in dem Spektrum der γ-Strahlen von
RaC. Z. Phys. 43, 354 (1927)
[Skobeltsyn1929] D.V. Skobeltsyn, Uber eine neue Art sehr schneller β-Strahlen. Z. Phys. 54, 686
(1929)
[Skobeltsyn1947] D.V. Skobeltsyn, G.T. Zatsepin, V.V. Miller, The lateral extension of Auger
showers. Phys. Rev. 71, 315 (1947)
[Smart2005] D.F. Smart, M.A. Shea, A review of geomagnetic cutoff rigidities for earth-orbiting
spacecraft. Adv. Space Res. 36, 2012 (2005)
[Snyder1949] H.S. Snyder, W.T. Scott, Multiple scattering of fast charged particles. Phys. Rev. 76,
220 (1949)
[SPACEWEATHER] https://www.spaceweather.com
[Spiering2012] C. Spiering, Towards high-energy neutrino astronomy. Eur. Phys. J. H37, 515 (2012)
[Spillantini2013] P. Spillantini, Early cosmic ray research in Italy. AIP Conf. Proc. 1516, 61 (2013)
[Stecker1968] F.W. Stecker, Effect of photomeson production by the universal radiation field on
high energy cosmic rays. Phys. Rev. Lett. 21, 1016 (1968)
[Steinke1930] E. Steinke, Über Schwankungen und Barometereffekt der kosmischen Ultrastrahlung
im Meeresniveau. Z. Phys. 64, 48 (1930)
[Steinmaurer1982] R. Steinmaurer, Erinnerungen an V. F. Hess, den Entdecker der Kosmischen
Strahlung, und an die Ersten Jahre des Betriebes des Hafeleakr-Labors, in [Sekido1982]
[Stevenson1935] E.C. Stevenson, T.H. Johnson, Coincidence counter studies of the variation of
intensities of cosmic ray showers and vertical rays with barometric pressure. Phys. Rev. 47, 578
(1935)
[Størmer1907] C. Størmer, Sur les trajectoires des corpuscules électrisés dans l’espace sous l’action
du magnétisme terrestre avec application aux aurores boréales. Arch. Sci. Phys. Nat. Genève 24,
5, 113, 221, 317 (1907)
[Størmer1930] C. Størmer, Periodische Elektronenbahnen im Felde eines Elementarmagneten
und ihre Anwendung auf Brüches Modellversuche und auf Eschenhagens Elementarwellen des
Erdmagnetismus. Mit 32 Abbildungen. Z. Astrophys. 1, 237 (1930)
[Størmer1955] C. Størmer, The Polar Aurora (Oxford University Press, 1955)
[STRATOCAT] https://stratocat.com.ar/indexe.html
[Street1935] J.C. Street, R.H. Woodward, E.C. Stevenson, Phys. Rev. 47, 891 (1935)
[Street1937] J.C. Street, E.C. Stevenson, New evidence for the existence of a particle of mass
intermediate between the proton and the electron. Phys. Rev. 52, 1003 (1937)
[Strohmayer2006] B. Strohmayer, Marietta Blau, Stars of Disintegration: Biography of a Pioneer
of Particle Physics (Ariadne, 2006)
[Strong2010] A.W. Strong et al., Global cosmic-ray-related luminosity and energy budget of the
Milky Way. Astrophys. J. 722, L58 (2010)
Bibliography 365
[Suga1962] K. Suga et al., Bolivian Air Shower Joint Experiment. J. Phys. Soc. Jpn. 17(Suppl.
AIII) (1962)
[Swanenburg1981] B.N. Swanenburg et al., Second COS B catalog of high energy gamma-ray
sources. Astrophys. J. 243, L69 (1981)
[Swann1933] W.F.G. Swann, Applications of Liouville’s theorem to electron orbits in the Earth
magnetic field. Phys. Rev. 44, 224 (1933)
[Swann1935] W.F.G. Swann, The corpuscular theory of the primary cosmic radiation. Phys. Rev.
48, 641 (1935)
[Swann1936] W.F.G. Swann, Can protons represent the primary cosmic radiation at sea level? Phys.
Rev. 49, 478 (1936)
[Swann1939a] W.F.G. Swann, The nature of cosmic ray phenomena at high altitudes. Phys. Rev.
56, 209 (1939)
[Swann1939b] W.F.G. Swann, Showers produced by penetrating rays and allied phenomena. Rev.
Mod. Phys. 11, 242 (1939)
[Swann1940] W.F.G. Swann, A theory of cosmic-ray phenomena based upon a single primary
component. Phys. Rev. 58, 200 (1940)
[Swann1941a] W.F.G. Swann, A single component for the primary cosmic radiation. Phys. Rev. 59,
770 (1941)
[Swann1941b] W.F.G. Swann, Further evidence for a single component in the primary cosmic
radiation. Phys. Rev. 59, 836 (1941)
[Swann1954] W.F.G. Swann, On H. Alfvén’s theory of the effect of magnetic storms on cosmic ray
intensity. Phys. Rev. 93, 905 (1954)
[Swann1961] W.F.G. Swann, History of cosmic rays. Am. J. Phys. 29, 811 (1961)
[Swider1967] W. Swider, M.E. Gardner, Environmental Research Paper No. 272, Air Force
Cambridge Research, Bedford, MA (1967)
[TA] Telescope Array Web site: http://www.telescopearray.org/
[Tanaka2001] H.K.M. Tanaka et al., Development of the cosmic ray muon detection system for
probing the internal structure of a volcano. Hyperfine Interact. 138, 521 (2001)
[Tanaka2007a] H.K.M. Tanaka et al., Imaging the conduit size of the dome with cosmic ray muons:
the structure beneath Showa-Shinzan lava dome. Geophys. Res. Lett. 10, L031389 (2007)
[Tanaka2007b] H.K.M. Tanaka, Monte-Carlo simulations of atmospheric muon production:
implication of the past Martian environment. Icarus 19, 603 (2007)
[Tanaka2014] H.K.M. Tanaka et al., Radiographic visualization of magma dynamics in an erupting
volcano. Nat. Commun. 5, 3381 (2014)
[Tanaka2016] H.K.M. Tanaka et al., Instant snapshot of the internal structure of Unzen lava dome
with airborne muograhy. Sci. Rep. 6, 39741 (2016)
[Tanaka2022] H.K.M. Tanaka, Cosmic time synchronizers (CTS) for wireless and precise time
synchronization using extended air showers. Sci. Rep. 12, 7078 (2022)
[Teller1954] E. Teller, Theory of origin of cosmic rays. Rep. Prog. Phys. 17, 154 (1954)
[Tennent1967] R.M. Tennent, The Haverah Park extensive air shower array. Proc. Phys. Soc. 92,
622 (1967)
[Terada2015] K. Terada et al., A new X-ray fluorescence spectroscopy for extraterrestrial materials
using a muon beam. Sci. Rep. 4, 5072 (2015)
[TerHaar1950] D. Ter Haar, Cosmogonical problems and stellar energy. Rev. Mod. Phys. 22, 119
(1950)
[Thompson1974] D.J. Thompson, A three-dimensional study of 30 to 300 MeV atmospheric gamma
rays. J. Geophys. Res. 79, 1309 (1974)
[Thompson2012] L. Thompson, The Discovery of Air-Cherenkov Radiation, CERN Courier, 18
July 2012
[Tioukov2019] V. Tioukov et al., First muography of Stromboli volcano. Sci. Rep. 9, 1 (2019)
[Tongiorgi1948a] V. Tongiorgi Cocconi, On the presence of neutrons in the extensive air showers.
Phys. Rev. 73, 923 (1948)
366 Bibliography
[Tongiorgi1948b] V. Tongiorgi Cocconi, On the origin of the neutrons associated with the extensive
air showers. Phys. Rev. 74, 226 (1948)
[Tonwar2013] S.C. Tonwar, Cosmic ray research in India, 1912–2012. AIP Conf. Proc. 1516, 72
(2013)
[Treat1948] J.E. Treat, K.I. Greisen, Penetrating particles in extensive air showers. Phys. Rev. 74,
414 (1948)
[Valat2020] D. Valat, J. Delporte, Absolute calibration of timing receiver chains at the nanosecond
uncertainty level for GNSS time scales monitoring. Metrologia 57, 025019 (2020)
[Vavilov1970] Yu.N. Vavilov et al., Bulletin of the Russian Academy of Science. Phys. Ser. 34,
1759 (1970)
[Vernov1967] S.N. Vernov et al., Radio emission of extensive air showers of cosmic rays. JETP
Lett. 5, 127 (1967)
[Walker1981] R. Walker, A.A. Watson, Measurement of the elongation rate of extensive air showers
produced by primary cosmic rays of energy above 2 × 1017 eV. J. Phys. G: Nucl. Phys. 7, 1297
(1981)
[Walter2012] M. Walter, A.W. Wolfendale, Early history of cosmic particle physics. Eur. J. Phys.
H37, 323 (2012)
[Watase1937] Y. Watase, Cosmic ray showers. Nature 139, 671 (1937)
[Watson2010] A. Watson, The discovery of the Cherenkov radiation and its use in the detection
of extensive air showers, in Proceedings of the Cosmic Ray International Seminars (CRIS2010),
Catania, Nucl. Phys. Proc. Suppl. B212–213, 13 (2010)
[WDC] World Data Center for Geomagnetism, Kyoto, https://wdc.kugi.kyoto-u.ac.jp/index.html
[Webber1974] W.R. Webber, J.A. Lezniak, The comparative spectra of cosmic ray protons and
Helium nuclei. Astrophys. Space Sci. 30, 361 (1974)
[Wibberenz1962] G. Wibberenz, Identifizierung von Protonen, Müonen und Elektronen in der
weichen Komponente der Ultrastrahlung und Messung ihrer Energiespektren. Z. Phys. 167, 284
(1962)
[Williams1940] E. Williams, G. Roberts, Evidence for transformation of mesotrons into electrons.
Nature 145, 102 (1940)
[Williams1948] E. Williams, The structure of large cosmic ray air showers. Phys. Rev. 74, 1689
(1948)
[Wilson1896] C.T.R. Wilson, Proc. R. Soc. Lond. 59, 338 (1896)
[Wilson1897] C.T.R. Wilson, Condensation of water vapour in the presence of dust-free air and
other gases. Philos. Trans. 189, 265 (1897)
[Wilson1898] C.T.R. Wilson, On the condensation nuclei produced in gases by the action of
Roentgen rays, uranium rays, ultra-violet light and other agents. Proc. R. Soc. Lond. 64, 127
(1898)
[Wilson1899] C.T.R. Wilson, On the comparative efficiency as condensation nuclei of positively
and negatively charged ions. Proc. R. Soc. Lond. 65, 289 (1899)
[Wilson1900a] C.T.R. Wilson, On the leakage of electricity through dust-free air. Proc. Camb.
Philos. Soc. 11, 32 (1900)
[Wilson1900b] C.T.R. Wilson, Atmospheric electricity. Nature 62, 149 (1900)
[Wilson1901a] C.T.R. Wilson, On the ionisation of atmospheric air. Proc. R. Soc. Lond. A68, 151
(1901)
[Wilson1901b] C.T.R. Wilson, On the spontaneous ionisation of gases. Proc. R. Soc. Lond. A69,
277 (1901)
[Wilson1912] C.T.R. Wilson, On an expansion apparatus for making visible the tracks of ionising
particles in gases and some results obtained by its use. Philos. Trans. R. Soc. A87, 277 (1912)
[Wilson1943] V.C. Wilson, D.J. Hughes, Cloud chamber and counter studies of cosmic rays
underground. Phys. Rev. 63, 161 (1943)
[Winckler1950] J.R. Winckler et al., A directional and latitude survey of cosmic rays at high altitude.
Phys. Rev. 79, 656 (1950)
Bibliography 367
[Winn1986] M.M. Winn et al., The cosmic-ray energy spectrum above 1017 eV. J. Phys. G: Nucl.
Part. Phys. 12, 653 (1986)
[Wolferton2019] M. Wolferton, How the First American Science Writer Found (Then Lost) God in
the Cosmic Rays, https://www.sciencehistory.org/distillations
[Wulf1909a] T. Wulf, A new electrometer. Phys. Z. 10, 251 (1909)
[Wulf1909b] T. Wulf, Uber die in der Atmosphäre vorhandene Strahlung von Hoher Durch-
dringungsfähigkeit. Phys. Z. 10, 997 (1909)
[Wulf1910] T. Wulf, Beobachtungen űber die Strahlung hoher Durchdringungsfähigkeit auf dem
Eiffelturm. Phys. Z. 11, 811 (1910)
[Xing-Ming2014] F. Xing-Ming et al., A position resolution MRPC for muon tomography. Chin.
Phys. C 38, 046003 (2014)
[Xu1987] Q. Xu, L.M. Brown, The early history of cosmic ray research. Am. J. Phys. 55, 23 (1987)
[Yock1986] P.C.M. Yock, Heavy cosmic rays at sea level. Phys. Rev. D 34, 698 (1986)
[Yodh2013] G.B. Yodh, Early history of cosmic rays at Chicago. AIP Conf. Proc. 1516, 37 (2013)
[Yoshida1959] S. Yoshida, M. Wada, Storm-time increase of cosmic ray intensity. Nature 183, 381
(1959)
[Zanini2001] A. Zanini et al., Neutron spectrometry at various altitudes in atmosphere by passive
detector technique. Nuovo Cimento C 24, 691 (2001)
[Zanini2009] A. Zanini et al., Cosmic rays at high mountain observatories. Adv. Space Res. 44,
1160 (2009)
[Zatsepin1951] G.T. Zatsepin, Dokl. Akad. Nauk SSSR 80, 577 (1951)
[Zatsepin1966] G.T. Zatsepin, V.A. Kuz’min, Upper limit of the spectrum of cosmic rays. ZhETF
Pisma Redaktsiiu 4, 114 (1966)
[Zhou1999] D. Zhou et al., Cosmic ray abundance at aircraft altitudes in the Earth’s atmosphere,
in Proceedings of the 26th International Cosmic Ray Conference, Salt Lake City (USA), vol. 3
(1999), p. 101
[Ziegler1996a] J.F. Ziegler et al., IBM experiments in soft fails in computer electronics (1978–1994).
IBM J. Res. Dev. 40, 3 (1996)
[Ziegler1996b] J.F. Ziegler, Terrestrial cosmic rays. IBM J. Res. Dev. 40, 19 (1996)
[Ziegler1989] Ch.A. Ziegler, Technology and the process of scientific discovery: the case of cosmic
rays. Technol. Cult. 30, 939 (1989)

Messengers From The Cosmos: Francesco Riggi

Uploaded by

Copyright:

Available Formats

Messengers From The Cosmos: Francesco Riggi

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Messengers From The Cosmos: Francesco Riggi

Uploaded by

Copyright:

Available Formats

UNITEXT for Physics

Messengers from the Cosmos

ISSN 2198-7882 ISSN 2198-7890 (electronic)

several colleagues is warmly acknowledged. A special thank to the Springer staff

Catania, Italy Francesco Riggi

1 The Discovery of the Cosmic Radiation . . . . . . . . . . . . . . . . . . . . . . . . . . 1

5 Developments of the Techniques for the Detection

9.4 Electrons and Positrons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

15 The Origin of Cosmic Rays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257

Appendix A: A Calculation of the Flux at the Top of the Eiffel

Appendix I: Gaisser-Hillas Parameterization of the Longitudinal

Abstract The balloon ascent made by Victor Hess in 1912, up to an altitude

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 1

phenomenon is due to something whose intensity undoubtedly increases with altitude

1.2 Terrestrial Radioactivity and First Experiences

The phenomenon of the spontaneous discharge of electroscopes had already been

subject of very extensive studies by numerous scientists. Although the knowledge

agents [Wilson1898, Wilson1899, Wilson1900a, Wilson1900b, Wilson1901a]. In

Fig. 1.4 Arthur Stewart Eve (1862–1948)

Fig. 1.5 Domenico Pacini (1878–1934). Courtesy of Pacini family, by A. De Angelis

1.3 Investigations in the Atmosphere

Fig. 1.6 Theodore Wulf (1868–1946)

by radioactive nuclei uniformly distributed in the ground, up to a depth of one meter

Fig. 1.9 Albert Gockel (1860–1927)

Fig. 1.10 Karl Bergwitz (1875–1958)

1.4 Victor Hess and the Evidence for an Extraterrestrial

Table 1.3 Summary of balloon flights performed by Hess in 1912 [Hess1912]

1.5 Towards a Confirmation of Hess’s Results

Fig. 1.14 Werner Kolhörster (1887–1946)

2.1 Further Investigations in Europe During and After

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 27

after 1913, published in Italian, were mainly devoted to spectrophotometric studies

study of the cosmic radiation, through quantitative investigations of the variations

2.2 Robert Millikan and the First American Contributions

2.3 The Acceptance of the Idea of a Cosmic Radiation

3.1 The Influence of the Earth’s Magnetic Field

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 41

3.2 Campaigns for Measuring the Intensity of the Cosmic

Fig. 3.6 Jacob Clay (1882–1955). Source Academictree.org

3.3 The Debate on the Corpuscular or Radiative Nature

3.4 Further Contributions in Europe and Other Countries

Fig. 3.14 Erich Regener (1881–1955)

possibility that allowed the use of a telescope of counters, capable of signalling

Fig. 3.19 Dmitry Skolbetsyn (1892–1990). Source Alchetron.com

3.5 Protons as an Essential Component of Primary

these particles as electrons, the range of electrons emitted by radioactive substances

4.1 The Discovery of New Particles and the Links

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 75

interact with matter—the photoelectric process and the Compton process—were

4.2 Properties of µ-Mesons

4.3 The Discovery of the Pion

numbers, while it became predominant in elements with high Z, such as reasonably

4.4 The Discovery of the Neutron

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 91

5.2 From Wulf’s Electroscopes to Automatic Recording

As we discussed in Chap. 1, reporting the first observations of the cosmic radiation,

Fig. 5.1 Sketch of the

The way to operate an electrometer for quantitative measurements is to charge

5.3 Ionization Chambers