SUNGRAZING

COMETS
SNOWBALLS IN THE FURNACE

By David A. J. Seargent MA, PhD, FRAS











© David A. J. Seargent 2012



TABLE OF CONTENTS
PREFACE

I. INTRODUCING THE SUN-DIVING COMETS

II. NON-KREUTZ SUNGRAZERS
III. PROVEN SUNGRAZERS
IV. WERE THESE COMETS SUNGRAZERS?
V. THE EVOLVING STORY OF KREUTZ GROUP EVOLUTION

PREFACE

Imagine that you are a disembodied spirit residing on a chunk of dust-
encrusted ice around 160 times the Earth’s distance from the Sun (160
Astronomical Units or AU in astronomical parlance). I invite you to imagine that
you are a disembodied spirit, because nothing physical could survive the trip that
is in store for you!
The year that you commence your journey is sometime around 1660 AD.
Your surroundings are incredibly cold and dark. None of the Solar System’s
planets is visible but amongst the stars there is one whitish luminary that is
greatly more brilliant than the rest. It is just a point of light like the rest of the
stars, but at a magnitude of around -16 it shines nearly 792,000 times more
brightly than its nearest rival, Sirius. The bright star is the Sun, and the ice
fragment on which you ride is beginning to fall ever so slowly toward it.
Years roll into decades and decades into centuries, and little appears to
change. The Sun gradually brightens, but does not change much in appearance.
Eventually, at the end of October 1991 as we measure time on Earth, you reach
Pluto’s mean distance from the Sun at 39.34 AU. The Sun has brightened by
about 3 magnitudes or 16.4 times but is still less than a minute of arc in diameter.
Picking up speed, your icy home reaches Earth’s distance from the Sun
(albeit well below the plane of the planets) on November 17, 2011. By
December 5, this distance has been halved and the Sun increases to a disc a full
degree in diameter and shining at a magnitude of around -28.3.
Things really happen quickly during the following 10 days. Already gas and
dust is boiling off the icy body on which you travel. At 1.83 hours on December
15, you reach just 0.1 AU from the Sun and the heat has become intense. The icy
body is surrounded by a large cloud of dust and glowing gas, sweeping back far
into space in a direction away from the Sun. The Sun itself appears as disc 5
degrees across and shining at a magnitude of -31.8.
By 23 hours on the next day, the Sun balloons to a diameter of 50 degrees
and its brightness increases to magnitude -36.8. Little more than an hour later,
you are flying at fantastic speed just 135,000 kilometres above a nightmare
landscape – the photosphere of the Sun! Below you the Sun’s surface boils and
erupts. If any prominences are close at hand, their tops tower far above you like
giant boiling luminous trees. You are flying at speeds well in excess of a million
kilometres per hour through the million-degree corona of the Sun, its tenuous
gas streaming by in a rarefied super hurricane. The icy surface of the shrinking
remnant fairly explodes into gas, yet somehow manages to hold together.
Soon, the distance from the Sun’s surface starts increasing and before that
eventful day of December 16, 2011 is over, the central star of the Solar System
transforms from an incredible landscape above which you have flown, once
more into a brilliant disc some 5 degrees across.

Congratulations, in your imagination you have just ridden the nucleus of
Comet 2011 W3 (Lovejoy) from aphelion (the point of its orbit most remote
from the Sun) in the middle of the seventeenth century, through perihelion (the
point of its orbit nearest the Sun) on December 16, 2011. This object was one of
a large family of comets whose incredibly small perihelion distances have earned
them the name of “sungrazer”. Their story is the subject of this book.


I
INTRODUCING THE SUN-DIVING COMETS

Ever since the Great Comet of 1680 sped to slightly over 0.006
Astronomical Units of the Sun (i.e. one 0.006th. of the distance from Earth to
Sun) astronomers have been intrigued as to how these small and apparently
fragile objects manage to survive such flights through the Sun’s corona. The
1680 comet – of C/1680 V1 as it is now more formally known – remained an
apparently one-time freak until the nineteenth century when a short but
spectacular parade of comets having similar perihelion distances blazed through
the skies. This cometary procession began with an exceptionally brilliant and
beautiful object in late February and March 1843 and continued with a
moderately bright object sporting a tail of colossal length in 1880, a very
spectacular comet in 1882 and a strange object described as a headless tail in
1887. Excepting the freakishly small perihelion distances of these comets, the
parade of the 1800s showed no orbital similarities with the earlier one of 1680,
but they did display an unusually close similarity with one another. So much so
in fact, that initially the 1880 comet was thought to have been a return of the one
in 1843, even though their orbits did not indicate periods measurable in decades.
Individual comets sharing similar orbits had been noted before and M. Hoek
suggested that these may be instances of a single comet having split apart at an
earlier perihelion passage only to return as two or more individual objects,
travelling in similar orbits but possibly arriving at widely separated times. By the
latter half of the nineteenth century this idea had ceased to be radical as an
instance had already been observed in the splitting of the short-period comet
Biela and its subsequent return as two separate bodies. If a short-period comet
could bifurcate in this way, there seemed no reason why a long-period comet
could not behave similarly. Nor did there appear any reason why splitting
should not produce more than just a pair of objects.
German astronomer H. Kreutz shared these opinions. He reasoned that the
comets of 1843, 1880, 1882 and 1887 were all fragments of a single comet that
passed very near the Sun at an earlier return. Kreutz also suspected that a bright
comet observed only briefly in early 1702 was another member of the same
group. No orbit could be calculated for this object due to the scarcity of even
tolerably good positional measurements, but from the little information that was
available, it seemed that the comet followed a very similar path through the sky
to what the others would have pursued at that time of year. Moreover, the general
appearance of the 1702 comet – small head and long, narrow but quite intense,
tail – agreed very well with the descriptions of the later objects.
Kreutz also found support for his argument regarding the genesis of this
comet group by noting the behaviour of the 1882 comet following its passage of
the Sun. The nucleus of this object divided into a line of condensations
graphically described at the time as a “string of beads” which slowly grew
further apart over time. One of these was bright and could be considered the
“main” nucleus. On the sunward side of this a very small condensation was
observed and on the tailward side, a far more prominent one, albeit not as bright
as the main nucleus. Tailward of this again, was another small condensation,
almost as faint as the sunward one. A fifth was suspected at times between the
two brightest condensations, but this was very faint and probably transitory. A
sixth condensation was also mentioned in Kreutz’s account of the comet. It is
likely, however, that only the two main nuclei ultimately survived, but the very
fact that this comet split into several pieces was certainly good support for his
hypothesis as to how this comet group came into being. Nowadays, nobody
seriously doubts the basic validity of Kreutz’s hypothesis, although the details
have evolved considerably and continue being worked out in finer detail. The
comet group has for long been known as the “Kreutz group” or the “Kreutz
family of sungrazing comets” in his honour.
Before going any further, it will be advisable to fix the meaning of the main
terms that we will be using in these pages. Terms like “sungrazer” are not precise
and can mean slightly different things to different people. Clearly, a “sungrazing
comet” is one that passes very close to the Sun, but just how close is deemed
very close? The term is sometimes used to denote objects passing between 0.01
and 0.1 AU of the Sun’s centre and I recall it once being used to describe a
comet that came to perihelion significantly over 0.1 AU from the Sun. Most
comet experts would consider this last instance to be a misuse of the term but
there is nevertheless uncertainty as to where to correctly draw the line. For most
people however, the “line” is 0.1 AU, and this is the sense in which I will be
using the term here (with 2 very minor exceptions which I will come to shortly).
For comets having perihelia between 0.01 and 0.1 AU, the term sunskirter will
be used. This term seems to have been coined by Z. Sekanina and P. Chodas and
is an excellent one for those comets which are not quite sungrazers but
nevertheless still pass unusually near the Sun. In a secondary sense, “sungrazer”
will also be used to describe any member of the Kreutz group ... and it is here
that the minor exceptions referred to above become relevant. Some of the Kreutz
group members have perihelia almost as “large” as 0.01 AU and there are two
known instances of very small comets (C/ 1984 O2 and C/2000 U3) observed in
data from the SOLWIND and SOHO space-based solar observatories, for which
a perihelion distances slightly in excess of 0.01 AU were computed. Without
splitting hairs, we can still refer to these objects (and any similar which might be
observed) as sungrazers on the grounds that they were clearly members of the
Kreutz group. It should also be mentioned that the exact value of perihelion is
difficult to determine for the small comets discovered in images from space-
based coronagraphs; another good reason for not splitting hairs over these
objects!
Returning to the Kreutz group in general, no further members were
observed in the decades following 1887 and it may have seemed that the family
of sungrazers had run its course. If astronomers of the time did think that
however, later history proved them well off the mark!
The first sign that the sungrazers were not yet finished came on December
11, 1945 when du Toit in South Africa discovered a seventh magnitude comet on
a series of photographs that he had taken of the southern skies. It was only
followed a further 4 days, after which it became lost in twilight. Curiously, it was
not observed in January 1946 even though it should have been a striking
southern object at that time.
After 1945, the Kreutz group became quiet again until 1963 when Z.
Pereyra of Cordoba Observatory in Argentina discovered a bright object with a
long tail in the morning twilight of September 14. Poorly placed and moving
away from Sun and Earth, C/Pereyra faded quickly and did not become the
striking object that it potentially could have been had it been better placed with
respect to Earth and Sun.
Following the appearance of Pereyra, a further 2 sungrazing comets were
discovered during the next 7 years; the magnificent Ikeya-Seki in 1965 and less
prominent, though briefly quite striking, White-Ortiz-Bolelli of 1970.
Then the fun really started!
The first indication that Kreutz sungrazers may be a lot more frequent than
ever imagined came in 1981 when details of a rather bright object apparently
striking the Sun back in August 1979 was released by the team responsible for
the SOLWIND coronagraph onboard the US Naval Satellite P78-1. A sufficient
number of positions was obtained from the SOLWIND images for an orbit to be
calculated, and this showed the comet to be yet another member of the Kreutz
family. Nevertheless, although it became about as bright as Venus (fortuitously
visible near superior conjunction in the same SOLWIND images), the intrinsic
brightness of this comet was inferior to those previously observed from the
ground – including, albeit not greatly, the faint object of 1945. Although the
initial thought that it hit the Sun proved incorrect, the comet apparently faded out
at perihelion, as it did not re-emerge from behind the coronagraph occulting
disk.
The big question though, was whether SOLWIND’s capturing of this object
was just a random fluke or whether it implied that an entire population of
intrinsically faint Kreutz comets existed. Faint objects such as the 1979 comet
would be unlikely discoveries from the ground and, if they all faded away near
perihelion, the only chance of finding them would be via the sort of space-based
solar observatories which SOLWIND was pioneering.
The answer to the question came quickly. Altogether, 6 Kreutz comets were
discovered by SOLWIND prior to its succumbing to target practice courtesy of
President Regan’s controversial “Star Wars” programme. Nevertheless, in more
recent times, with the availability of SOLWIND images via the Internet, Reiner
Kracht found a further 3 faint Kreutz comets in SOLWIND images of 1981
November 20, 1983 July 7 and 1984 August 22/23, in addition to a previously
unrecognized non-Kreutz non-sungrazer in images for 1984 September 15/16.
Prior to the demise of SOLWIND, the Solar Maximum Mission (SSM) was
launched and began its independent monitoring of the Sun. SSM was more
sensitive than SOLWIND, but only imaged segments of the solar corona at any
one time, not the entire solar neighbourhood in the manner of the older
coronagraph. Nevertheless, before the decade was over, SSM added a further 10
Kreutz comets to the total. One comet was imaged by both spacecraft,
confirming beyond all shadow of doubt that these were real comets and not freak
lens reflections or whatever.
The SOLWIND and SSM discoveries were as nothing, however, by
comparison to what happened following the launch of the ESA/NASA space-
based solar observatory SOHO (Solar and Heliospheric Observatory) in 1995.
Although discovering comets was not part of the purpose of this spacecraft, it
has turned out to be by far the most prolific comet-hunting instrument in history.
More comets have been found by SOHO than had previously been recorded in
the whole of human history! From its fist light until July 2012, some 2,337
SOHO comet discoveries have been confirmed, more than 60% of these
belonging to the Kreutz group. The majority of these comets have been
intrinsically very faint objects moving in sungrazing or sunskirting orbits. Very
few have been observed from the ground and all of the sungrazers faded out as
they approached perihelion. Conversely, most of the sunskirters, and the handful
of others whose perihelia lie outside of the sunskirter range, survived their
encounter with the Sun. Remarkably, some of these sunskirters have turned out
to be of short period.
In recent years, Kreutz comets have also been discovered in images beamed
back by the STEREO spacecraft and others have been observed in both
STEREO and SOHO data. Nevertheless, SOHO remains the sungrazer catcher
per excellence and the publication of its data on the Internet in near real time has
encouraged a number of dedicated comet hunters to maintain a constant scrutiny
of the data, snapping up every comet that comes along!
Over the years of SOHO data, the frequency of these mini–Kreutz comets
has clearly increased, from around 50 per year in the mid – 1990s to near 200
per year at present. This is not a selection effect spawned by the growing number
of Internet comet hunters nor is it due to the finer honing of their skills. Some of
the most skilled and successful of their number (a notable example being R.
Kracht) have gone back over archived images right to the beginning of SOHO’s
career and several faint and previously overlooked comets have been discovered
in this way. Yet, the earlier years still fail to tally with more recent times, clearly
indicating that the increase in the numbers of these objects is intrinsic. In fact,
one might suppose that after some 17 years of exposure to the rigors of outer
space, SOHO’s instruments would be netting fewer comets now than when they
were fresh and new!
As these words are being written – in October 2012 – the remarkable fact is
that most of the comets known today are members of the Kreutz sungrazing
group! Furthermore, the overwhelming majority of these have little in common
with the Kreutz objects known prior to the late 1990s. The majority are tiny,
faint, invisible from the ground and unable to survive their freakishly small
perihelia. Some have tails visible in the SOHO/LASCO coronagraph images.
Others are mere specks of light. The majority would not be accessible to the
naked eye even if they could somehow be transported into a dark night sky. How
very different they appear from the “classical” sungrazer; those magnificent
comets with glorious tails sweeping far across the sky and heads that, at times,
have been visible without optical aid in full daylight almost at the Sun’s limb!
Yet, certain astronomers saw the increase in numbers of these tiny comets as
evidence that one of the “big ones” might be on its way. During the festive
season of 2011, these predictions seem to have been fulfilled with the arrival of
the unexpectedly spectacular C/2011 W3 (Lovejoy), aka the “Great Christmas
Comet of 2011”. Although at the lower end of the spectrum of major sungrazers,
C/Lovejoy nevertheless provided a wonderful spectacle (at least for southern
observers) and was far and away the most impressive of its family to have
passed our way since 1965. Its appearance has aroused interest in the group as
well as heightening expectations that other spectacular sungrazers will appear in
the not-too-distant future.
In the course of the following chapters, we will look more closely at the
major comets that are positively included within the Kreutz group, as well as
giving briefer consideration to a number of others whose Kreutz credentials have
been raised by various people at different times. Some of these are very likely
candidates for family membership. Others are very doubtful and some can
almost certainly be excluded. We will then look at the more recent attempts at
modelling the structure and evolution of the family as a whole, say a few
speculative words about the diverse patterns of behaviour shown by different
members of the group and dare to make some predictions as to the future
evolution of the group, especially considering the possibility of further bright
members in the near future. First of all however, we will digress somewhat from
the Kreutz group and have a quick look at the only known sungrazers that appear
unconnected with this family. These will be the subject of the following chapter
and, as we shall soon see, they are surprisingly few in number. I say
“surprisingly”, because dynamically speaking, comets having high inclinations
and perihelia within the Earth’s distance from the Sun have some probability of
evolving into sungrazers. For instance, the Great Comet of 1997 – Comet Hale-
Bopp – has about a 15% chance of becoming a sungrazer eventually. The reason
that we don’t see more is probably because most comets are too small to persist
as their perihelion distances shrink all the way down to sungrazing dimensions.

II
NON-KREUTZ SUNGRAZERS

This will be a very short chapter! Although most of the comets known today
are Kreutz sungrazers, courtesy of the SOHO observatory, sungrazers unrelated
to the Kreutz group are very, very rare indeed. As of October 2012 when these
words are being written, just 3 such comets are known. One is the great Comet
of 1680 which we have already mentioned and about which more will shortly be
said. The other two were faint mini-comets discovered and observed by SOHO,
both of which faded away completely as they moved toward perihelion.
The first of these is designated C/2001 N1 (SOHO) or, in order of discovery
with this observatory, SOHO 340. With a very Kreutz-like perihelion distance of
just 0.0052 AU on July 6, 2001, the other elements of this comet’s orbit were
nevertheless very different from those of the Kreutz comets and C/2001 N1
appears to bear no relation to this group.
The second non-Kreutz was found by SOHO on June 25, 2007 and is
officially designated C/2007 M5 (SOHO) and, informally, as SOHO 1331. Like
C/2001 N1 this sungrazer pursued an orbit entirely dissimilar to that of the
Kreutz group in spite of its extremely small perihelion distance. The most
interesting aspect of this comet is the extremely small value of this distance; just
0.0011 AU from the centre of the Sun! This is by far the smallest perihelion
known, well inside that of any of the Kreutz members. In fact, the perihelion
point of this comet’s orbit lies some 0.00355 AU – more than 530,000
kilometres – beneath the Sun’s photosphere! In theory, the comet should have
actually hit the Sun or else burnt up in the lower chromospheres as a solar
meteor. In practice however, there is little chance that any remnants of this tiny
object survived into the lower corona. As observed in SOHO images from the
LASCO C2 coronagraph, the comet had already faded to invisibility before
reaching the limb of the coronagraph’s occulting disc.
Little more can be said about these two very small comets, except to repeat
that although they were clearly sungrazers, the orbits of each bore little
resemblance to that of the Kreutz group or, for that matter, to each other or to the
very different specimen to which we shall now turn; the Great Comet of 1680.

 THE GREAT SUNGRAZER OF 1680
The Great Comet of 1680 – more formally known as C/1680 V1 – was one
of the finest on record and has secured its place in astronomical history not
simply because of its brightness and length of tail, but also by being the first
comet discovered by telescope and through the role it played in the development
of Newtonian gravitational theory, earning it a place in Newton’s epoch making
Principia. It rates alongside the periodic comets Halley and Encke in respect of
its importance to the development of our understanding of cometary orbital
mechanics.
The story of its discovery is an interesting one. On the morning of
November 14, 1680, Gottfried Kirch of Coburg, Germany, was busy observing
the 23-day-old Moon and the planet Mars when he happened upon a star close to
the Moon that did not appear in the star catalogue of Tycho Brahe that Kirch was
using. During the course of his endeavours to plot an accurate position for this
star, Kirch came across another object unknown to him; something that he later
described as a “sort of nebulous spot, of an uncommon appearance.” He thought
that this might be “a nebulous star, resembling that in the girdle of Andromeda”
or, alternatively, a comet. As it turned out, it was the latter!
At the time of discovery, Kirch did not see any tail and there is no mention
of the comet being visible with the naked eye. Maybe the nearby Moon played a
role in the object’s visibility, as just two days later Kirch noted a half degree tail
visible in his telescope. This, plus the object’s movement over that period of
time, betrayed the true nature of Kirch’s “nebulous spot”.
On November 20, by which time the Moon was out of the way, the comet
was seen with the naked eye from the Philippines, on the following morning
from England and on the 22nd. in China where it appeared to the naked eye as a
“white broom star” sporting 1.5 degrees of tail. The description “white” in
Chinese chronicles is normally taken as implying “bright”, so it seems that the
comet had brightened rather quickly since Kirch’s first sighting, aided no doubt
by the timely departure of the Moon from the morning skies.
By month’s end, the comet had become a striking spectacle in the pre-dawn
sky. Observing from Rome on November 27, J. Ponthio saw some 15 degrees of
tail with the unaided eye and on the 29th. of the month, A. Storer of Maryland in
North America measured the tail as between 15 and 20 degrees in length. Up to
36 degrees of tail were seen by an observer in Dresden on that same morning, by
which time the brightness of the comet was also said to match or exceed a star of
first magnitude.
 Even though the comet continued to increase in brightness, observations
started becoming more difficult after the early days of December as it plunged
ever deeper into twilight, becoming invisible to most observers after the first
week of that month. On December 18, the comet passed just 0.006 AU from the
Sun’s centre. At the time of perihelion passage, it was actually eclipsed by the
Sun, although even had it not been eclipsed, it is unlikely to have been visible so
close to the Sun’s surface. Daylight observations were, however, made earlier
that day from the Philippines, some 9 – 10 hours prior to perihelion when the
comet was still around 2 degrees from the Sun. Then, on the following day,
according to the writings of Dutch settlers at Esopus in New York, “there
appeared an extraordinary comet, which caused great consternation throughout
the province.” They wrote of its appearance “about two o’clock in the afternoon,
fair sunshine weather, a little above the Sun, which takes its course more
northerly, and was seen the Sunday night, right after about twilight, with a very
fiery tail or streamer in the west, to the great astonishment of all spectators, and
is now seen every night with clear weather.”
From December 20, the comet became widely seen in the western evening
sky with a very long and bright tail which, in the estimation of some observers,
possessed a distinct golden colouration. In England on December 21, John
Flamsteed described the tail as a beam of light about the same width as the
Moon and extending straight up from the western horizon. Several estimates
indicated a length of about 70 degrees – even as long as 90 degrees according to
R. Hooke on December 28!
 Fig. 1. The Great Comet of 1680 over Rotterdam. Painting by Lieve
Verschuier.


Telescopic observations of the inner coma around the end of December
revealed what may have been the first sightings of jet activity similar to that
observed in many active comets of later times. Hooke, for example, saw a stream
of luminous matter issuing from the nucleus “in the manner of a sudden spouting
of water out of an engine”. This reads very much like a description of the sort of
jet activity that has been noted in, for example, Comet Halley on several returns;
not least at its most recent apparition in 1986.
By December 31, the comet’s brightness had declined to third magnitude
according to Flamsteed’s estimate, who also remarked that the previously star-
like condensation had by then expanded somewhat into a “hazy light”.
Fading continued during early January, although the tail remained as long as
75 degrees according to some observers on January 5. Kirch observed an
interesting feature on January 7 and 8; a sunward-pointing anti-tail, apparently
far fainter than the main one. Regrettably, he gave no description of this feature
and it does not appear to have been recorded by other observers. From that fact
alone, we can surmise that it was very faint, but its length and width remain
unknown.
The head of the comet fell below naked-eye visibility in early February
(being down to magnitude 7 on the 4th. according to Flamsteed) although up to
12 degrees of tail remained visible to the unaided eye until February 17 for
observers with very favourable sky conditions. This naked-eye persistence of the
tail after the head has slipped into the telescopic realm is not uncommon with
sungrazers nor, for that matter, other comets displaying intense dust tails. Most
astronomers had given up on the comet by the end of February, but Newton
continued to follow it telescopically until as late as March 19, by which time it
was “scarcely discernible” in his telescope.

ODD COINCIDENCES ... OR SOMETHING ELSE?
Excepting the very small perihelion distance, the orbit of this comet has
essentially nothing in common with those of Kreutz comets. These move in
highly inclined retrograde orbits; the 1680 comet in a prograde orbit of moderate
inclination. The Kreutz comets have periods of several hundreds of years –
relatively short by the standards of comets. Although the orbital period of the
1680 comet is poorly determined, the best estimates place it in the region of
9,000 years, around an order of magnitude longer than the Kreutz. Surely there
can be no association between the great sungrazer family and this lone object!
And yet ... H. Kreutz himself drew attention to a strange coincidence
concerning this comet’s orbit and that of the other sungrazers. Or, at least, the
sungrazer of 1882. The orbits of both comets – though totally dissimilar to one
another – nevertheless almost cross in space! The very low probably of two
unrelated comets having such highly anomalous perihelia distances moving in
orbits that almost intersect is just too much for some astronomers to swallow and
the very reasonable suggestion has been made that even this comet may be
related in some distant way to the rest of the sungrazing fraternity. Perhaps the
1680 comet split away from the original Kreutz sungrazer – the ultimate parent
comet of the family – at some long ago perihelion passage. The problem with
this however, is the requirement for a very violent schism of a totally different
order to the gentle splitting observed in the comets of 1882, 1965 and
(presumably) the original sungrazer, as well as any other fragmentation events
that may have occurred along the way. A split sufficiently violent to separate
1680 from 1882 would also be unprecedented amongst the many other (non-
sungrazing) comets that have been observed to break up. Perhaps the apparent
orbital relationship is just a coincidence after all. Very extreme coincidences do
admittedly happen from time to time. It is said that during World War II, a bomb
fell through the roof of a house in England and failed to explode. That was
fortunate but not in itself highly unusual. But a short time later, a second bomb
fell through the hole in the roof made by the first ... and also failed to explode! Is
the probability of this happening any greater than the orbits of two unrelated
sungrazing comets intersecting? I leave this to mathematically minded readers to
calculate, but I would be surprised if the probability was greater!
Nevertheless, the plot has continued to thicken in a most unexpected way.
On September 21, 2012, V. Nevski and A. Novichonok of the International
Scientific Optical Network (ISON) found a slow-moving object of 17 – 18
magnitude which, although initially suspected to be an asteroid, was
subsequently found to be a distant comet. Initial orbital computations hinted at a
very exciting prospect. One of the possible orbits was that of a sunskirter!
Moreover, if the orbit was correct, perihelion was more than a year away.
Fortunately, pre-discovery images of the new object were found in the data from
other programmes and these supplied a pair of very early positions on December
28, 2011 and January 28, 2012. From these positions, plus the more recently
obtained ones, a good orbit was quickly computed and this revealed that the
object was indeed a sunskirter, having perihelion on November 28, 2013 at just
0.0125 AU from the Sun. This is a small perihelion distance, even by sunskirter
standards. The only known comets, other than true sungrazers, that approached
the Sun more closely were 2 minisunskirters discovered in SOHO data, each of
which had computed perihelia marginally smaller than that of ISON. In fact, had
the perihelion of ISON been just a little smaller, it too would have been a bona
fide sungrazer. (Actually, one Kreutz minisungrazer – C/ 1984 O2 SOLWIND –
had a computed perihelion distance of 0.0154 AU; slightly larger than that of
ISON!)
But that was not all. As soon as the orbital elements were published, comet
expert John Bortle drew attention to the rather striking similarity between this
comet’s orbit and that of 1680. Now, unrelated comet orbits may show
similarities by pure chance, but in this instance both orbits are made highly
atypical by their very small perihelia. Putting this in context, if we take the
Kreutz family and the abovementioned pair of minicomets out of the equation,
the 1680 comet and ISON have the smallest perihelion distances of all known
comets. Two typical comet orbits might be similar by pure chance, but the odds
that a pair of such atypical ones would line up in this way are a lot longer!
There is not the slightest shadow of a chance of the two comets being
identical of course, and nobody even suggested that. But Bortle wondered if they
might be fragments of a single object that broke apart at an earlier perihelion
passage in prehistoric times. If this is true and if it is also true that the 1680 and
1882 comet are likewise associated in some roundabout manner, then it would
follow that C/2012 S1 (ISON), as the new object was officially designated, is
also a sort of distant cousin far removed of the Kreutz family!
Not everyone agreed with Bortle’s suggestion. Thus, Maik Meyer pointed
out that even though the orbits of both comets come close near perihelion, they
greatly diverge as we look outward from the Sun. This does not suggest true
association. The same thing may be said – and indeed is even more striking – in
the case of the 1680 and 1882 comets as well. Moreover, the orbit of ISON
seems to be just about as close to a perfect parabola as we can expect to find in
nature. Early computations of the comet’s orbit have shown very slight
divergences from an exact parabola; some falling a little on the elliptical and
some on the hyperbolic side of the parabolic limit. The former suggest orbital
periods anywhere between about a million to over 200 million years, with very
small differences in the calculated eccentricity making huge differences in
“period”. All of these values should be taken with a pinch of salt. Even if the
present orbit really is elliptical, it is still so close to the parabolic limit that the
aphelion distance – the point furthest from the Sun – lies within the remote
sphere of comets known as the Oort Cloud. Comets approaching from these
distant regions are so weakly held by the Sun’s gravity that, after a single orbit,
they are likely to either drift away from the Solar System altogether or be
slightly retarded in their motion by the pull of the planetary system and fall into
elliptical orbits of much shorter period. In other words, if we find a comet
coming in from Oort Cloud distances, chances are it is making its first (and
maybe only!) trip to the inner reaches of the Solar System. Although we cannot
be absolutely certain of this, the orbital evidence strongly suggests that ISON is
one of these maiden-voyage objects and as such, very unlikely to be physically
associated with the 1680 comet – to say nothing of the Kreutz group.
Another possible piece of evidence supporting this opinion is the early
activity of this object. At discovery, it was over 6 AU from the Sun – beyond the
orbit of Jupiter – and yet was already surrounded by a fuzzy coma and showing
signs of quite a high level of activity in the nuclear regions. Deep images even
revealed a small tail. Clearly, sublimating water ice cannot cause this sort of
activity so far from the Sun, so something more volatile than water must be
present in rather large quantities in the surface layers of the comet’s nucleus.
Arguably, this stock of volatiles would have been depleted had the comet
formerly been roasted near 0.01 AU from the Sun.
In short, it seems that the similarity between the orbits of the comet of 1680
and C/2012 S1 (ISON) is, after all, simply coincidental and, although this does
not relate directly to the apparent association between the orbits of the former
comet and the Kreutz sungrazer of 1882, it seems very likely that this is pure
coincidence as well. There the matter must rest unless some startling new
information can be gleaned from observations of ISON. But for now, let us turn
to the principal subject of this volume; the Kreutz family of sungrazing comets,
beginning with an overall review of the objects that we know to be members of
the group, before moving on in the subsequent chapter, to those more or less
strongly suspected of being members.

III
PROVEN SUNGRAZERS

We will now leave the very rare non-Kreutz sungrazers and take a closer
look at the comet family whose history we briefly skimmed over in Chapter One
of this book – the fascinating family of Kreutz comets that have inspired such
interest in recent times. Although probable members of the group were recorded
in earlier centuries, the earliest undisputed member of this comet clan was the
magnificent object that appeared early in 1843 and which is widely considered to
be one of the most brilliant and spectacular comets ever seen. Let’s take a closer
look at this object.

C/ 1843 D1
This may have been one of the finest comets of the 1800s, yet it bears the
name of no well-known astronomer. Indeed, nobody knows the name of the first
person who saw it. The only knowledge of its discovery is a report in a New
York newspaper – not even a scientific journal – simply stating that a comet had
been seen in the evening sky on February 5. Later, another news report spoke of
a comet sighted in the evening close to the star Beta Ceti on February 11.
Scattered reports subsequently came in on the 19th. 23rd. and 26th. of the month of
sightings of a comet tail having been espied in the evening twilight.
A strange report was received late in the month of a daytime comet seen
very near the Sun on February 27.66. This came from P. Ray at Conception,
Chile who estimated that the object was just 5 arcminutes from the Sun’s limb.
That, as became evident later once the comet’s orbit had been calculated, was
not correct. The comet would have been about 1.5 degrees from the Sun’s centre
at the time, but there is no reason to think that the sighting itself was false.
Perhaps the report somehow became garbled. The actual distance from the solar
limb was 75 arcminutes, so maybe the “7” somehow got lost in reporting. We
simply do not know.
Although nobody knew at the time, the comet passed through perihelion just
0.0055 AU from the Sun’s centre on February 27.91. By the middle of the
following day, the comet had withdrawn far enough from the Sun to be widely
noted shining brightly against the daylight sky. It was seen in China as “a large
broom star” and in the occident as a “beautiful star” and “a short dagger-like
object” around 4 – 5 degrees from the Sun. Although obviously very brilliant –
presumably in the range -9 to -10 – the long tail was probably the main feature
that made this comet so conspicuous.
On the evening of March 1, the tail was seen by several people rising out of
the evening twilight and by the 4th.of that month, 30 degrees of tail were
observed streaming away from a head of second magnitude. The tail continued
to lengthen even as the comet faded, reaching a maximum of 64 degrees on
March 21 according to J. Schmidt’s estimates. Estimated lengths in the range of
50 – 40 degrees were, however, quite common between March 10 and 20 by
which time the comet was being observed as far north as Germany. These
lengths refer to the main dust tail. Earlier in the month, several observers
reported a fainter, albeit even longer, tail that may have been an ion tail although
its exact nature does not appear to have been studied. According to W. Clerihew,
observing from India on the evening of March 11, this fainter appendage was
“twice as long” as the main one. Clerihew did not give an estimate of the actual
length of either tail, but the fact that he could observe the fainter one as well as
he did implies that his observing conditions must have been good and that his
view of the bright tail would have equalled the best observations by other people
on that night. As other observers generally made the main tail between 36 and 45
degrees at the time, this implies that Clerihew probably saw between 72 and 90
degrees of secondary tail! The fainter tail was said to form an angle of about 20
degrees to the main one.
Typical of bright sungrazing comets, the tail remained far more conspicuous
than the brightness of the head might seem to imply. By March 17, J. Herschel
estimated the head to be just 5th. magnitude and the “nucleus” to be very
indistinct, even though he could still see 30 degrees of tail with the unaided eye.
Earlier, what may or may not have been a multiple nucleus was noted by E.
Herrick at New Haven, Connecticut, when he observed the comet through a 13-
cm. refractor at 55x on March 12. According to Herrick, the head of the comet
presented as “an indefinite globular body, somewhat elongated behind, with a
concentration of light near or a little in advance of the centre, which at times
seemed to consist of three faint stellar points.” Superficially, this reads like an
observation of a broken nucleus; even more so in hindsight following the careers
of the later sungrazing comets of 1882 and 1965. Nevertheless, as no confirming
observations of a split or multiple nucleus appear to have been made by other
observers, we would be ill advised to jump to any conclusions about this matter.
Maybe the features Herrick saw were simply knots of tail material or, at best,
transitory fragments that had broken away from the main mass. Without further
reports, we simply do not know.
From the end of March, the comet faded more rapidly and by all reports
quickly became a difficult object. Observing on April 3, Herrick noted that it
was “barely discernible” with the unaided eye and the last sighting appears to
have been by W. Smyth and T. Maclear at the Cape of Good Hope on April 19,
by which time it had become barely discernible telescopically, being “of the last
degree of faintness” according to these observers.
Reliable magnitude estimates, at least those equal to the standard expected
today, are few and far between for this comet, but taken over all, the absolute
magnitude of C/1843 D1 appears to have been about 5, but with a steeper fading
late in the period of visibility.
 Fig. 2. The Great Comet of 1843.
C/1880 C1
Like the Great Comet of 1843, the Great Comet of 1880 was first espied by
a non-astronomer whose name has not come down to us. We are merely
informed that a “gentleman in the northern part of [New South Wales,
Australia]” saw it on February 1. The next evening, several more reports were
made, including one by B. Gould at Cordoba, Argentina. Although clearly not
the initial discoverer, Gould’s name has often been associated with this comet,
albeit in an unofficial capacity.
Initially, what most of the observers saw was a long tail beaming up from
beneath the south-western horizon. On February 2, about 20 degrees of tail were
visible, quickly extending to between 50 and 60 degrees by the 5th. and 6th. of the
month. The greatest measured extent seems to have been by David Gill in South
Africa, who managed to trace it for a full 75 degrees!
The head was far less impressive. L. Eddie compared it to the globular star
cluster 47 Tucanae on February 5. Presumably, he meant the similarity to relate
to the comet’s brightness as well as its general diffuse appearance, which would
indicate a magnitude at that time of around 4. From this very rough brightness
estimate, an absolute magnitude of about 8 is derived; remarkably faint for such
a spectacular comet.
The comet faded rapidly. By February 13, it had become quite difficult to
see without optical aid and by the 20th. of the month, it had disappeared
altogether.
Remarkably, the orbit turned out to be surprisingly similar to the comet of
1843. Perihelion had occurred on January 28, at which time the comet passed
just 0.005 AU from the Sun but, although the orbit was given as a parabola, the
similarity with 1843 D1 led many astronomers to conclude that this was really a
return of the former comet, although it was admitted that the display in 1880 had
not been so brilliant as the former apparition. Estimates of the period make
somewhat strange reading today as various attempts were made to identify
previous “returns” of the comet. One suggestion had the period varying from
several decades to as little as 7 years! The solution to this puzzle was, however,
not long in coming.

C/1882 R1
The mystery of what seemed to be a short-period comet moving in a near
parabolic orbit was finally solved by the appearance of a third object just over
two and a half years after the 1880 object. Not only did this new object solve the
mystery, but it quickly evolved into one of the most spectacular comets on
record, equally as magnificent as that of 1843 and with a period of visibility long
outlasting that of the earlier comet.
Once again though, the actual discoverer of this object is not known.
Indeed, there appears to have been several “discoverers” as reports of a bright
comet in the early morning sky of September 1, 1882, came in from the Gulf of
Guinea and the Cape of Good Hope at about the same time. Another independent
discovery was made on the 3rd. this time from New Zealand. The first “official”
word seems to have been the announcement on September 6 from Gould at
Cordoba that a comet “as bright as Venus” had been observed in the morning sky
the previous morning. It should be noted with respect to this comparison with
Venus, that this was not Gould’s estimate, but the impression of the unnamed
individual who reported it to the observatory. That it cannot be taken too
seriously as an accurate brightness estimate is demonstrated by the fact that, on
September 8, W. Finlay at the Cape estimated the comet’s brightness as just third
magnitude. An object of that brightness would, however, never be said to have
resembled Venus unless one was given to preposterous exaggerations!
Part of the discrepancy is surely explained by the fact that Finlay saw the
comet in twilight and that his estimate appears to have been of the central
condensation only. That probably indicates a total brightness of around first
magnitude at the time – still a lot less than Venus! The comparison with Venus
most likely refers as much to the size and conspicuous nature of the comet as to
its brightness. A bright golden object with an intense tail (as the report also
mentions) glowing in the morning twilight would be as conspicuous as Venus
and it seems natural that the comparison would be made, especially by someone
not experienced in making formal magnitude estimates of astronomical objects.
Further discoveries were made by J. Tebbutt of New South Wales on
September 8, by J. Reed on board HMS Triumph off Cape Verde Islands on the
10th. and by L. Cruls (with whose name the comet is sometimes associated) at
Rio de Janeiro on the 12th..
B y September 15, L. Eddie at Grahamstown in South Africa estimated the
comet’s brightness as equalling that of Jupiter and measured the tail’s length as
12 degrees. The following day, Gould tracked the comet throughout the daylight
hours with a finder scope and later that day, Tebbutt found it with the naked eye
in full daylight just 4 degrees from the Sun. The following morning, Eddie saw
the comet rise just 14 minutes prior to the Sun. After sunrise, it remained visible
with a tail approximately 1degree long. A little later that day, A. Common at
Ealing in England found it near the Sun during a deliberate search for near-Sun
comets following the discovery of a small object during a total solar eclipse in
Egypt the previous May. We shall say more about that object in the course of the
following chapter.
A very interesting observation was made by W. Finlay and W. Elkin on
September 17. Tracking the comet in late-afternoon daylight through a pair of
telescopes at the Cape of Good Hope, these astronomers were able to watch it
right up to the limb of the Sun. Finlay used a 15-cm telescope at 110 power
magnification fitted with an early type of solar filter. He described the comet as
being silvery in colour, in contrast to the reddish hue of the Sun as it appeared
through the filter, and measured it with a micrometer as just 4 seconds of arc in
diameter. A very short tail was also visible. These astronomers were probably
observing what had formerly been the innermost condensation of the head,
although so close to the Sun that was probably all that remained of the head as
the coma dust rapidly evaporated and was swept away by solar wind and
radiation pressure. Elkin timed the disappearance of the comet “among the
undulations of the Sun’s limb” at September 17.6506. Finlay saw it vanish 8
seconds later “when the Sun’s limb was boiling all about it” and may have very
briefly glimpsed it again about 3 seconds later, although he could not be sure of
the reality of this. After the comet had vanished, the face of the Sun was
searched for any sign of a transiting body, but none was seen.
It is interesting to note that these observations give us a relatively good idea
as to the comet’s brightness at the time. Assuming that the comet’s head and the
Sun’s surface were of equal intensity, the difference in diameter between the
Sun’s disk and the comet’s coma as measured by Finlay allows us to calculate
the difference between the total brightness of the Sun and that of the comet. The
difference comes out at almost 13.3 magnitudes, yielding an estimate of
magnitude -13.4 for the comet. This is surely too bright however. The
phenomenon of limb darkening means that the solar limb has an intensity of just
40% that of the face of the Sun itself. Had the comet been as intense as the face
of the Sun, it would have appeared bright against the solar limb. From Finlay’s
description, we can assume that the intensity of the comet and solar limb were
fairly close, although it is likely that the comet was somewhat fainter. On the
hypothetical assumption that the two were equally bright however, we get a
cometary magnitude of approximately -12.5. It is interesting to note that Elkin
likened the appearance of the comet vanishing at the Sun’s limb to the
occultation of a fourth magnitude star by the Moon. The difference in magnitude
between a star of that brightness and the full Moon is 16.7 magnitudes. If the
comet was that much fainter than the Sun, its magnitude would have been -10.1.
Had it been of equal intensity with the solar limb (and therefore, of magnitude
-12.5) the difference would have equalled that between a 1.6 magnitude star and
the Moon, which does not fit Elkin’s description. Pulling all of this together, it is
probably pretty safe to say that the comet’s magnitude at that time was in the
-10 to -11 range.
The next sighting of the comet appears to have been made on September
10.06 when J. Tebbutt of Windsor, New South Wales, observed it in broad
daylight less than 1 degree from the Sun’s western limb. Tebbutt did not attempt
an estimate of the comet’s brightness – only saying that it “far exceeded” any
other comet that he had observed – but judging by its relative ease of observation
it was probably in the region of -9. Later that same day, in terms of Universal
Time, when dawn was breaking in South Africa, David Gill watched the comet
rise just before the Sun over the mountains on the eastern side of False Bay. At
sunrise, it appeared pure yellow and after the Sun appeared, became clearer with
altitude rather than fainter, remaining visible to the unaided eye throughout the
daylight hours as a bright object with a tail around 0.5 degrees long. An observer
needed only to screen the direct light of the Sun with an outstretched hand to see
the comet. Daylight observations were made at various locations on the same,
and on the following, day and even as late as September 22nd. E. E. Barnard at
Nashville followed the comet with his unaided eyes until 15 minutes after
sunrise. Prior to the Sun’s rising on that morning, this same observer measured
the tail length as 12 degrees in bright twilight. On that same day, telescopic
daylight observations of the comet were also reported by J. M. Schaeberle at
Michigan, who saw it just before noon, and in Rome shortly after noon by E.
Millosevich. As late as September 27, 10 degrees of tail were still visible in the
rising dawn after first magnitude stars had faded from sight! At the end of the
month and early in October, the brilliant tail extended for about 20 degrees in a
dark sky and was observed to have a dark lane running down the middle like a
“shadow” of the nucleus. Despite its length, the tail was narrow, estimated as just
1 degree wide at the end of September. One observer in India likened the tail to
an elephant’s tusk. Remember this description, as we will meet it again in the
following chapter relating to a different and far earlier comet that may or may
not have been another member of the Kreutz group.
As October began, several observers noticed that the central condensation
of the comet began to take on an elongated appearance when viewed through a
telescope. Then, on the 3rd. of the month, F. Terby (Leuven, Belgium) noted that
it had definitely split into two segments and, 3 days later, other observers
reported 3 distinct nuclei being visible within the comet’s head. On October 15,
Eddie at the Cape of Good Hope found that when observed at low power through
a 24-cm reflector, the comet exhibited one distinct nucleus that he described as
resembling “the colour of the electric light” in addition to 2 further (and
apparently less conspicuous) condensations within “a bar of light”. Furthermore,
when he increased the magnification to 100 times, the number of condensations
within this bar “seemed again doubled, so that the whole nucleus resembled a
string of 5 ill-defined luminous beads.” It seems that up to 6 sub-nuclei were
observed at times within this nebulous “bar”; one a good deal brighter than the
rest.
It was this breaking up of the comet that gave the clue to solving the
mystery of why several comets had been found moving along almost identical
orbits, yet with periods that precluded their being simply multiple returns of the
same object. The main nucleus, and possibly the largest of the secondary nuclei,
were probably large enough to return in the future (though, as we will see later,
not necessarily as single comets). It seemed reasonable therefore, to think that
the comets of 1843, 1880 and 1882 had also been united as a single object once
upon a time and that this parent comet suffered a similar split at an earlier
perihelion passage. This was the hypothesis of Heinrich Kreutz and, although
this simple picture has undergone (and continues to undergo) modifications and
complications, the basic scenario has remained essentially unchanged to this day.
During October, another peculiar phenomenon associated with this comet
was also widely reported. The head of the comet became shrouded in a nebulous
sheath extending in a sunward direction for at least 4 – 6 degrees. Though a good
deal fainter than the main tail, this anti-tail was nevertheless visible with the
naked eye and appeared as a “tube of light” with the edges being rather more
intense than the central regions.
Even stranger was the appearance of apparently disconnected blobs of
nebulosity several degrees away from the comet itself; some of which took on
the appearance of separate comets. The first hint of these objects came on
October 5 when E. Marwick noticed two nebulous wisps in front of the comet’s
head. At first, he dismissed them as unrelated background nebulosities, however
no nebulae were shown on his star charts at the relevant positions. Then, on
October 8, J. Schmidt discovered what he thought to be a comet about 4 degrees
from the main object. He subsequently found this object (or what he presumed to
be the same object) on October 10 and 11 and from these observations, an
approximate orbit was calculated. This orbit had little in common with that of
the main comet, but suggested a sunskirting perihelion on September 25. As we
can appreciate from the continuation of the saga of these nubulosities, any
orbital solution assuming a regular Keplarian orbit around the Sun is probably
less than reliable, especially as we cannot even be sure that Schmidt’s
observations all related to the same thing.
On the same date as Schmidt’s second observation (October 10) either this
object or a similar one was also found by E. Hartwig at Strasbourg and described
as “a comet with a bright nucleus and a fan-shaped tail.” A second attempt to
locate this object on the 13th. failed to find any trace of it.
Other experienced observers also found nebulous objects in the vicinity of
the Great Comet of 1882. Thus, on October 14, E. E. Barnard came across a
nebulous mass about half the diameter of the full Moon located south of the
comet. Remarkably, he then noticed a second one apparently in physical contact
with the former and, on the opposite side of the first, yet another somewhat
fainter one. Sweeping with his telescope, Barnard found a total of 6 or 8 separate
nebulosities within 6 degrees of the comet’s head. One of these, as he recorded,
appeared very elongated. Somewhat later that month, on the 21st. in fact, W. R.
Brooks also discovered a comet-like nebulosity about 8 degrees from the 1882
comet. He described this object as being about 2 degrees long.
Although these objects are sometimes referred to as “satellite comets” in the
literature, this term should probably not be taken too literally. Certainly, they
could not have been “satellites” in the true sense of that word, but it is equally
unlikely that they were real comets. Had they split away as fragments from the
main nucleus, the velocities of separation would have been unacceptably high.
But if they were fragments accompanying the main comet on its way to
perihelion, they would need to have been quite large to survive. But, had they
really been that substantial, one would expect that they should have developed
impressive tails and appeared, not as mere nebulous patches, but as great comets
in their own right. What a spectacle that would have been; a veritable cloud of
great comets! The very rough orbital calculation for the object(s) seen by
Schmidt does not closely resemble that of the main comet, as already remarked.
The difference is too great to be explained by a fragment splitting away near the
Sun, although the chance that Schmidt saw a different object each night is not to
be discounted, especially as all other reports suggest that the individual
nebulosities were only seen on a single night.
Gary Kronk suggests that the nebulosities and the tube-like anti-tail were all
part of the same cloud of particles surrounding the head of the comet. The
nebulosities were not, in his opinion, evaporating icy bodies in solar orbit (i.e.
they were not “comets” in any genuine sense of that word) but transitory
condensations of particles within the broader and more nebulous cloud. They
were probably forming and dispersing quite rapidly which is why – with the
single dubious exception of Schmidt’s object – they were only observed on a
single night.
Unlike most other Kreutz sungrazers, the Great Comet of 1882 faded only
slowly as it moved away and remained a conspicuous object sporting at least 30
degrees of tail throughout October and November. The brightness of the head
seems to have still been around second magnitude in early November, fading to
perhaps fifth by the end of the month. Typical of sungrazers however, the tail
remained clearly visible by naked eye even as the head grew faint and naked-eye
sightings continued being reported even into the early days of March the
following year. The final telescopic observation of the comet was made as late as
June 1, 1883, when J. Thome at Cordoba, Argentina, described it as “an
excessively faint whiteness” discernible in his eyepiece.
Without doubt, this comet was the intrinsically brightest and, presumably,
the largest of the Kreutz objects of recent centuries. Conceptually placing it at
the standard distance of 1 AU from both Earth and Sun, the comet is estimated to
have shone with a brightness of at least first magnitude in rounded figures;
significantly more luminous than any other sungrazer of relatively modern times.
This absolute magnitude actually places it amongst the 10 intrinsically most
luminous known comets in history! This is a far cry from the next Kreutz
sungrazer that was to suddenly burst upon the scene in the 1880s!

 Fig. 3. The Great Comet of 1882 as photographed by D. Gill on November
7, 1882

C1887 B1
Once again the story repeats itself! The name of the true “discoverer” of this
comet is not known to us. All we learn is that he was a farmer living in South
Africa and that he spied the comet on the evening of January 18, 1887. The
following evening, the comet was independently found by Thome at Cordoba,
the astronomer who last saw the previous sungrazer of 1882. Thome’s name has
often been associated with this comet, although only in an unofficial capacity.
The comet presented a striking spectacle in the evening sky. On January 22,
Finlay described the tail as a “pale narrow ribbon of light” some 35 degrees
long. Tail lengths of around 40 degrees were widely reported between January
23 and 25. In strong contrast to the 1882 comet however, this one faded very
rapidly and appears to have been last detected by J. Tebbutt at Windsor on
January 30.
The oddest feature of this comet, was the apparent lack of any true head.
The only known report of anything remotely resembling a genuine head was an
observation by C. Todd on January 27 of a “diffused nebulous mass” which,
curiously, seemed to be separated from the tail. On the other hand, most
observers found that the tail simply dwindled away to nothing within the field of
view. For that reason, the comet is sometimes known as the “Headless Wonder”!
The essentially headless appearance of this comet made orbital
determination a very difficult task and a number of orbits have been given over
the years. Questions were even raised as to whether it was really a Kreutz object
at all, although the weight of evidence clearly supported its credentials in that
respect. The most reliable orbit for this object is the one determined by Z.
Sekanina, yielding elements very similar to those of the comets of 1843 and
1880 and indicating a perihelion passage of just 0.0048 AU on January 11.93,
1887.
Sekanina’s study revealed some remarkable behaviour explaining the
strange appearance of this object. His analysis of the tail found that it could best
be explained as a single synchrone, i.e. a single burst of particles of differing
sizes released simultaneously from the nucleus. Similar features are sometimes
seen as broad and diffuse streamers within the spreading dust tails of comets
(very notably, in Comet West of 1976), but are not to be confused with either
the “stria” such as displayed in spectacular fashion by Comet McNaught of
January 2007, or with the thin rays observed within the ion tails of many comets.
True synchronous streamers are quite rare and result from bursts of activity on
the surface of a comet’s nucleus or some similar event capable of producing
“puffs” of particles having assorted sizes. In the case of 1887 B1 however,
Sekanina demonstrated that the single isolated synchronous streamer that
constituted the entire tail was created by the total disintegration of the very small
nucleus just 6 hours after it passed through perihelion.
Because of the appearance of this comet, it is difficult to estimate what its
absolute magnitude may have been. Except for one very vague mention of a
weak condensation, there was really nothing to measure – just a free floating
streak of dust! Undoubtedly, the nucleus was very small; probably just 200 – 300
metres across and the comet’s absolute magnitude prior to perihelion well below
10. Sekanina gives 10 as the absolute magnitude after perihelion, albeit
hypothetically.
Following the 1887 comet, nothing was heard from the Kreutz sungrazing
group for decades. Then, quite unexpectedly, a sungrazer of very different
appearance showed up on photographs of the southern sky late in 1945.

 Fig. 4. The “headless” comet of 1887 as tracked through the southern sky.
C/1945 X1 (du Toit)
At long last, we have a sungrazer with an acknowledged discoverer! On
December 11, 1945, D. du Toit of Bloemfontein in South Africa found images of
a diffuse seventh magnitude comet on photographs of the southern sky. Rapidly
moving into twilight, the comet was only photographed by its discoverer on 4
more nights and, it seems, these were the only observations made of this object.
No tail was apparent on the photographs and the comet’s brightness suggests an
absolute magnitude of about 11.
The orbit of this comet turned out to be remarkably similar to that of the
comet of 1882, with perihelion on December 28 within 0.008 AU of the Sun. In
theory, it should have become very bright – albeit at small solar elongation –
around that date, before emerging as a spectacular object, for Southern
Hemisphere observers at least, early in the New Year. Alas, nothing of the sort
happened and the comet was never seen again. Not even a headless tail of the
1887 variety emerged from perihelion and there can be little doubt that the comet
completely faded away, probably even before perihelion was reached.

C/1963 R1 (Pereyra)
 On September 14, 1963, Z. Pereyra of the Cordoba Observatory in
Argentina discovered a comet in the morning twilight. He estimated the total
brightness as second magnitude and saw 1 degree of tail, though what (if any)
instrument was employed was not given in the discovery announcement.
Confirmation came from A. McClure of Hollywood on the 16th.. According to
McClure, twilight was already becoming bright when, using binoculars, he
noticed the long straight tail rising like a distant searchlight beam from the
dawn-lit horizon. When the head emerged, he initially did not recognize it as it
appeared simply as a small and condensed spot at the tip of the tail. The tail was
about 10.5 degrees long as recorded by McClure’s photographs and almost as
long visually, but his brightness estimate of the head was just magnitude 6,
considerably fainter than Pereyra had indicated.
It has been suggested that the comet may have been experiencing an
outburst at discovery but, while that is a possibility, the issue is actually more
complicated than this. There is a rather large scatter in the brightness estimates
of this comet and there is some thought that its low elevation and twilight
location may have led to consistent underestimates of its true brightness. For
instance, on September 20, G. de Vaucouleurs of the McDonald Observatory
estimated the comet’s central condensation as 2 – 3 minutes of arc in diameter
and of magnitude 6. Both of these values are very close to those given as the
total coma diameter and brightness by most observers around that date.
Apparently, de Vaucouleurs used a telescope for these estimates as he
specifically remarks that through binoculars, the coma measured about 15
minutes of arc in diameter and he suggested that the total integrated magnitude
was probably closer to 3!
Moreover, on September 20, C. Capen measured the tail as 18 degrees long
and gave the magnitude as 5.5 – 5.8, describing the head as “small” and
possessing an “ill-defined nucleus” as observed with 10x50 binoculars.
Apparently, the tail length was measured by naked eye, as Capen gave this
measurement separately from his comments about the comet’s head, which he
specifically stated was made using binoculars (I am indebted to John Bortle for
drawing this observation to my attention as a probable naked-eye sighting of the
tail). Also, according to the Wikipedia entry on this comet, famed English comet
and nova discoverer G. E. D. Alcock saw a pencil-like beam of light in the
twilight of September 12 (i.e. prior to the comet’s official discovery) which was
later found to closely match the position of the tail. All of these observations
suggest something brighter than many of the brightness estimates imply.
However, even the more conservative estimates still yield an absolute magnitude
almost as great as that of the Great Comet of 1843. Those of Pereyra and de
Vaucouleurs on the other hand, place the comet only a little fainter than that of
1882! Yet, if that is true, the comet must have faded rather rapidly, at least during
the early stages of its retreat from the Sun, as observations in November by A
Jones in New Zealand suggest brightness levels (magnitude 12.5 on November
23) closer to the “conservative” earlier estimates (although these visual
telescopic estimates may also have been of the inner coma if the outer regions
had grown very large and diffuse by that time).
Whatever the true story of the comet’s intrinsic brightness, it was clearly
numbered amongst the brighter members of the Kreutz group and it is a pity that
it was so badly place as seen from Earth. Perihelion occurred on August 24 at a
distance of just 0.005 AU and the orbit revealed a strong similarity with those of
the comets of 1843 and 1880. The perihelion distance was, however, slightly
smaller than either of these objects, actually being the smallest well-determined
for any major comet (that of 1887, like the values given for the SOHO comets, is
regarded as less well established).
The tail grew faint during the latter days of September, although S. C.
Venter in South Africa could still trace it for 4 degrees as late as October 1. In
common with the comet of 1882, but unlike many other members of the group,
the central condensation of this comet remained visible for an extended time. On
November 9, E. Roemer at the US Naval Observatory (Flagstaff)
photographically observed it at magnitude 17.2 when a possible secondary
condensation was noted just 0.1 minute of arc from the main nucleus. No further
observations of this feature were made, although the reality of the secondary is
generally accepted. Later observations of the main condensation were, however,
obtained by K. Tomita (Dodaira, Japan) on November 16 and 26 and again by
Roemer on December 14 and 18. On this last date, the brightness of the nuclear
condensation was estimated as magnitude 18.2. The relatively long period of
observation, together with the early intrinsic brightness of this comet, is
evidence that this was one of the larger members of the Kreutz family.

C/1965 S1 (Ikeya-Seki)
This comet has acquired the (not undeserved!) reputation of being one of
the brightest and most spectacular ever recorded. It is also one of the few comets
to have the honour of being commemorated in a musical composition.
At the time of its discovery however, little of this future glory was evident.
It was found, independently but almost simultaneously, by Japanese amateur
astronomers K. Ikeya and T. Seki on the morning of September 18, 1965 as a
small and tailless nebulous ball of magnitude 8. The first very preliminary orbit
suggested a perihelion near the Earth’s distance from the Sun but as further
observations were made, it started to become apparent that something more
remarkable was destined for this comet. By September’s end, it was clear that
this was yet another sungrazer, moving in toward a perihelion date of October
21.18 at a distance of just under 0.008 AU from the Sun, on an orbit that
appeared almost indistinguishable from that of the comet of 1882.
The comet brightened noticeably from one morning to the next. On
September 26, a photograph by James Young revealed a small circular coma that
had already sprouted a thin single ray of plasma tail 1.5 degrees long. By the
month’s end, it had crossed the naked-eye limit.
During the early days of October, Ikeya-Seki presented as a fairly typical
“gas” comet as it developed an ion tail consisting of a bunch of rays diverging
from the region of the nucleus. A photograph by Elizabeth Roemer at Flagstaff
on October 7 revealed this structure most strikingly. The dust tail was, at that
time, faint. The comet’s brightness continued to increase quite rapidly and on
October 11, NSW Government Astronomer Dr. Harley Wood saw the comet
with the naked eye from near the CBD of Sydney. Its brightness had risen to
magnitude 0.2 on the 16th. according to A. Page of Brisbane, Queensland, and by
then it had taken on more the form of a parabolic head and straight tail as
expected of a bright and dust-rich comet. Up to 10 degrees of tail were visible in
the morning twilight around that time.
Ikeya-Seki became visible with the naked-eye in broad daylight next to the
Sun on October 20. At 18:00 Universal Time on that day, G. de Vaucouleurs at
McDonald Observatory in Texas estimated its brightness as magnitude -10;
equivalent to the quarter Moon, but concentrated into a spot the size of a
planetary disk. Half an hour later, N. Roth and D. Fernald at the Smithsonian
Station at Organ Pass in New Mexico measured the tail as 1 degree long and
estimated its intensity as equal to that of the 25 ½ day old Moon, likewise visible
in the daylight sky. Two hours later, Roemer at Flagstaff saw 2 degrees of tail
and estimated the brightness of the head as -10 or -11 magnitude while C. Capen,
at 22 hours U. T. estimated the head as -10.5 magnitude as seen by the naked eye
and measured the tail as 4 degrees in length.
Many daytime sightings and daytime photographs were obtained from
around the world on that day. One photographer used an ordinary “box” camera
to snap the comet by simply obscuring the Sun behind a nearby tree branch. The
photograph clearly showed the comet as a bright star-like object sporting a tail of
1.5 degrees, readily visible against the bright sky background.
Astronomers at the Mount Norikura Coronagraph Station of the Toyko
Observatory watched the comet pass through perihelion on October 21,
describing it as “10 times brighter than the full Moon” at the time and noting a
disruption of the head into 3 separate pieces. The disruption began about half an
hour before perihelion and the 3 pieces had separated from one another shortly
after perihelion passage itself. Two of these apparently faded out as later
observations revealed just a single nucleus.
There appears to have been a temporary dimming of the comet soon after
perihelion passage, but the following day it had again become a naked-eye
daylight object as it drew away from the Sun. Telescopic daytime observations
were also made on the 22nd. and the following day saw the reappearance of the
comet deep in the dawn twilight. On the morning of October 25, observers at the
Smithsonian Station in Arequipa, Peru, observed a tail 20 degrees long and 3
degrees wide at the extremity, tapering from an almost stellar head of magnitude
-2. The greatest visual tail estimate seems to have been 45 degrees as observed
by R. B. Minton on the morning of October 28, but estimates of at least 30
degrees were common place during the latter days of October. A striking feature
of this tail was its high intensity over much of the length, plus a series of stria
that gave it the “corkscrew” appearance sometimes commented upon in the
literature.
The most spectacular views of the tail appear to have been those of October
31 and November 1, by which time the comet had receded far enough from the
Sun to be visible in a dark sky from suitable latitudes, yet remained close enough
to maintain a high level of brightness and activity. My most memorable view of
the comet was on the morning of October 21 as it rose over an ocean horizon in
a very clear sky. The head was about magnitude 2.5 and the tail stretched upward
into the sky like an intense beam of light for over 20 degrees. On that same
morning, John Bortle began to observe the comet when he noted that a thin mist
was rapidly thickening into a full fog. Soon, the sky had become so opaque with
the fog that stars fainter than magnitude 2.5 or 3 faded from naked-eye view.
Yet, the comet’s tail was neither dimmed nor diminished in length, continuing to
shine through the fog unhindered!
 Fig. 5. Ikeya-Seki, approximately 10 days past perihelion. Photograph
courtesy Victor
R. Boswell Jr., National Geographic Magazine.
On the morning of November 2, Bortle carried out an interesting series of
observations measuring the intensity of the tail at various distances from the
head by comparing it to the out-of-focus images of stars of known brightness.
Not far from the head, he found that the tail matched defocused images of
second-magnitude stars. At 8 degrees from the head (where the tail was still less
than a degree wide), its intensity matched those of third magnitude and around
20 degrees from the head (i.e. close to the furthest extremity of the tail) it
matched stars of magnitude 4.5. Close to that distance, the tail ended suddenly,
unlike the tails of the majority of comets which fade away more or less gradually
until they no longer can be discerned against the background sky. The head
itself, at that time, was about third magnitude and almost overwhelmed by the
intense tail.
During the early days of November, interesting changes were observed
within the head of the comet. Telescopic observations revealed the nuclear
condensation to have become distinctly elongated at the beginning of November
and on the 4th. of the month H. Pohn at Flagstaff found that it to be distinctly
double. A third and much fainter condensation was observed further along the
tail at that time, but seems to have faded out soon afterwards. Other possible
condensations were also reported around that date but these do not seem to have
been confirmed and apparently were very short lived. An interesting observation
was made by A. D. Andrews at Boyden on November 6 who found the
secondary component to be a close triplet on the negative of a photograph taken
of the split nucleus that morning.
The comet continued to fade during November, but like Pereyra before it,
there appeared to be quite a scatter in the range of estimated magnitudes toward
the end of the month. Most observers saw a small coma of about magnitude 7
during the closing days of November, however some in the southern hemisphere,
where the comet rode high in the sky during the second half of the month, saw
far larger outer comas and consequently made much brighter estimates. For
instance, J. Davies on Glenbrook in New South Wales estimated the naked-eye
brightness as between 3 and 3.5 with some 30 degrees of tail visible on
November 25. I can confirm that the comet – both head and tail – were visible
with the naked eye at that time and, in my opinion, the head looked very similar
to the globular star cluster 47 Tucanae, i.e. a condensed central “core”
surrounded by an extended outer coma, probably some 20 minutes of arc across
and having a total brightness in the same ballpark as Davies’ estimate.
Incidentally, photographs taken around that time recorded tail lengths as great as
35 degrees. A section of tail was visible with the naked eye on December 5,
although I could not detect the head without binoculars at that time. Just 10 days
later, no tail was visible and all I could find in 20x65 binoculars was a diffuse
mass at least 20 minutes of arc in diameter and showing no sign of central
condensation. Several nights later, only the faintest glow could be discerned,
looking more like a condensation stain on a glass window than a definite object.
Clearly, the latter stages of the comet’s period of visibility saw a dramatic fading
of its light and a rapid diffusing of material. The last positive observation came
on January 14, 1966 although a possible images of the comet was detected on
photographs taken with Baker-Nunn cameras as late as February 12. Even at this
later date however, predictions based upon the comet’s earlier brightness
indicated that it should still have been visible in small telescopes.

Fig.6. Comet Ikeya-Seki. Courtesy Maynard Pittendreigh.

C/1970 K1 (White-Ortiz-Bolelli)
Shortly after the Sun had set on the evening of May 18, 1970, Graeme
White of Barrick Point in New South Wales was sweeping the low western sky
for possible comets using a pair of 12x50 binoculars when he struck pay dirt; a
first-magnitude strongly condensed object sporting a short tail, located just 12
degrees from the Sun. He located it again on the 20th. by which date it had
retreated far enough from the Sun to be visible with the unaided eye in the
twilight sky, revealing some 10 degrees of tail.
The following evening saw independent discoveries by Air France pilot
Emilo Ortiz who spotted the comet whilst flying east of Madagascar and, just a
few hours later, in Chile by Cerro Tololo Interamerican Observatory technician
Carlos Bolelli whose attention was drawn to the long tail projecting upward from
beneath the western horizon. Many other independent discoveries were reported
around the same time as the comet was by then a conspicuous sight low in the
evening sky.
Tail lengths of 15 degrees and longer were being reported by May 24, when
the magnitude of the head was estimated as fourth magnitude. After fading from
naked-eye visibility, the comet nevertheless remained a striking sight in large
binoculars into the first week of June. On the 2nd. of that month, around 6
degrees of tail were visible in the 20x65 binoculars that I was using at the time,
tapering out from a very small head that was hardly distinguishable from the end
of the tail. By June 6, the tail had faded noticeably and diminished to about a
degree in length. Paradoxically, the head was seen more distinctly as a feature in
its own right on that evening, not simply as the sunward terminus of the tail.
Nevertheless, although distinct, the head was far from impressive, appearing
simply as a diffuse and transparent nebulous globule showing no signs of
condensation. It rather gave the appearance of swelling and diffusing away not
unlike Ikeya-Seki had done in December of 1965. A very rough estimate of the
brightness placed the comet’s head as magnitude 9. The same magnitude
estimate was given the following evening by A. Jones in New Zealand and, later
because of the longitude difference, by M. Jones in Queensland. A. Jones found
the head and tail to be pretty much indistinguishable at that time whereas M.
Jones saw the comet as just a faint patch of nebulosity. It is unclear if it had
become even more diffuse since my observation on the previous evening or
whether the difference in appearance was simply due to the difference in the
instruments used but, in any case, these observations by the Jones’ marked the
final time that this comet was seen. Having reached maximum elongation of just
22 degrees around that date, it soon began to slowly sink back into evening
twilight and searches in the morning sky following its emergence from
conjunction failed to detect any trace of the comet.
Because of the fairly brief observational arc of this comet, its orbit could not
be computed to the accuracy of those of Pereyra and Ikeya-Seki, but it is
nevertheless clear that it was a Kreutz sungrazer, albeit one having an orbit
differing to a certain degree from that of earlier members of this group, more so
from those of 1843, 1880, 1887 and 1963 than from the comets of 1882, 1945
and 1965. Perihelion occurred on May 14.49 at a distance of nearly 0.009 AU
from the Sun; rather large by the standards of Kreutz comets. According to the
brightness analysis by M. Meyer, its absolute magnitude was 6.9 and fading
according to the inverse 4.7 power of its distance from the Sun. It seems that the
comet’s brightness fell away rather faster than average as it retreated from
perihelion. Nevertheless, using these parameters and calculating the comet’s
brightness back to the time of perihelion yields a value of -17 for its peak
magnitude (necessarily very close to the solar limb) and brightness levels close
to that of the full Moon when the comet was about 1 degree from the Sun. The
lack of daylight sightings casts very grave doubts on these figures, but even
more telling is the lack of observations during the time that the comet was
approaching perihelion. The orbital geometry of the Kreutz group is such that a
member comet reaching perihelion in the middle of May is significantly better
placed prior to, rather than following, perihelion. Of course, these comets
normally display more spectacular tails after perihelion and for that reason it is
not surprising that White-Ortiz-Bolelli was widely seen then, despite the far
from favourable solar elongation, but the complete lack of earlier observations is
surprising. If Meyer’s brightness parameters also held prior to perihelion, the
comet should already have been magnitude 9.6 on April 1, when located 48
degrees from the Sun and visible in the western sky after twilight. As its
elongation decreased during April, its brightness should have increased quite
dramatically until the final evening of that month should have found it as bright
as magnitude 5.4 and located at an elongation of 22 degrees – the same as at the
time of its final observation in June by which time it was, however, some 3.6
magnitudes fainter.
Back in 1970, before the days of automated search programmes, visual
comet searches were being regularly carried out by many amateur astronomers
and the western evening sky was one of their most popular hunting grounds. It is
almost inconceivable that a comet could have been located in this region for over
a month whilst brightening from ninth to fifth magnitude without someone
having found it. It would have been the proverbial sitting duck. By comparison,
during late October of that same year a stump-tailed comet of magnitude 6 or 7
came out from behind the Sun and into the evening sky and was snapped up by 5
observers in the space of 2 days when at 34 degrees elongation (C/1970 U1
(Suzuki-Sato-Seki)). That alone proves that the relevant region of the sky was
being well patrolled back in 1970!
Although this sort of circumstantial evidence cannot be offered as absolute
proof, the early non-discovery of this comet surely arouses a powerful suspicion
that it’s pre-perihelion and post-perihelion behaviour differed. Likely, it was
initially very faint, did not become excessively brilliant near perihelion (nothing
approaching magnitude -17!) but surged significantly in intrinsic brightness very
soon after its closest approach to the Sun. I suspect that had the western sky been
scanned, during April 1970, using the type of wide-angle CCD instruments that
we have today, a very faint but rapidly brightening comet would have been
found racing toward the Sun and probably given no hope of surviving its
approaching ordeal, according to the opinion of most astronomers. It might have
reached telescopic daylight levels of brightness, probably after perihelion rather
than before and surprised everyone by emerging from its solar encounter as a
spectacular object. Does that scenario sound familiar? Think Christmas 2011
before answering that question!

TINY SUNGRAZERS OBSERVED FROM OUTER SPACE
In 1981, US Naval research scientists R. Howard, M. Koomen and D.
Michels were examining data from 1979 August 30 obtained by the SOLWIND
coronagraph on board Naval satellite P78-1, when they were surprised to find
images of a brilliant comet very close to the Sun. Fortuitously, Venus (then near
superior conjunction) was also in the field of view and this enabled the scientists
to estimate the brightness of the comet’s head. It appeared to them somewhat
brighter than Venus, possibly close to magnitude -4. The comet continued to
approach the Sun as shown in subsequent images, until it disappeared behind the
disc of the coronagraph about 2.4 hours later. The tail, however, remained visible
from behind the disc for several more hours and even brightened for a time. But
the comet’s head was never seen again. A notable brightening of the Sun’s
corona, roughly opposite the point at which the comet had disappeared behind
the disc, is widely (though not universally) supposed to have been caused by the
dust of the tail remnant forward scattering sunlight.
Initial orbital calculations indicated that this was yet another member of the
Kreutz sungrazing group of comets. They also yielded a perihelion distance
smaller than the radius of the Sun, implying that (had the comet persisted for
long enough throughout its dive into the Sun) it would actually have hit the
photosphere. Nevertheless, the chief investigator of the orbit, B. G. Marsden,
was not happy with this result. The problem was not that a comet could strike the
Sun, but rather that any realistic orbit also implied that it had struck the Sun at its
previous return as well – and that was impossible! It is amazing enough to have
comets pass through the lower corona and survive to do it all again another day,
but it is certainly out of the question to have one dive beneath the very surface of
the Sun and later re-emerge to repeat the performance at its next apparition. A
subsequent reassessment of the orbital computation by Marsden showed that the
comet did not, in fact, quite hit the Sun but passed just over 22,000 kilometres
from the photosphere on August 30.95 UT. This was still so close that the comet
completely disintegrated, probably even before reaching its perihelion.
Variously named Comet Howard-Koomen-Michels or (later) SOLWIND 1
and formally designated C/1979 Q1, this object was intrinsically faint; about
absolute magnitude 11.5 judging by its appearance in the SOLWIND images.
The question was raised at the time as to whether it was a rare object that simply
happened to coincide with the date of SOLWIND’s scrutiny of the immediate
solar environment or whether it was representative of an entire population of
intrinsically faint members of the Kreutz group.
At one level, the question was answered with the second alternative.
Continued examination of SOLWIND images found that other Kreutz sungrazers
had been recorded by this instrument on 1981 January 27, 1981 July 20, 1981
November 4, 1983 September 25 and 1984 July 28. Nearly 3 decades later, with
the images available on the Internet, R. Kracht found a further 3 fainter ones; in
data for 1981 November 20, 1983 July 7 and 1984 August 22/3. Furthermore,
following the destruction of P78-1 in the name of space-based defence, the new
Sun-monitoring satellite Solar Maximum Mission (SSM) found a further 10
between 1987 October 6 and 1989 September 28, after which it too was forced
to give up the ghost as its orbit deteriorated due to friction with the outer fringes
of Earth’s atmosphere. Like the 1979 object, none of these survived perihelion,
but unlike the earlier object, none left any residual brightening within the solar
corona. Clearly though, C/1979 Q1 was far from being a “one off” object!
On the other hand, none of these subsequent comets were as bright as the
initial one and the myriad of others discovered since 1996 (about which we shall
soon be speaking) came nowhere close to the magnitude of this comet. All of
these objects are now lumped together as “minicomets” or “minisungrazers”,
and the one in 1979 is normally included within that class, chiefly because of its
failure to survive perihelion. Yet, it was clearly an intrinsically brighter object
than any of the subsequent comets, many of which have absolute magnitudes
fainter than 20. Indeed, it was not very much fainter than the 1945 sungrazer
(which, we recall, also failed to show up after perihelion). Maybe – just maybe –
if the perihelion distance of C/1979 Q1 had not been so small, it might have
survived perihelion, at least for a short while, and appeared in the morning sky
during September. Because of the time of year, it would not have been
favourably placed but, given a bright enough tail, some comet hunter may well
have swept it up. Rather than being classified with the subsequent minicomets, it
may then have been remembered as the final member of the cluster of (“major”)
sungrazers that appeared during the middle years of last century.
A new milestone was reached late in 1995 with the launch of the Solar and
Heliospheric Observatory (SOHO). Unlike the earlier space-based solar
observatories, SOHO does not orbit the Earth but is parked at one of the
Langrangian points where, in theory, a body affected by gravity alone can
remain station with respect to two other objects. An enormous amount of
information about our nearest star has been acquired through the battery of
instruments on board this observatory. Serendipitously, it has also given us much
new information about comets; the Kreutz group in particular. With respect to
near-Sun comets, the chief sources of information have been the LASCO
coronagraph C2 and C3 images and, for comets in general, the SWAN ultraviolet
images of the solar wind. The latter has, with one interesting exception which we
shall come to later in this chapter, supplied little detail about sungrazing comets,
so LASCO has been the main player in the SOHO sungrazer saga.

Fig. 7. The “Christmas Comet of 1996” C/1996 Y1 (SOHO), also known as

SOHO-6,
December 23. Courtesy of SOHO/LASCO consortium. SOHO is a project
of
international cooperation between ESA and NASA.

Small sungrazers began to be found in SOHO data in 1996, soon after the
spacecraft had been placed in orbit around the Langrangian point. It probably
came as no great surprise that sungrazers showed up in the data transmitted back
to Earth – after all, both SOLWIND and SSM had already demonstrated the
presence of small members of the group. But what was surprising was the sheer
number of these objects, as well as how dim many of them were. Many of the
objects discovered in SOHO data were nothing more than small specks of light,
barely as bright as stars of seventh or eighth magnitude. Some were
demonstrably diffuse, others were not. Some had short and faint tails, others
were tailless. Every now and then a brighter one (perhaps as bright as second or
even first magnitude) with an impressive tail would appear. Such was the bright
object that appeared on December 23, 1996 and became appropriately nick-
named the “Christmas Comet”. It’s very small head was probably in the range of
magnitude 1 – 2 and it displayed a long bright tail that made it a spectacular
sight in the LASCO images. It looked almost like a Lilliputian Ikeya-Seki. Most
of the SOHO sungrazers were, however, far less impressive, but what they
lacked in spectacle, they certainly compensated for in number. Moreover, over
the years since 1996, their annual influx has been increasing, albeit not regularly
and not without fluctuations on the shorter time-scale. Overall though, the
numbers have increased quite dramatically; from about 50 per year in 1996 to
around 180 in the year 2010. Several years into the SOHO programme, the
general public was encouraged to participate in the search for comets in images
posted on the Internet and this led to an explosion in the discovery of comets,
including members of other “sunskirting” comet groups unrelated to the Kreutz
family. Nevertheless, as SOHO archived images remain accessible on the Web
and as these have also been the subject of close scrutiny by the growing army of
SOHO comet hunters, the level of accessibility is not the cause of the increased
number of Kreutz comets being found in more recent years. There is no doubt
that the increase is real and unrelated to any possible artefact of discovery.
Probably the most remarkable influx of Kreutz comets was the comet swarm
(some called it the “comet storm”) of December 2010 when 10 Kreutz comets
were discovered between December 3 and 9, with a further 25 arriving during
the following 10 days. This naturally aroused debate as to whether this event was
simply a statistical fluctuation or whether it presaged the arrival of a really large
sungrazing comet in the near future.
Anyone with even a passing interest in comets now knows what happened
just 12 months further on, but the display of December 2011 was preceded by a
continuing high influx of minisungrazers throughout that year. A couple of these
were rather noteworthy in their own right. One such object, C/2011 N3 (SOHO),
coming to perihelion on July 6, was not only bright by the standards of SOHO
discoveries but was also observed in ultraviolet light from the Solar Dynamics
Observatory satellite (SDO) as it transited the Sun’s disc. At that time of year,
Kreutz comets transit the Sun at the time of perihelion passage and 2011 N3 was
actually observed during the process of disintegration into a trail of evaporating
debris just prior to what, in theory, was its arrival at perihelion.
A second noteworthy comet in 2011 arrived at perihelion on October 1 and
briefly became what was arguably the brightest of the SOHO Kreutz discoveries,
possibly reaching magnitude -1 before fading away closer to perihelion. Then, of
course, the long awaited “Big One” arrived in the form of C/2011 W3 (Lovejoy),
which we shall look at in the following section. Arriving at perihelion on
December 16, Lovejoy was far from alone however. Some 31 other Kreutz
comets were found in SOHO data that month. Many comet enthusiasts, whilst
watching SOHO images of Lovejoy’s dash toward perihelion on December 16,
noticed the appearance of a small comet travelling slightly ahead of it, like a
pilot guiding it in to perihelion. Then, a little later, a similar small sungrazer was
observed following the big one. Moreover, although C/Lovejoy has been called
the “Great Christmas Comet of 2011” because of its spectacular display through
the Festive Season, there were actually 3 genuine “Christmas Comets” that year,
discovered in SOHO data on December 25. Clearly, December 2011 was a very
good month for sungrazers.
Small sungrazers have continued to arrive during 2012, although the
numbers have not been as great during the latter part of the year. Writing these
words in October, it is too soon to tell whether this is or is not a trend and, if it is,
what its possible significance (if any) might be. With one interesting exception
about which we shall be speaking in a moment, the comets of 2012 have also
been pretty faint and unimpressive.
The big exception was the strange object found, not in LASCO data but in
SWAN images by V. Bezugly on March 8. Several earlier comets were
discovered in SWAN data and many other known ones became clearly visible in
the UV images. It has been known since 1970 when the bright comets Tago-
Sato-Kosaka and Bennett were first observed at UV wavelengths from beyond
Earth’s atmosphere that the heads of comets are surrounded by very large
hydrogen clouds radiating at the wavelengths of ultraviolet light. The hydrogen
comes from dissociation of water molecules within the visible coma and it is this
extended feature of typical comets that renders them visible in SWAN images.
Nevertheless, all comets discovered or observed in SWAN data prior to March
2012 were non-Kreutz objects. No sungrazer – not even C/Lovejoy – had been
recorded in SWAN data and when it became clear that the new SWAN object
was moving in a Kreutz orbit, it is understandable that the level of excitement
rose amongst astronomers. Was this object set to become even bigger and
brighter than Lovejoy?
Perihelion was calculated for March 15.04 at a distance of just 0.0053581
from the centre of the Sun. In the days following discovery, several attempts
were made to observe C/2012 E2 (SWAN), as it was officially designated, from
the ground. One such attempt was made by Mr. Terry Lovejoy (the discoverer of
the great sungrazer of the previous year) who obtained images into the evening
twilight along the projected Kreutz track. At the time, the position of the new
comet was only very imperfectly known and an initial search of the images
failed to find anything comet-like, even though an eighth-magnitude star close to
the sungrazer path was recorded. Subsequently however, once a better
determination of the comet’s path had been computed, Lovejoy subjected his
images to a more thorough scrutiny plus computer processing and found that he
had indeed captured an image of the comet, appearing like an elongated star
deep in the twilight. This seems to have been the only ground based observation
of 2012 E2, but it makes this comet the third Kreutz sungrazer observed from the
ground thus far during the present century. Its two predecessors were 2011 W3
of course, plus the bright SOHO minisungrazer C/2008 C1, captured on one of
Miloslav Druckmuller’s deep images of the totally eclipsed Sun taken from
Mongolia during the eclipse of August 1, 2008. It is also interesting to note that
early March 2012 was the only time when 2 Kreutz comets were observable
from the ground at the same time, albeit only on computer-processed images. As
we shall see in the following section, the tail of 2011 W3 remained discernible
on enhanced images until the middle of March 2012.
Alas, those of us hoping for a repeat performance of the previous December
were to be disappointed. Comet SWAN did indeed enter the field of the
STEREO spacecraft on March 11 and LASCO the following day, but it quickly
became clear that this was not to be a second Lovejoy. It nevertheless gave a
nice display in the coronagraph images and probably reached a peak brightness
of around first magnitude, similar to the “usual” brighter SOHO objects but
fainter than the extra bright one of the previous October. Not surprisingly, it
failed to survive perihelion passage.
This comet has left us with a mystery. Why did this object produce enough
hydrogen to be observable in SWAN images when other bright sungrazers –
including significantly brighter objects such as that of October 1, 2011 and
(especially) 2011 W3 – did not? It was suggested that the comet might have been
experiencing a brightness flare (probably triggered by a solar event) at the time
of discovery. However, continuing observation over a number of days at UV
wavelengths indicated the sort of increasing activity expected from the “normal”
brightness evolution of a comet approaching perihelion. It seems likely that this
comet simply had more fresh ice exposed to the heat of the Sun than the majority
of sungrazers of similar size. If most of the other objects have a sufficiently thick
dust coating to delay their activation, they may not fully switch on until too close
to the Sun to be observed by the SWAN instrument. The “fresh” nature of 2012
E2 may be telling us something about the immediate origin of this fragment and
how it fits into the Kreutz scheme of things. This is something that shall be taken
up again at the end of the present chapter when the possible association between
2011 W3 and 2012 E2 is again raised. For the present however, it might be
interesting to compare the history of this object with that of the 1945 sungrazer.
That comet was also active well in advance of perihelion. Indeed, it was
significantly brighter than 2011 W3 at the same location in its orbit. Yet, it too
did not survive perihelion. We suggest that both this object and 2012 E2 were
simply small fragments – minisungrazers not greatly different from the brighter
SOHO objects – that had unusually fresh, icy, surfaces causing them to switch
on at distances from the Sun where most sungrazers of similar size remain
dormant or, at least, too feebly active to be visually discoverable or (nowadays)
show up on SWAN images. Incidentally, if this line of reasoning is correct, it
serves as a warning that if a new sungrazer happens to be discovered relatively
far from the Sun and appears rather bright in telescopes, it does not
automatically follow that it will evolve into a spectacular sight. The converse, as
we shall soon see, can also be true!
The discovery of the myriads of tiny sungrazers by SOHO has transformed
our picture of the Kreutz group from that of a relative few large comets to a
population exceeding the total number all other known comets. In recent years,
the STEREO spacecraft have also added to the total number of sungrazer
discoveries, although SOHO’s LASCO coronagraph continues to supply the
majority of discoveries, with seldom more than a very few days passing without
a new find. Several discoveries in a single day is not especially exceptional!
 Fig. 8. Comet STEREO-37, visible March 8 – 10, 2010, revealed a twin tail.
Courtesy
STEREO/NASA.
Nevertheless, despite the large number of these objects, their total mass is
small. Few appear to be as large as 50 metres and the majority are probably less
than 10 metres in diameter. The very faintest ones have been estimated as no
more than 5 metres across. If all the SOHO sungrazers were snowballed
together, the resulting sphere would be less than 150 metres in diameter.
So small are these objects that, initially, not all astronomers were happy
about calling them “comets”. “Fragments” of even “cometary fragments” were
preferred, however placing size limits in this way appears arbitrary and raises its
own set of problems. For instance, although we admit that the SOHO objects
truly are fragments, that does not make them any different from, say, Ikeya-Seki
or the Great Comet of 1882. They were also fragments of former comets, but to
object on these grounds to the classification of them as comets would be simply
absurd! Again, the behaviour of the SOHO objects – sublimating ices, gas/dust
comas and even tails in many instances – certainly looks like cometary activity,
so why should the objects displaying this not be called “comets”? After all, if
something has feathers, waddles like a duck and quacks like a duck .... Happily,
the critics lost the day and I doubt if anyone is unhappy about naming the SOHO
objects as “comets” today!
The SOHO comets repeatedly display rather interesting behaviour with
respect to their brightness evolution and tail development, where the latter is
present. In the majority of instances, these comets initially brighten very rapidly
as they approach the Sun. This surge of activity stops well short of perihelion
however, with maximum brightness usually being reached between about 10 and
13 solar radii from the Sun’s centre, i.e. about 0.047 – 0.06 AU from the Sun.
Brightness levels off and then starts to fade, presumably fading out completely
sometime prior to perihelion. Apparently, maximum brightness does not coincide
with the peak of tail-dust production, as studies by Z. Sekanina show that the
form and orientation of the tails displayed by these comets are best explained if
these consist of dust released between 20 and 30 solar radii, or 0.09 – 0.14 AU,
from the Sun with little contribution thereafter. This behaviour may provide
valuable insights into the behaviour of comets at very small solar distances. Or it
might be telling us something about the survival of tiny dust particles released at
distances smaller than 0.09 AU from the Sun.

C/2011 W3 (LOVEJOY)
On the morning of November 27, 2011, Queensland amateur astronomer
Terry Lovejoy captured a faint cometary image of magnitude 13 during the
course of his automated sky patrol. Despite its faintness, the new comet was
moving rapidly and heading into the morning twilight. Lovejoy obtained a
second image, and Alan Gilmore and Pamela Kilmartin of Mt John Observatory
in New Zealand officially confirmed the comet on December 1, with the official
announcement being made the following day. Early visual observations by the
comet’s discoverer and Michael Mattiazzo in South Australia indicated a rapid
brightening. Mattiazzo also noted that the comet appeared to be following the
Kreutz track, something which was soon confirmed as more positional
measurements were obtained. Perihelion was calculated for December 16, at
which time the comet was predicted to pass just 0.0056 AU from the Sun.
There were not many observations in early December as the comet plunged
into morning twilight. Nevertheless, the few magnitude estimates that were
obtained indicated a steep intrinsic brightening trend; far steeper than Ikeya-Seki
during its pre-perihelion phase. The absolute magnitude was, however, very
faint. Around the time of discovery it was just 15, some 4,400 times fainter than
Ikeya-Seki and even around 40 times fainter than the 1945 sungrazer under
similar circumstances. Clearly, even granted its rapid rate of brightening, it
lagged well behind earlier sungrazers observed from the ground. For this reason,
most observers looked upon it as being in the same league as the SOHO
minisungrazers rather than the larger objects of earlier decades. It is doubtful if
anybody expected it to survive perihelion, although John Bortle raised the
(small) possibility that an 1887-type headless tail might emerge from perihelion.
Only super-optimists would have considered anything more than this a serious
possibility however.
This “pessimism” has been criticised in hindsight, however the comet’s total
destruction prior to perihelion really did seem the most likely possibility at the
time. Comparison with the 1945 object suggested as much. If that comet
disappeared at perihelion, leaving not even a headless tail remnant to emerge
from the encounter with the Sun, what reason was there to think that this
intrinsically fainter object should fare any better?
A photograph taken at the end of the first week of December, by which time
the comet had become a difficult visual object in the twilight, showed a striking
similarity with James Young’s image of Ikeya-Seki on September 26, 1965. Both
comets revealed small and very compact heads with ion tails consisting in each
instance of a singly ray emerging from the centre of the coma. The big difference
being that when Young photographed Ikeya-Seki it was still almost a month out
from perihelion whereas Lovejoy had only about a week to go. Though it was
closer to both the Sun and the Earth than Ikeya-Seki had been at the time of the
Young photograph, Lovejoy’s level of development was clearly lagging far
behind the earlier comet andt its apparent brightness was fainter by some 2
magnitudes. Still, the similarity, as well as the differences, in the evolution of
both comets is interesting.
From December 13, the comet entered the field of view of the STEREO
spacecraft and during the middle part of the month, while too close to the Sun
for ground based observation, it was closely monitored by a battery of space-
based solar instruments. Although the rate of brightening slowed somewhat, the
comet reached a total magnitude of near -3 while some 0.05 AU from the Sun,
after which it began to fade. This is very similar behaviour to that of a typical
SOHO sungrazer and seemed to reinforce the majority opinion that this comet
was effectively an unusually large member of the same class of object. The tail
was, however, bright and in addition to the main tail a faint, narrow and straight
secondary was visible in SOHO images. No previous sungrazer had displayed a
similar feature visible in SOHO instruments, although 2 earlier Kreutz objects
(STEREO-23 and STEREO-37) did show similar tails in STEREO data, on
December 30/31, 2009 and March 8 – 10, 2010 respectively. The tails of these
STEREO comets, as well as that of Lovejoy, are generally assumed to have been
plasma (ion) tails, however it is not clear if they displayed the fine ray structure
common to these appendages. It has been suggested that they may have been
neutral iron tails similar to the one observed in STEREO images of C/2006 P1
(McNaught). The iron atoms comprising this tail are thought to have been
supplied by the evaporation of grains of troilite (FeS), a mineral first found in a
meteorite.

Fig. 9. Comet Lovejoy imaged by STEREO-A, December 14, 2011.

Courtesy
STEREO/NASA.

Whatever the nature of the secondary tail, an amazing change soon became
evident in the main one. After the head of the comet disappeared behind the
occulting disc, the main tail detached completely from it and drifted through
space as a headless arc of dust! Had the comet finally succumbed to the rigors of
perihelion?
That was the most reasonable conclusion. Or so it seemed! But sometimes
the thing that appears most reasonable is not that way at all! In what John Bortle
described as the most amazing even in the history of comets, a small starlike
remnant of the head (minus all trace of tail) emerged from the opposite side of
the occulting disc, speeding away from the Sun. The comet’s tail was on one side
of the Sun and its head on the other!
 Initially, the head was faint, but it quickly brightened and within hours was
outshining the preperihelion brightness peak. Moreover, a new tail started
forming, once more making the comet a striking sight in the coronagraph
images. When the comet was about 4 degrees from the Sun, it had become so
bright that several telescopic and photographic daylight observations were made,
although there do not seem to have been any naked-eye sightings similar to those
of the sungrazers of 1843, 1882 and Ikeya-Seki. The following day, the comet’s
discoverer estimated its brightness as -1.2 magnitude and other estimates in the
low negative magnitudes were also reported around the same time. The nuclear
condensation appeared as a very small and intensely bright disc during the days
immediately following perihelion, but between December 19 and 20 it
experienced profound and dramatic changes that radically altered the appearance
of the comet. The previously bright discrete condensation suddenly elongated
into a bright streak or “spine” extending away from the position of the former
nucleus and into the principal tail. What just about everybody expected to
happen just prior to perihelion actually occurred some 36 hours later; the nucleus
totally disintegrated into a cloud of dust which, in the days following, elongated
into a long bright streak. According to Sekanini, the destruction of the nucleus
was most probably caused by thermal stress, but because it took time for the heat
to penetrate down toward the core of the nucleus, the final breakup was delayed
until well after perihelion had been passed. The loss of the nucleus did not,
however, mean a rapid fading away of the comet’s head or (still less) of the tail.
Although the head of the comet appeared weak and uncondensed as it became
increasingly visible in the morning twilight, the comet’s absolute magnitude was
estimated at around 10 following perihelion, representing a brightness increase
of 100 or thereabouts compared with early estimates prior to perihelion.
Moreover, although it did fade rather more quickly than average, the brightness
decrease with increasing solar distance was still a good deal less steep than the
increase in brightness had been before perihelion.
 Fig. 10. Dramatic changes occurred in the nucleus of Comet Lovejoy
between
December 19 (image on left) and December 21. Courtesy Jakub Cerny, Jan
Ebr, Martin
Jelinek, Peter Kubanek, Michael Prouza and Mikal Rings, Czech
Astronomical Sosiety.


Within a week of perihelion passage, the comet had evolved into a glorious
sight in the early morning skies for observers south of the equator. My first good
view came on the morning of December 23 (late in the UT day of December 22)
as the tail rose from behind the horizon like the beam of a powerful searchlight.
Against the background of a dark sky, the tail was estimated as measuring 21
degrees long and delicately curved, with a fainter straight tail separating from it.
 Fig. 11. Comet Lovejoy photographed from the International Space Station
as it passed over Tasmania on December 21. Courtesy Dan Burbank/NASA



A curious phenomenon regarding the tail was evident on that morning. It
appeared to be increasing and decreasing in brightness simultaneously over its
entire length; fluctuating in the manner of an aurora! This effect had also been
remarked upon by at least one observer of Ikeya-Seki, although I must admit that
I did not notice it then. But with Lovejoy, the effect was very obvious. Similar
events have appeared in the records of historical comets. Sometimes a sort of
aurora-like flickering almost instantaneously traversing the length of the tail has
also been reported and it is likely that this effect – termed “coruscations” in the
older literature – is closely related to what I saw. It is difficult to accept these
events as actual processes within the comet tails themselves. For a tail as long as
Lovejoy’s to be affected simultaneously over its entire length, some effect would
need to travel along it at speeds exceeding that of light. The Earth’s atmosphere
is usually named as the culprit and the effect explained as being similar to the
scintillation of stars, although the large size of the extended objects affected (21
degrees for Lovejoy and as much as 30 degrees for Ikeya-Seki) requires
something of entirely different scale to that affecting the point sources of stars. It
is also possible that some form of optical illusion is involved, although I have
not actually heard that idea put forward and, quite frankly, I would take some
convincing that what I saw was illusory. As for clouds of thin atmospheric mist,
I can only remark that at the time I noticed this effect the sky was perfectly clear
and the temperature quite warm. The morning the Ikeya-Seki observation was
made was also warm, very clear, and without any trace of cloud, haze or mist. In
this respect too, it is worth mentioning that neither morning had much in
common with the cold frosty skies normally associated with excessive
scintillation of stars. It would be interesting for anyone reading these words to be
aware of this phenomenon next time a bright comet appears and, if similar
effects are seen in the tail, to note the meteorological conditions prevailing at the
time and anything else that might possibly be relevant and report the observation
to an internet forum of comet observers. There is unlikely to be anything very
profound awaiting discovery here, but it is intriguing as one of the “little
mysteries” that tend to be passed over very quickly but which continue to elude
satisfactory explanation.
Coming back to Comet Lovejoy after this diversion, it was notable that the
comet’s head on that morning was surprisingly weak compared with the
magnificent tail. When it appeared over the horizon, it was nothing more than a
small blob at the end of the tail and estimated as only about magnitude 4.5. By
the morning of the 28th. the head seemed to be just a slight brightening at the
sunward extremity of the tail and was estimated as no brighter than magnitude
5.3 with the naked eye. The tail, by contrast, was conspicuous with the naked
eye (even though it now extended along the Milky Way band and so had
considerable “competition”) and extended for a full 27 degrees.
By the last days of December, the naked-eye tail had become essentially
headless. For instance, on the morning of December 30, it stretched for 34
degrees as seen with the unaided eye, yet the “head” (a slight “swelling” at its
sunward tip) required opera glasses to see and was estimated as magnitude 6.3.
On the morning of New Year’s Day, the naked-eye tail was measured as 39
degrees, but the magnitude of the head just 6.6 as determined with the aid of
6x35 opera glasses. Oddly, the tail seemed very faint on the following morning
when just 26 degrees could be seen by eye, despite a very clear sky. On the other
hand, the next morning found it far more easily visible. The “head” (or sunward
tip of the tail) was hidden behind a cloud, but the naked-eye tail extended a full
42 degrees from the ephemeris position of the head. Rob McNaught at Siding
Spring Observatory also noted that the comet seemed fainter the previous
morning. My final naked-eye observation was on the morning of January 5th.
(January 4.69 UT), when I traced the tail very faintly for 31 degrees. The
brightness of the head, as estimated in the 6x35 binoculars, was then 7.1. On that
same morning, Michael Mattiazzo in South Australia saw the tail naked eye
using averted vision and traced it for 30 degrees photographically. Two days
later, just 5 degrees of tail were glimpsed by Rob McNaught at Siding Spring
with the naked eye using averted vision.
The Full Moon became a problem following the abovementioned
observations, although fortunately the comet passed close to the south celestial
pole and shifted from predominantly morning to predominantly evening about
the same time that the Moon vacated the evening sky. The break in the
observational record was therefore kept to a minimum, although after the first
few days of January, most observations were wide field images of the tail rather
than visual sightings. R. Kaufman and L. Barnes were the principal imagers of
the comet’s tail in the weeks that followed. According to their observations, the
greatest tail lengths were recorded between January 14 – 16. Several images
revealed between 45 and 46 degrees of tail on those dates, with the greatest
length of 47 degrees recorded by L. Barnes on January 15.5 UT. An odd feature
of the comet’s appearance in these and later images was the lack of anything
visible at the exact ephemeris spot of the comet. The tail stopped short of the
position predicted for the head! Indeed, even visual observations around the New
Year found the brightest part of the tail a little displaced in the anti-sunward
direction from the head. This may have been similar to the dark “breaks”
reported by Todd in the comet of 1887.
On January 16, I barely managed to see the comet in a 25x100 binocular
telescope. All that I could detect was a very slight band of misty light that
appeared a little brighter at one end. I judged the brightness of the “brighter”
region to be magnitude 9.6 and deemed this to be the head, although it did seem
a little displaced in the tailward direction from where I would have expected to
find the actual “nucleus” – had it still been in existence. My only other sightings
were on January 22 and 23 when the field of the binocular telescope appeared to
be crossed by what could best be described as a band of sky that was very
slightly less black than the rest of the background sky. To call it a band of light
would be a gross exaggeration, but it did appear to lie at the position and have
the correct orientation of the tail. Nothing resembling a head was visible.
Also on the night of January 22, Alan Hale (of Comet Hale-Bopp fame)
turned his 41-cm reflector toward the comet’s predicted position and suspected
“something” at the limit of vision. He described it as “an extremely vague pale
glow” but suspected that it was moving along the comet’s predicted path. It was,
however, little more than a slight brightening of the background field and he was
less than convinced of its reality. So, it must be added, were most observers who
read his account. A relatively deep CCD image of the ephemeris position taken
by Paul Camelleri about the same time revealed nothing. After a brief spell of
poor weather, Hale tried again on January 25 and confirmed his earlier
observation. The object – which he described as nothing more than a vague
“presence” – had the right position and direction of motion to be the comet. He
even suspected occasional hints of “sky brightening” in the predicted orientation
of the tail. The “head” appeared to him to be about 3 minutes of arc in diameter,
although on occasions it seemed around twice that size, and he estimated its
magnitude as about 12, commenting however that it was a far more difficult
object than the average twelfth magnitude comet. Two other observers, David
Levy and Thom Peck, also managed to see the comet from Jarnac Observatory in
Vail, Arizona, on January 27. They also used a 41-cm telescope and described
the comet in similar terms to Hale; a faint almost imperceptible glow trailing off
in the direction of the tail. Theirs appears to have been the last visual observation
of Comet Lovejoy.
 Wide-field images of the tail were, however, made throughout February,
principally by Kaufman and Barnes in Australia, but also in the Northern
Hemisphere by Man-To Hui and colleagues in northern China. On February 12,
Kaufman traced the tail for some 7 degrees, but it had contracted to just 1.2
degrees as imaged by Barnes as late as March 16.
This comet has left us with several questions, possibly the most obvious
being how such an intrinsically faint object could blossom into the magnificent
display that greeted us just prior to Christmas. Apparently, the comet switched
on activity only when well within the Earth’s distance from the Sun and then
brightened unusually rapidly, but why did it behave that way and is this sort of
behaviour typical or atypical of Kreutz comets?
Previously, the sungrazer that had been most extensively observed prior to
perihelion was Ikeya-Seki. This object brightened approximately as the inverse
fourth power of its distance from the Sun which is often considered the average
for long-period comets (in so far as “average” makes sense when speaking about
objects as seemingly capricious as comets!). The sungrazers of 1843 and 1882,
also observed prior to perihelion although not as well as Ikeya-Seki had been,
seemed to behave more or less similarly, as far as we can determine. On the
other hand, the minisungrazers found in SOHO data surge in brightness far more
rapidly.
When Lovejoy was discovered, it was assumed that its light curve would
follow that of Ikeya-Seki, albeit with a far fainter absolute magnitude. When it
became obvious that it was brightening more like a SOHO comet (although
admittedly not as steeply as many of these) it seemed to confirm the suspicion of
many that this really was a minisungrazer after all – simply an unusually bright
one. Yet, after it emerged as a spectacular object, the question was raised as to
whether some of the spectacular comets of earlier years may have followed a
similar evolution of brightness. We have already spoken about the odd fact that
White-Ortiz-Bolelli approached perihelion in the western evening sky during a
period when many comet hunters were active, yet somehow eluded discovery
until after perihelion. It is also worth noting that the comets of 1880 and 1887
were likewise seen exclusively after perihelion, even though they were well
placed on both their inward and outward treks. Maybe Lovejoy is not the
anomaly. Perhaps Ikeya-Seki is!
Although this topic will be taken up more fully later, there are good reasons
for thinking that Comet Lovejoy separated from a larger fragment near the time
of its previous aphelion. We can speculate a little here as to why the comet may
have behaved as it did, remembering that this is only speculation. Thus, having
passed perihelion several centuries ago, we suggest that the proto-fragment had
most of the volatile materials cooked out of its top layers and therefore was
effectively covered with a non-volatile crust preventing further sublimation of
the ices lying deeper beneath its surface. We may suggest that this is the reason
that most of the Kreutz sungrazers have faded quickly soon after their peak
display. In some instances (as with Lovejoy for example) this has marked the
total disruption of the comet, but that may not have been the reason in every
case. But be that as it may, if we assume that the Lovejoy proto-fragment was
covered by an insulating crust, when it broke into several pieces around the time
of aphelion, many of these fragments would also have been partially covered;
the exceptions being fragments of fresh ice from deep within the proto-fragment.
Suppose the proto-fragment was elongated in the manner of the nuclei of
comets Halley and Borrelly, and further supposing that its axis of rotation lay
very close to the plane of its orbit, it would be possible for the Sun-ward tip of
the elongated nucleus to break away and approach the Sun quite closely without
any fresh ice becoming exposed to solar heating. This broken “tip” – shaped
more or less like a bullet – would still be encased in its refractory coat, except
for the “base” that had broken away from the main fragment. But if it
approached the Sun “tip first” rotating about its long axis so as to have the icy
underside of the base always turned away from the Sun, it could conceivable
remain dormant until the crust became so hot that some of the heat started to
penetrate to the underlying ices and trigger sublimation. As the pressure of gas
built up beneath the crust, we could imagine sections of this giving way under
the pressure, thereby allowing more heat to reach the underlying ice. This
ensures even greater sublimation, more gas and more crustal material blown
away ... and so the self-perpetuating process gathers momentum. The
observational effect of this would be a late switching on of the comet followed
by a rapid increase in its brightness as the crust was progressively purged and
fresh underlying ice exposed. A process such as this would explain the late
activation, initial extreme faintness and steep brightening curve of Comet
Lovejoy.
It is also interesting that the same scenario could explain the behaviour of
Comet SWAN. We suggest that this was a relatively small but very icy fragment
from deep within the broken proto-fragment. Being essentially a lump of fresh
ice, it became active unusually early but was completely consumed as it swept
through perihelion passage.
 By the way (and this topic will be taken up in the final chapter), the fact that
there are some orbital differences between comets Lovejoy and SWAN does not
preclude these two objects from having broken away from the proto-fragment
about the same time. Fragments splitting away near aphelion can differ quite
significantly in their orbital elements, although their periods will alter little and
they will arrive at perihelion only months apart, as indeed was the situation with
Lovejoy and SWAN.
In view of the increase in the numbers of SOHO comets in the years
preceding the arrival of Comet Lovejoy, it seems likely that most or all of these
were spawned by the same proto-fragment, though probably from schisms at
different sections of its orbit, giving their arrival times at perihelion a greater
temporal spread. By the way, we need not think of this proto-fragment as being
very massive. An object of just 600 metres, or thereabouts, across would contain
enough material to spawn all the minisungrazers seen to date – and many more
besides – plus, maybe, a couple of other Lovejoy-sized objects as well.
With these speculative thoughts, we conclude our account of those comets
that certainly belong to the Kreutz sungrazing family. But it is almost certain that
the family is larger even than depicted here. A surprising number of older comets
have been suspected, for one reason or another, as having been members of the
group. The “Kreutz potential” of some of these has been pretty weak, although
others appear very likely to have been bona fide members and some may even
have been the immediate parents of ones seen in more recent years. These
possible candidates for Kreutz membership form the subject of the following
chapter.


IV
WERE THESE COMETS SUNGRAZERS?

In the previous chapter, we looked at the comets whose Kreutz membership
credentials are beyond all reasonable doubt. Nevertheless, it is widely believed
that the Kreutz family of comets is considerably larger than this list of “proven”
members. Throughout the centuries – right back as far as human records of
astronomical events extend – records are to be found of objects that seem to have
had some of the characteristics commonly associated with Kreutz sungrazing
comets. They may have been in the right part of the sky for a Kreutz comet
coming to perihelion around that time or they may have been noted as having the
familiar appearance of a Kreutz comet, namely, a long, narrow but unusually
intense tail emerging from a small and maybe rather insignificant-looking head.
Or a comet may have been recorded as having been visible in broad daylight
close to the Sun or very low in twilight just after sunset or before sunrise. Other
things being equal, all of these features could be signs that the comet in question
was a sungrazer. But, of course, without sufficiently accurate and/or numerous
positional measurements to derive an orbit for these objects, a question mark
must remain over their Kreutz credentials. Moreover, not all of the “potential”
Kreutz comets have family-membership credentials of equal weight. Some are
almost certainly sungrazers, others are “probables”, yet others mere “possibles”
while some that have been proposed at different times can almost certainly be
rejected.
The main lists of possible Kreutz candidates were drawn up by I. Hasegawa
in 1979 and K. England in 2002. The latter contains all the former’s proposals,
plus others that the author has gleaned from general comet catalogues according
to his rather liberal criteria. These include suggestions made by earlier writers
going back to Kreutz himself, so it will be worthwhile to begin by looking at the
England list, later adding a small number of other suggestions that he did not
include.
England’s list only includes objects that appeared prior to the early
eighteenth century (which means that a couple of proposed sungrazers of later
years were not included) but it is especially useful in that its author attempts to
quantify the possibility of each listed object really being a Kreutz member. He
assigns each entry a “score” between 0 (a positive non-Kreutz) and 10 (a definite
member). Not surprisingly, no 0 entries appear in his list, as these have been
eliminated before the list was made. There are no scores of 10 either, probably
because true certainty really requires an orbit to be calculated, and none of the
entries in his list was observed sufficiently accurately to allow this. In most
instances, scores of less than 5 can probably be scratched from the list and the
majority will not be discussed here. However, there are a few that do require
some further remarks for one reason or another, and these are given below with
the specific entry. The most interesting objects though, are those with scores
higher than 5, and especially, the ones with scores of 7 or higher. These will be
discussed in more detail below. Some of them are widely accepted as bona fide
sungrazers and as we shall see in the following chapter, have been included in
the evolving Sekanina-Chodas model of the sungrazing family’s formation.
England’s list is as follows; the number given in parenthesis following each
entry being the probability score that he assigned to that entry:
Eclipse phenomenon of 1375 (?) BC (4). [Ancient Syrian tablet records a
solar eclipse, stating that the “Sun goddess set with Rashap as her gatekeeper”.
This was considered to be a bad omen. “Rashap” is identified with Mars,
however England argues that Mars is rather faint and too well known for its
presence near the Sun to be considered such a bad omen, and he suggests that
“Rashap” in this instance may refer to a comet near the Sun. This seems to me to
be reading too much into too little. The position of the red planet – so often a
symbol of blood and war – next to the darkened Sun may have been omen
enough, without the need to introduce a comet!]
Eclipse phenomenon late second millennium BC (2). [Eclipse seen in
China, during which “three flames and a great star” were seen. The “flames were
probably coronal features or even prominences, and the “star” may have been
Venus. Evidence for a comet is weak].
Great comet of 372 BC (7).
Eclipse comet of 117 BC (?) (5).
Comet 110 AD (3).
Comet 133 (5).
Great Comet of 191 (4). [Said to have had a tail 100 degrees long in the
morning sky in October or November. Although the region of sky corresponds to
a sungrazer at that time of year, a sungrazer’s tail would have been pointing
more or less away from Earth and somewhat foreshortened. A length of 100
degrees seems excessive, and more like an Earthgrazer or, if not exactly an Earth
grazer, at least a close approach with the tail seen broadside.]
Roman phenomenon of 193 (stars seen “near Sun”) (2).
Comet 252 (4).
Comet 254 (4).
Comet 283 (2).
Daylight comet 302 (4). [Observed in China some time during May or June
& record repeated in one Korean chronicle. One translation says that it was seen
“in the morning”. There may be some ambiguity in the wording here. A Kreutz
comet may have been visible after sunset, especially if it appeared within the
early days of the time period].
Roman daylight “star” (meteor train?) of 334 (2).
Great Comet of 349 (2).
Great Comet of 423 (5).
Great Comet of 467 (5).
Comet of February 501 (3).
Comet of April 501 (3). [This may have been same as above. If so, then
definitely not
a Kreutz].
Comet of 560 (1).
Comet of 582 (3).
Eclipse “star” seen in Ireland in 612 (2).
Comet of 675 (3).
Comet of 762 (1).
Object resembling “two moons joined together” seen 813 (1).
Comet of 815 (3).
Another object (this time called a comet) like “two moons joined
together” in 822
(1).
Comet 840 (1).
Comet 852 (5).
Comet 867 (2).
Comet 892 (3).
Comet 939 (2).
Comet 943 (3).
Comet 947 (3).
Comet 948 (2).
 Comet 957 (4).
Comet 961 (3).
Comet 1034 (4).
Comet 1041 (2).
Comet 1041 October/November (3).
Comet 1056 (1).
“Star” near Sun (possibly Venus) 1077 (1).
Comet 1080 (2).
Great Comet of 1106 (7).
“Star” seen just before sunrise, 1137 (1).
Thin crescent with 3 “stars” near Sun February 20, 1148
(meteorological
phenomenon?) (1).
“Star” near Sun in daylight August 1, 1179 (2).
“Comet” on Sun (!), May 1, 1184 (possible sunspot or even white-light
flare?) (1).
Comet 1232 (4).
“Star” seen in daylight on March 30, 1282 (2).
Comet in Virgo, October 1314 (2) [This was almost certainly a misplaced
record of
comet of 1315. Definitely not a Kreutz]. Revised score (0)!
Comet 1368 (4). [Tail described as broad and like a pyramid. Does not
sound like the
tail of a Kreutz comet].
Comet 1381 (4).
Comet 1392 (3).
Comet 1406 (1).
Comet 1407 (3).
Comet 1434 (2).
Comet 1529 (3).
“Star” seen in daylight in Japan on August 30, 1587 (3).
Southern comet 1668 (8).
Southern comet 1689 (9).
Southern comet 1695 (9).
Southern comet 1702 (9).

Of the pre-seventeenth century objects catalogued by England, Hasegawa
listed those of 133, 191, 252, 423, 1034, 1106, 1232 and 1381 as possible
sungrazers and, although he did not name it in his specific list of Kreutz
candidates, a note that the comet of 302 might have been a Kreutz sungrazer was
included with that comet’s entry in his general catalogue of ancient comets.
In addition to England’s list, B. G. Marsden noted that a comet seen in
November 1075 may have been a sungrazer, as well as mentioning another
possible member of the family seen in February 1666 and a daylight star-like
object witnessed near the Sun by several people at Broughty Ferry in Scotland
between 10 and 11 a.m. on December 21, 1882.
The first of these was located in the right part of the sky, but judging by the
account of its direction of motion, was almost certainly moving toward
perihelion. If this object was a sungrazer, it is likely to have become a very
spectacular sight following perihelion passage, however the comet disappears
from the records after sinking into morning twilight, with no hint of a striking
display later. A southerly declination is probably not the whole story for the
complete lack of post-perihelic records. Carrying on England’s system of scores,
I would think a 3 might be a fair probability estimate.
A report that a bright comet sporting a long and narrow tail was observed in
Ceylon during February 1666 has often been dismissed as a misplaced reference
to the 1668 comet. The description does sound like the latter comet, and (we
might add) like any bright Kreutz comet! Marsden appeared inclined to believe
that the date was right and that this was a separate object and, it should also be
noted, a comet was also observed in Korea at the time. G. Kronk concludes that
the Korean and Ceylonese comets were the same. On the scale of scores, I would
be inclined to rate this as a 6 or even a 7 as a possible sungrazer.
The object of December 21, 1882 is interesting in view of the Kreutz
activity of that decade. It was said to have been visible near the Sun, to have had
a “milky” appearance with the naked eye (in contrast to the sharp appearance of
a star at night) and to have revealed a crescent shape when viewed through a
spyglass. The possibility that the object was Venus has been mooted, but this
opinion has also been criticized on the grounds that the object was apparently
closer to the Sun than the position of Venus at that time. A sungrazer arriving on
or around December 21 would, however, be well placed for southern observers
later in December. It would, indeed, follow a path similar to that of Comet
Lovejoy in 2011 and, because the 1882 object was visible in daylight by eye
alone whereas Lovejoy required optical aid, we might assume that (if a
sungrazer) it must have been intrinsically brighter and presumably capable of an
even more spectacular display after perihelion. As the Great Comet of 1882
continued to be well visible even in late December, observers in the southern
hemisphere would have had the unique opportunity of seeing a pair of bright
Kreutz comets simultaneously in the same general region of the sky!
Unfortunately, nothing of the sort happened, which alone tends to cast doubt
on the Kreutz credentials of the object. My own feeling is that the December 21
object was probably a comet, though not a sungrazer. An intrinsically rather faint
object of small perihelion distance that both approached and receded from
behind the Sun more or less along the line of sight may explain its sudden
daylight appearance coupled with a lack of sightings in the evening or morning
twilights. We might assume that the object came from behind the Sun and spent
only a brief time near perihelion between Earth and Sun (albeit far closer to the
latter) at high phase angles where forward scattering of sunlight from dust grains
close to the nucleus greatly enhanced its apparent brightness. Following
perihelion, the comet whipped back behind the Sun – out of forward-scattering
geometry – and rapidly faded as it moved away. Such an orbit would of necessity
be very different from that of a Kreutz comet. On the probability scale, this
object scores just 2 in my opinion.
Another object which Sekanina and Chodas referred to in their 2007 paper
as being a possible Kreutz candidate was the “Schmidtt phenomenon” of July
16, 1949. This took the form of a 70 degree-long, 1 degree-wide beam of light
stretching across the northern sky from the north-west to the north-east in the
early evening. It was reported by A. Schmidtt of the Algiers Observatory and
seems to have been first noticed by his wife through the trees from the
observatory’s car park. Schmidt described it as reddish near the horizon and to
have the same width throughout its length. The Schmidtts watched the
phenomenon for an hour before it sank beneath the horizon. No other sightings
of anything unusual in the sky on this or surrounding nights are known.
Schmidtt suggested that the beam may have been a comet tail, or a twilight
phenomenon (an unusual instance of crepuscular rays?) or a rather unusual
auroral streamer. It would have been an atypical example of any of these
alternatives, making it difficult to assess which is the more likely. We may
question whether it is realistic to think that a bright 70-degree comet tail could
appear in the evening sky, visible from the northern hemisphere, and yet be seen
by just two people at one location for an hour on a single night. A Kreutz
association (even if it was a comet) might appear to be ruled out by the date of
the observation. July sees the worst placement for Kreutz comets. They remain
at very small elongation, beyond the Sun (except for a very brief time near
perihelion, part of which is spent actually transiting the Sun) with their tails
pointing away from Earth. Yet, a very preliminary analysis by Sekanina and
Chodas suggest that the phenomenon might have been the remnant tail of a
sungrazer that experienced an 1887-type cataclysmic disruption just prior to
perihelion, a few days earlier than the Schmidtts’ observation. As it stands
though, a score of just 1 seems appropriate for this phenomenon.
An earlier candidate fares much better however. Back in 1882 (again!) a
total eclipse of the Sun crossed Egypt on May 17. When totality arrived, several
astronomers who had travelled to observe the phenomenon were surprised by
what they at first mistook for an unusual bright streamer at the outer edge of the
corona. The feature was also photographed and upon more careful examination
proved to be a small comet with its head almost enveloped by the corona. With
the first emergence of the Sun from behind the Moon, the comet disappeared and
nothing more was seen of it. However, following the discovery of the Great
Comet the following September and subsequent developments in our knowledge
of the sungrazing comet family, there was a suspicion that the eclipse comet of
May 17 might have been another – albeit far smaller – member of the same
group. This suspicion was raised to near certainty in the late 1960s when B. G.
Marsden demonstrated that this comet lay very close to where an object pursuing
the orbit of the Great Comet of 1843 is predicted to have been about 5 hours
prior to perihelion. The fit with an object in the orbit of the Great Comet of 1882
is not as good. This object, subsequently known simply as “Tewfik” in honour of
the Khedive of Egypt, was probably similar to the brightest of the sungrazers
discovered by space-based coronagraphs in recent years, for example, the SOHO
object of October 1, 2011. There is little doubt that it represented numerous
minisungrazers arriving at perihelion during the decade of the 1880s. This object
deserves a rating of 9.
A further list of suspected objects was compiled in 2002 by R. Strom. This
is a list with a difference, as the objects concerned are not clearly comets.
Instead, they are bright “stars” which Chinese solar observers noted in the
daytime close to the Sun. Strom argues that they were comets, as no other
phenomenon appears to fit the description. Moreover, because (as he states) 60%
of those comets which became visible with the naked eye in broad daylight
between 1800 and 2000 were Kreutz sungrazers, he concludes that most or all of
these “sun stars” or “sun comets” were most probably members of this group.
(Actually, he is incorrect in saying that 60% of naked-eye daylight comets during
the 1800 – 2000 period were Kreutz. Presumably, he was only thinking of the
non-Kreutz comets widely seen very close to the Sun sans optical aid – viz.
those of January 1910 and December 1927 and comparing their number with the
daylight sungrazers of 1843, 1882 and 1965. Yet, the non-Kreutz comets of
1861, 1976, 1947, 1948 and even Halley’s at its 1910 return – not to mention the
probable comets of December 21, 1882 and 1921 – were also seen naked-eye in
daylight, albeit not as widely as the ones of 1910 and 1927, or the 3 Kreutz
comets).
Personally, I was initially sceptical of the nature of the “sun comets”. My
reasons were twofold; none of the reported objects corresponded to a suspected
Kreutz comet appearing later (or earlier for that matter) in the night sky. This
worked both ways of course. Most of the objects catalogued by Strom appeared
at times of the year when a Kreutz comet would not have been well placed at
night. But from the opposite viewpoint, there appeared no obvious reason why a
sungrazer well placed at night should be missed near the Sun in daylight. After
all, those of 1843, 1882 and 1965 were well observed both at night and in
daylight. Daylight sightings of the 1843 comet appear to be recorded in Chinese
chronicles (although there is apparently some ambiguity as to whether they are
truly daylight or evening twilight observations), but the object is described as a
“broom star” (comet with a tail); quite a distinct description from the objects
which Strom notes.
Another problem was the way a number of the records were worded. They
appeared to be saying that the “star” was located “in” the Sun, which was
impossible if a comet really was being observed. I came to wonder if the
Chinese were seeing naked-eye solar flares, especially as most of the reports (as
we will shortly see) cluster in the seventeenth century. Did the Sun do crazy after
the Maunder Minimum? Nevertheless, an inquiry to comet observer Man-To Hui
as to the exact meaning of the Chinese word translated as “in” changed my
mind. The word does not imply that the object was actually within the disc of the
Sun. It is more accurate to say “beside” or maybe “in the vicinity of” than simply
“in”. The objects were not seen on the face of the Sun as a flare would be, but
beside the Sun in the sky in the manner of a daylight comet.
The problem with the non-correspondence between these daylight sightings
and the night-time comets remains, but Strom’s suggestion that seasonal sky
clarity in China may help somewhat. He suggests that rain-washed skies during
the summer monsoon season (when most of the sightings were made) may have
made it easier to see objects in close proximity to the Sun at that time. This
makes those Kreutz comets that come to perihelion during the summer months
easier to see in daylight than those during the winter, when they are better placed
in the night sky but when wind-borne dust may make the daytime sky too bright
in the near vicinity of the Sun to easily allow daylight comet sightings. It is also
possible that yet another variety of Kreutz comet behaviour is apparent here. We
have already seen that not all of these comets behave alike, and it may be
suggested that the “Sun comets” were basically small fragments which flared
tremendously (to extinction?) when close to the Sun but were otherwise too faint
to be seen. As we can see from the dates of these objects listed below, some of
them (if major Kreutz comets) may well have been seen from somewhere in the
twilight following perihelion, yet no such observations are known.
Be that as it may, the objects listed by Strom were observed in the years 15
AD (sometime between March 10 and April 11 – the Han shu also mentioning a
“star” seen at noon that year; probably the same object), 1539 (between July 15
and August 13), 1564 (between August 7 and September 4), 1625 (September 2),
1630 (August 5), 1643 (between June 16 and July 15), 1644 (between March 9
and April 6), 1647 (July 28), 1648 (during the “summer”), 1665 (August 27),
1774, 1792 (between April 21 and June 20), 1839 (August 14) and 1865 (July
18). The one recorded in the summer of 1648 may have been the same daylight
“star” that was seen in London on May 29 at the time King Charles I visited St.
Paul’s Church to give thanks for the safe delivery of Prince Charles.
Little more can be said about the earliest of these objects, but the
concentration of reports during the 1600s, especially during the 1640s, is little
short of outstanding. If these were Kreutz objects, and if the other suspected
sungrazers observed during that century were also bona fide members of the
group, it seems that the seventeenth century was the century of sungrazers. The
appearance of the non-Kreutz sungrazer of 1680 is interesting, but probably has
no significance other than being an odd coincidence. No greater significance, we
might suppose, than the deep sunskirting, but apparently dynamically new,
Comet ISON coming to perihelion a little less than 2 years after Comet Lovejoy!

THE HIGH-SCORING SUSPECTS
We now come to look a little more closely at those objects rated by England
as scoring 5 or better on his probability scale, in other words, those comets that
have a 50% or greater probability of being true Kreutz sungrazers. On any
estimate of the size of the sungrazing family and its evolution, these are
obviously the ones most worthy of careful consideration.

THE GREAT COMET OF (circa) 372 BC
This Greeks noted a very large comet sometime during the winter at a time
when Alcisthenes was the archon at Athens. The date appears to have been
around 372 BC although we cannot be certain of the precise year. Most of the
information that has come down to us is second or third hand, although the most
extensive account given in Aristotle’s Meteorologica is probably from his own
memory of having seen the comet as a child of about 12 years of age. It certainly
left a vivid impression in his mind!
According to Aristotle, the comet first appeared immediately after sunset at
a time when the weather was very clear and frosty and set later each evening as
it trekked eastward. It sported a long tail that stretched across the sky “like a
great ribbon” and gradually rose higher in the sky, eventually becoming lost in
the region of the belt of Orion. Aristotle mentions 4 comets in his book, but this
is the only one he refers to as “great”. Moreover, this description is repeated
several times during his account of it. Various other reports refer to it as having
the appearance of a “beam”, at least initially, although becoming more like a
comet as time went by. This reads rather like a sungrazer – a very prominent tail
and small head that later became larger and more diffuse.
An interesting record preserved by Diodorus, who was not a contemporary
of the event, tells of a “torch” that appeared sometime around 370 BC so bright
as to cast shadows like those cast by the Moon! This is sometimes seen as a
reference to the comet, with the differences in dates explained as having arisen
from a mistake in converting Greek dates to Roman ones. If this is correct, it
implies that the comet must have been exceedingly bright, although the casting
of shadows as clear as those of the Moon sounds exaggerated and if this really is
an account of the comet, we might suspect that it has grown by word of mouth
rather like the notorious fishermen’s tales of the “big one that got away”!
However, scholars agree that most or all of the “torches” that appear from time
to time in Roman accounts of astronomical phenomena are to be distinguished
from comets. The term seems to refer to bright meteors or fireballs and, as these
not infrequently are bright enough to cast shadows like the Moon, it may be that
Diodorus’ object was a fireball and not the Great Comet. It is not impossible that
an account of a meteor and one of the comet became mixed together. When
Halley’s Comet was about to return in 1986, the news media interviewed several
elderly folk who had witnessed its previous apparition in 1910 and some of the
memories recalled had clearly become distorted over the decades. One witness in
particular began by describing a very realistic account of the comet, but then
went on to recount how it burst into a shower of fragments and disappeared!
Although this gentleman showed no evidence of mental confusion, it seemed
clear that he had combined childhood memories of the comet and a bright
meteor into a single object. It is not impossible that an occasional early record –
especially by the Romans who, unlike the Chinese, were not especially
concerned with astronomical events – has been similarly distorted.
Whether or not the comet cast shadows, it was certainly long remembered.
An account by the historian Ephorus, allegedly mentions that it split in two.
Unfortunately, the original account by Ephorus has long been lost and the only
information about this that has come down to us is the critical account of Seneca
who, of course, lived long after the comet appeared. The supposed splitting of
the comet was one of the reasons that astronomers first came to wonder whether
this object was a Kreutz comet. Once it became accepted that the sungrazers of
the nineteenth century were fragments of a single comet and not different returns
of the same object, the Ephorus account started to sound very interesting. Could
he have actually seen the splitting away of the comets of 1843 and 1882?
The answer is a clear negative. The evolution of the Kreutz group is a lot
more complex than this, but it is clear from more recent studies of the breakup of
comets that the velocities of separation involved in these events are too small for
the process to be observed with the naked eye. Whatever Ephorus saw, it could
not have been the splitting of the comet’s nucleus in the manner of the events of
1882 and 1965 – or of any other similar event recorded over the years. I suggest
that what he may have witnessed was a disconnection event in an ion tail, but
without his actual description, any suggestion is really only guesswork. Perhaps
Seneca deliberately re-worded his statement for the purpose of ridicule, as he
clearly wished to perform a demolition job on Ephorus’ reliability. Maybe the
“split” (or whatever descriptive word or phrase he really used) was nothing more
than a division of the tail by the type of dark central lane that is not at all
uncommon in bright comets.
Nevertheless, even discounting (as we surely must) the alleged “split” as
evidence favouring a Kreutz connection, there are other reasons for thinking that
this comet might have been an early sungrazer. For instance A. Pingre, writing
before the Kreutz group was discovered, attempted to compute a very
approximate orbit for this comet and suggested that it had a “very small”
perihelion distance in the direction of Virgo or Libra and an orbital inclination of
about 150 degrees. The “very small” perihelion distance is interesting (but how
small is “very small”?) and the orbital inclination is in the right ballpark, but
Sekanina and Chodas draw attention to the fact that the direction of perihelion is
well off course. Just how much stress we should place on any of these values is,
however, open to debate considering the vagueness of the records with which
Pingre had to work.
Since the 2007 paper by Sekanina and Chodas, about which more will be
said in the course of the following chapter, the Kreutz credentials of the Great
Comet of 372 BC has taken something of a nosedive. Especially since Marsden’s
1989 paper on the Kreutz group, in which he suggested that this might actually
be the parent of the entire Kreutz family, this comet had gained quite a degree of
prominence in both technical and popular literature on the subject. The comet’s
obviously spectacular appearance seemed fitting for the Grand Progenitor of the
group (let’s refer to this progenitor comet as “Comet Kreutz” for the sake of
convenience) and Aristotle’s description, plus the region of sky through which it
moved, all appeared more or less consistent with the Kreutz orbit.
The recent and continuing work of Sekanina and Chodas, however, is
moving toward a more comprehensive model of the group and in the process
some old ideas are being shattered and former suggested identifications seriously
questioned. Computing back from recent to ancient perihelion passages, these
authors find that the date for the final intact appearance of Comet Kreutz is
incompatible with this having taken place in 372 BC. The proposed
identification with this comet and the sungrazer parent is, it seems, a casualty of
our increasing knowledge of this comet family.
Whether the 372 BC comet is a fragment that broke away from Comet
Kreutz at an earlier time is something that has not been examined, as far as I am
aware. It is not likely though if Pingre’s direction of perihelion is correct,
although the thought remains an intriguing possibility.
As a completely irrelevant aside, it is interesting to speculate about the
effect that seeing this great comet at such a young and impressionable age might
have had on Aristotle. His father was a physician and it may, perhaps, have been
expected that he would follow in this profession as well. Had that happened,
Aristotle would likely have been lost to history. But did seeing this grand sight in
the heavens turn his young mind toward cosmological and philosophical
speculations; speculations which led him down the path he followed in later
years?
We can only speculate of course, but if that was really the case, the course
of a good deal of intellectual history was given direction by this comet. The
history of logic would obviously have been different without Aristotle, but a
good deal else would have changed as well. The philosophy of Thomas Aquinas
and the Schools is largely Aristotle’s philosophy Christianised. I recall that,
during my years as a philosophy undergraduate, one of my lecturers pointed out
that the two most influential philosophical systems of the day were Marxism and
Thomism. Thomism, because Roman Catholics are, in theory at least, Thomists.
And, as we said, Thomism is Aristotle Christianised. The doctrine of
Transubstantiation, for instance, was presented by St. Thomas in terms of
Aristotle’s philosophy of substance and accident. Just consider how much of
European history since the Reformation hinged on whether one accepted or
rejected various aspects of Thomist teaching and it becomes apparent just what a
large role Aristotle has played, and continues to play, in fields not obviously
related to the philosophy seminar room!

THE ECLIPSE COMET OF 116 (?) BC
Seneca notes, in his Naturales Quaestiones, that the Greek philosopher
Posidonius reported seeing a comet close to the Sun during a solar eclipse. Born
sometime around 135 BC, Posidonius travelled widely in his early years before
settling on the island of Rhodes and writing his numerous books on general
science and geography.
Unfortunately, the year of the eclipse and comet is not known, although
there were several solar eclipses, mostly partial, during the philosopher’s
lifetime that were visible from locations which he may have visited. The sky
would very likely have been too light during a partial eclipse, however an
annular eclipse occurred on June 29 in the year 94 BC, visible across the western
Mediterranean, and some astronomical historians think that this may have been
the one in question. Nevertheless the sky would still have been pretty bright
during this enent and any comet that became conspicuous during the deepest
phase might also have remained faintly discernible afterwards in full daylight.
Had this happened, it surely would have been noted. A better candidate is the
total eclipse that crossed Spain and North Africa on August 29, 116 BC.
Posidonius may indeed have witnessed this event as a young man sailing around
the western Mediterranean.
A comet approaching and receding from the Sun along the line of sight from
Earth would only have a chance of being visible either in daylight or during an
eclipse. A sungrazer coming to perihelion at the time of the eclipse of 94 BC
would have been in that situation. One arriving during the 116 BC eclipse may
have become visible in the morning sky after perihelion, however, if it was
visible only during totality, it was probably not an intrinsically bright object and
may have faded out shortly afterwards. Perhaps this is the first record of a
minisungrazer of the Tewfik/SOHO variety!? Pure speculation of course, as we
cannot even be sure if it was a sungrazer at all. If we knew the comet’s location
in relation to the Sun we could make a better assessment, but unfortunately even
that information has been lost.

COMET 133 AD
On February 8 of the year 133, Chinese astronomers noted the appearance
of a comet near the star Delta Eridani. It was described as having a white (i.e.
intense and bright) tail some 75 degrees long and 3 degrees wide. Unfortunately,
nothing further is known about this object, but its general appearance and
position in the sky are consistent with its having been a member of the Kreutz
family. This is one of the objects that Hasegawa suspects of Kreutz membership
and finds that its recorded position fits a time of perihelion of February 5.

COMET 423
The Chinese first noted this comet near Gamma Pegasi and Alpha
Andromedae on February 15. Its tail was described as being “white” and over 30
degrees long. A separate record placed the comet in the Andromeda/northern
Pisces region and described the tail as 45 degrees in length. The comet moved
toward the southeast, disappearing in central Eridanus on March 5.
Hasegawa included this as a possible sungrazer, with perihelion on February
6, although he also suggested that the position of the comet was consistent with
it having been an earlier apparition of C/1931 P1 (Ryves). This object was a
sunskirter, having a perihelion distance of around 0.07 AU, but has an orbit
otherwise very different from that of the Kreutz group. Although the basic orbit
of this comet is secure, the values of its eccentricity and orbital period vary far
and wide, from slightly hyperbolic to a relatively short-period ellipse of just over
200 years. Hasegawa’s proposed identification with the 423 comet rested upon
the latter orbit. More recent reassessments of this comet’s orbit suggest,
however, a far more elongated ellipse with a period of several thousands of
years. It seems that its previous return took place way back in prehistory, long
before the year 423.
The comet of 423 seems a bit too far north for a sungrazer, taking the
records at face value. Furthermore, in 1997 H. Zhou and his colleagues
computed an orbit from the records and found a radically non-sungrazing orbit
having a perihelion distance of 1.1 AU on February 3. The description of the
object’s physical appearance does not, however, read like the average comet with
a perihelion as large as 1.1 AU. Furthermore, the Zhou orbit should have made it
visible during the previous autumn, unless it experienced a great surge in
brightness following perihelion.

THE GREAT COMET OF 467
A strange phenomenon was noted by the Chinese on February 6 of that year.
Described as a “white vapour” it possessed a bright nucleus and a pair of tails
that stretched halfway across the sky from the southwest to the southeast.
Because the astronomical chroniclers did not use the terms normally designating
comets, Ho Peng Yoke, in his catalogue of ancient and mediaeval eastern
comets, suggested that this phenomenon may have been an aurora. Nevertheless,
records from other parts of the world leave us in no doubt that this was a comet;
and an especially magnificent one by all accounts.
Byzantine chronicles, for instance, tell of a large prodigy visible in the skies
for 40 days. According to one description, the comet was shaped “like a
trumpet”, presumably implying that it was very long, began quite narrow and
gradually fanned outward along its length. Other Byzantine witnesses described
it as a “spear” and yet others as a “beam”. Observers in Portugal and Italy also
noted the comet.
This comet’s appearance and position in the sky makes it a good candidate
for Kreutz family membership. A perihelion passage very early in February is
indicated.

COMET 852
Japanese astronomers recorded a bright comet on March 14 and their
Chinese counterparts sometime between March 25 and April 22, most likely late
in March. Both locate the comet in the western evening sky, within the
constellation of Orion and possibly rather close to the star Betelgeuse. It was
described as having a distinct tail, which the Japanese estimated as 75 degrees
long. The position and appearance of this object suggest possible Kreutz
membership and Hasegawa suggests that, if it was a sungrazer, perihelion would
have taken place around March 7.

THE GREAT DAYLIGHT COMET OF 1106 –X/1106 C1
 We now come to one of the chief Kreutz suspects of ancient and mediaeval
times and one which just about everyone who has speculated on possible
sungrazing comets in earlier years has duly noted. It also seems to have been the
brightest comet of the middle ages; indeed, one of the most luminous ever
recorded.
The comet was first noticed on February 4 as a “blazing starre” visible in
broad daylight only some 2 degrees from the Sun itself. Some records even refer
to it as a comet, suggesting that some of the tail was also visible against the
daytime sky. It was visible for 6 hours, according to Sigebertus, after which it
remained hidden in the Sun’s glare until February 7, when it was sighted from
Constantinople and Jerusalem following sunset.
The Japanese saw the comet on February 9, describing the tail as “white”
and estimating its length as 100 degrees, pointing eastward between the stars
Zeta and Pi Ceti. Korean astronomers also noted the comet that same evening,
although they only recorded 15 degrees of tail.
Chinese observations of the comet are unusually brief, although
astronomers there saw it on February 10, noting a tail 90 degrees long and
(fanning to?) 5 degrees wide. It was pointing toward the northeast and was said
to be broken into fragments. Did this refer to striations in the tail or to something
else? Curiously, some European records of the comet mention the presence of
“other strange stars” seen around the same time as the comet. This reference may
have simply been to an unrelated meteor shower, but it is also possible that these
“strange stars” and the “fragments” mentioned by the Chinese referred to
brighter counterparts of the odd comet-like clouds and nebulous wisps observed
near the head of the Great Comet of 1882. What it surely does not refer to is the
splitting of the nucleus a la 1882 and 1965!
Another curious reference was made by Armenian observers on February 13
who described the tail as “like a flowing river”. Again, this may have been a
simple description of the great length of the tail but “flowing” seems to imply
some sort of perceptible motion within it. Could this refer to the “coruscations”
noted in some comet tails by later observers? This effect, as we have already
mentioned, is difficult to explain as a real process in the tail and is most probably
a trick of our atmosphere or (less likely) a form of optical illusion. Nevertheless,
it can appear very real to an eyewitness!
The comet remained widely observable until the middle of March, and a
small number of observers tracked it for considerably longer. William of Tyre,
for example, mentioned that it was last observed on March 25 and one Armenian
record implies that it was visible until as late as April 4.

THE GREAT COMET OF 1668 – C/1668 E1
This comet was first spied from the Cape of Good Hope on March 3 as a
first-magnitude object sporting a bright tail. Observers at Lisbon next saw it on
March 5, when its tail was measured as 45 degrees in length. P. Valentin Estancel
of San Salvodor (Brazil) observed it from March 7 to 11, by which time it had
moved to the Lepus-Eridanus border. Chinese observers also found it on March
7, describing it as “a stretch of white light to the southwest”. They followed the
comet until March 30 and noted that the tail had increased to over 60 degrees as
measured on the 18th. of that month.
The comet was also observed in Japan, Italy and India where Gottignies
(whose name is sometimes unofficially associated with this object) made
accurate positional measurements of the head between March 5 and 21. As an
indication of the level of absurdity that superstition about comets can reach, in
one region of Europe the comet was blamed for an unusually high mortality rate
noted amongst cats at that time!
Many alternative orbits have been computed for this comet. Most
catalogues of comet orbits give a sunskirting orbit with a perihelion distance of
approximately 0.07 AU on February 28, but other orbits yield perihelion
distances ranging from 0.6 AU to sungrazing values and dates from February 23
to March 28. Marsden found that the best formal fit to available positions
yielded a perihelion distance of 0.3 AU, yet from the comet’s appearance and
behaviour, he considered a sungrazing orbit more probable. Perihelion would
then have occurred on March 1. In fact, this comet is almost universally assumed
to have been a member of the Kreutz group, even though the non-Kreutz
sunskirting orbit is (confusingly) the one most frequently given.

THE GREAT SOUTHERN COMET OF 1689 – C/1689 X1
Once again, the first people to report this object were located at the Cape of
Good Hope in South Africa, and they found it on the mornings of November 24
and 25. Following these early sightings, the comet temporarily passed from sight
until early December when Father Richaud in India (whose name is sometimes
unofficially given to the comet) saw the tail rising up from the horizon in the
constellation of Centaurus. Both Richaud and the Cape observers saw it again on
December 9 and measured the tail as over 4 degrees long. By December 14, a
curving tail 60 degrees long curled out from a head of magnitude 3 or 4. The
comet was also observed by the Japanese and remained visible until early
January 1690.
An orbit computed by Holetchek in 1891 yielded a sunskirting perihelion of
0.064 AU on November 30, but the comet has long been suspected of Kreutz
family membership and a sungrazing orbit having perihelion on December 2 at a
distance of 0.009 AU was calculated by Marsden. Note that, if Marsden’s orbit is
close to the truth, this is one of the relatively few Kreutz comets and Kreutz
suspects that was discovered prior to perihelion, excepting of course, the
minisungrazers of recent decades.

THE GREAT SOUTHERN COMET OF 1695 – C/1695 U1
The tail of this bright comet was first spied by Father P. Jacob, Brazil, as it
rose from behind the dawn-lit eastern horizon on October 28. An independent
discovery was made on the 30th. by Bouvet in India, when the comet’s head was
located in Libra and the tail stretched some 30 degrees into Virgo. The comet
was picked up by the Chinese on November 2 and from the Arabian Sea on
November 17. A maximum tail length of 40 degrees was recorded on November
6 and the final view of the comet came on the 19th. after which the full Moon
obscured the fading object.
Uncertainties in the measured positions have yielded orbits having
perihelion distances from 0.04 to 0.84 and perihelion dates from October 24 to
November 19. The object’s appearance, however, strongly suggests a sungrazing
perihelion distance and the position and motion of the comet is consistent with
Kreutz membership. A Kreutz orbit yields perihelion on October 23 at a distance
of 0.009 AU.

COMET OF 1702 – X/1702 D1
The tail of a bright comet was first espied in the evening twilight by Dutch
navigators at the Cape of Good Hope on February 20. Further sightings came
from others in South Africa and other southern sites, including from ships at sea
on the following evenings. The tail was also observed from Bengal for half an
hour after sunset on the 23rd. of the month. Observers near the Bay of Bengal
estimated the tail’s length as 45 degrees on February 28 and about the same time
the comet was seen from Rome by J. Maraldi and by LaSueue in Louisiana on
the evenings of February 27, 28 and March 1. It was last observed on March 2.
Unfortunately, so few positional measurements were obtained that not even
an approximate orbit for this object has been calculated, however its appearance
and position strongly suggest Kreutz membership. The couple of reasonable
positions that do exist lie very close to the locations of an object moving in the
orbit of the 1882 comet assuming a sungrazing perihelion on February 15. The
fit to the orbit of the 1843 comet is somewhat less satisfactory.

This completes the list of the most likely candidates for Kreutz membership.
Leaving aside the very early possibilities for the moment (anything prior to
1500) and the objects discovered by space based coronagraphs in recent decades,
it is interesting to make a combined list of the above comets, the “sun comets”
listed by Strom and the definitive bright Kreutz objects that have appeared since
1843. The result is as follows;
1539, 1564, 1625, 1630, 1643, 1644, 1647, 1648, 1665, 1666, 1668, 1689,
1695, 1702, 1774, 1792, 1839, 1843, 1865, 1880, 1882 (May), 1882
(September), 1887, 1945, 1963, 1965, 1970 and 2011.
That is quite a list, but the most striking aspect of it is the concentration of
entries in the 1600s, especially during the 1640s. Although some of the objects
listed here may not have been Kreutz comets, most probably were and, of
course, the more recent entries definitely were.
The second notable aspect of this list is the way the entries tend to clump
together. Radom lists display clumps of course, but there are good reasons
(which we will discuss in the following chapter) for believing that something
more than randomness is at work here. From the above list, we may discern the
following clusters;
1625 – 1630, 1643 – 1648, 1665 – 1668, 1689 – 1702, 1839 – 1843, 1880 –
1887, 1963 – 1970.
Of course, it is somewhat subjective just where to draw the line, and others
may (and do) draw it somewhat more liberally to include the objects that I have
omitted from the clusters. It is also likely that objects passed unseen and this
raises the possibility that the ones left out of the above list of clusters may have
had unseen objects clustered around them or even that what I have listed as two
adjacent clusters may really have been parts of a single extended cluster; the
intermediate objects having passed unseen and unrecorded. A continuous cluster
may have extended from 1839 right through to 1887 for instance, or from 1945
to 1970, especially if the 1949 phenomenon turns out to have been a bona fide
sungrazing comet.
Be that as it may, it is becoming clear that the Kreutz sungrazing group is a
good deal more complex than initially believed. Some of these complexities, and
the attempts at solving them, form the subject of the final chapter of this short
book.




































V
THE EVOLVING STORY OF KREUTZ GROUP
EVOLUTION

Although the basic model derived by Kreutz himself has remained the
bedrock of hypotheses as to how the Kreutz family of sungrazing comets has
evolved, our increasing knowledge of the family has brought with it a level of
complexity that this astronomer could scarcely have imagined. The model of a
comet breaking up into a family of comets, as presumably witnessed by
observers of the Great Comet of 1882, has stood the test of time, but the version
of this model which we have in 2012 differs radically in a number of respects
from its original form as proposed by Kreutz. In part, this has come about by the
discovery of other members of the family, but also by the recognition that
fragmentation of comet nuclei is not confined to the parts of their orbit close to
perihelion or even to times when the comet is active. But more about this anon.
For the present, let’s take a quick look at how the basic fragmenting-comet
model has developed since the time of Kreutz’s initial hypothesis.
It is probably fair to say that with the apparent absence of further
sungrazing comets during the years and decades following 1887, the Kreutz
comet group increasingly became looked upon as something belonging to the
past. Moreover, the years following the return of Comet Halley in 1910 were
ones in which cometary matters drew less and less attention from professional
astronomers in general and in that atmosphere it is hardly surprising that the
Kreutz group was less than a hot astronomical topic. Even the appearance of a
small sungrazer in 1945 did little to enliven interest in this subject.
From about 1950 however, something of a revival of interest in comets
began to take place. Publication of the opposing theories of the nature of comets
– R. Lyttleton’s “sandbank” model and F. Whipple’s “dirty snowball” – plus the
hypothesis put forward by J. Oort of a vast sphere of comets surrounding the
planetary system (independently proposed some time earlier by E. Opik, but
largely ignored by the astronomical community of the time) brought some life
back into discussions of what we might call the cosmology of comets. The
steady stream of new comet discoveries (especially by a dedicated team of visual
searchers associated with the Skalnate Pleso Observatory in Czechoslovakia)
kept the observational side of the subject alive for a small group of astronomers,
even though nothing very noteworthy by way of bright comets had appeared in
the northern hemisphere since 1910. The brightest comets of these decades had
favoured more southerly latitudes and had also, for the most part, appeared quite
suddenly when close to their maximum brightness, giving little time for
professional astronomers to plan observing programmes. That dramatically
changed in 1957 with the appearance of a pair of very bright and spectacular
comets well visible from northerly climes. The first of these – C/Arend-Roland –
was discovered some 6 months before perihelion, allowing astronomers precious
time to plan its scrutiny. This comet became the most thoroughly observed since
Halley in 1910 and, considering the advances in observing technology that had
taken place between 1910 and 1957, was undoubtedly the most intricately
scrutinised ever at that time. Observations of this comet were supplemented by
those of C/Mrkos later that same year, although this second comet was already at
its peak when discovered and did not allow the preparation time afforded the
earlier object. These two comets broke the long comet drought, with a parade of
bright objects during the 1960s and beyond.
It was during this period of reawakened interest in matters cometary that a
pair of sungrazers made their appearance; Pereyra in 1963 and Ikeya-Seki in
1965. While the faint and transitory object of 1945 might have been ignored,
these new objects – especially the latter one – could not be. Fortunately, Ikeya-
Seki was discovered about 5 weeks prior to perihelion which, although not
giving a long time to plan, was enough for astronomers (realizing the importance
of this object as a potentially very bright sungrazer) to prepare an extensive
observing programme. This object knocked Arend-Roland off its perch as the
most thoroughly and scientifically scrutinized comet of all times and much was
learned about the nature of comets in general as well as how they behaved when
very close to the Sun.
One lesson taught by the sungrazers of 1963 and 1965 was that the Kreutz
group was not something belonging to the past. Members of the family were still
around and more could potentially be learned about them.
The challenge to find out more about these comets was taken up by
astronomer B. G. Marsden, a specialist in the study of the orbital evolution of
comets and other small Solar System bodies. Marsden’s initial paper on the
sungrazing group was published in 1967 and included some important
developments of the original fragmented-comet model. In particular, he noted a
pair of characteristics of the list of sungrazing comets (in which, by the way, he
included not only the undisputed family members, but also the more
controversial ones of 1668, 1689, 1695, 1702 and Tewfik). First, he noted that
the distribution of these objects was far from even. There was a clear cluster in
the late 1600s/early 1700s, the 1880s (of which he included 1843 as a possible
early member) and in the middle years of the twentieth century (1945 – 1965).
Secondly, he noted that the orbits of the various comets concentrated around
two sets of values of certain orbital elements, in particular the argument of
perihelion, longitude of node and perihelion distance. Two quite distinct
subgroups emerged, one having larger values for the argument of perihelion,
small values for the longitude of the ascending node and very small perihelion
distances and the other combining smaller values for the argument of perihelion,
large values for the longitude of the node and larger perihelion distances (albeit,
still significantly within 0.01 AU of the Sun’s centre). The first set he referred to
as Subgroup I and the second as Subgroup II. The archetypical Subgroup I is the
comet of 1843; that of the second subgroup, 1882.
These features had not gone entirely unnoticed previously, but it was
Marsden who first drew attention to their possible implications. Unfortunately,
the two were not entirely in harmony. Thus, when the orbits of the different
members of the clusters were also assigned to one or other of the subgroups, the
results were not as might have been hoped for an easy analysis. Considering the
first subgroup (and assuming that the Kreutz orbits computed for these objects
were correct), Marsden found that the measured positions of the comets of 1668
and 1689 were better represented by Subgroup I orbits while that of 1695 was
better represented by Subgroup II. Furthermore, although no orbit has been
computed for the 1702 comet, the best available positions fall closer to a
Subgroup II orbit than to a Subgroup I.
Moving to the better established clusters of the 1800s and 1900s fared no
better. The comets of 1843, 1880 and 1887 all belonged to Subgroup I. Also, the
positions determined for the eclipse comet Tewfik were found to be a better
match to Subgroup I. On the other hand, the Great Comet that appeared in
September of that year was clearly Subgroup II. As for the twentieth century
cluster, the comets of 1945 and 1965 were Subgroup II while that of 1963 was
clearly a member of Subgroup I.
What was happening here? On the face of it, the facts appeared to place an
investigator of the Kreutz group squarely on the horns of a dilemma. EITHER,
abandon the clusters as being nothing more than coincidence OR abandon the
subgroups for the same reason. One could not accept both as having real,
physical, significance.
 Marsden abandoned the clusters. Assuming that nucleus splitting is
confined to perihelion or very close thereto, he employed the existence of the
pair of subgroups to develop a scenario in which a progenitor comet split apart
several revolutions ago; possibly as many as 20. The period of revolution of this
progenitor comet was about 700 – 800 years, the same as its more recent
fragments, and during the course of the several returns following its split, the
action of planetary perturbations caused the 2 fragments to assume somewhat
different orbits. Finally, each fragment of the split nucleus broke up and the
second-order fragments of this later disintegration have been returning since the
1600s as sungrazing comets.
Taking the best available orbits of the second subgroup (viz. those of 1882
and 1965) and tracing them back in time, Marsden found that they converged
near their previous perihelion in the early 1100s. Moreover, the velocity of
separation required to set them on their separate journeys was effectively
identical (within the bounds of error) to the mutual velocities of separation
determined for the fragments of the split nuclei of the comets of 1882 and 1965
themselves. This was a very strong indication that this pair of comets began their
individual careers as fragments of a comet that split apart as it passed through a
sungrazing perihelion passage sometime during the early 1l00s. The comet of
1106 sprang immediately to mind as the possible parent.
The best-determined Subgroup I orbits are those of the 1843 and 1963
comets, but tracing these back through time did not yield such satisfactory
results. The published orbit for the first of these comets gives a significantly
shorter period than that of the latter. The 1880 comet (for which only a parabolic
orbit had been computed, but which presumably moved in an ellipse similar to
that of the 1843 one) was also difficult to fit into the picture. Marsden opined
that the parent of the first subgroup must have broken up at an earlier return,
probably the one that presumably took place early in the first millennium.
This initial paper was not, however, Marsden’s last word on the subject. A
second paper was published in 1989 in which some of the conclusions of his
earlier work were revised. Also, between 1967 and 1989, some further
developments had taken place in the unfolding story of the Kreutz group of
comets. The year 1970 saw the discovery of White-Ortiz-Bolelli and the 1980s
provided the first evidence that the group was a lot more extensive than had
hitherto been imagined as SOLWIND and Solar Max gleaned the first fruits of
what would in subsequent years become a vast harvest of small sungrazers
observed from coronagraphs located beyond Earth’s atmosphere.
 White-Ortiz-Bolelli proved to be something of a headache for Marsden. The
problem was, its orbit did not fit into either of the two subgroups. It was clearly
closer to Subgroup II than to Subgroup I, but not sufficiently close to be included
as a bona fide member. Despite the fact that it arrived just 5 years after Ikeya-
Seki, it did not appear to be closely associated with that comet.
Marsden argued, in this second paper, that the Subgroup I orbit represented
the “core” of the Kreutz system and that the period of the comet of 1843 was
closest to that of the original progenitor comet. This, he figured to be about 400
years. Tracing back, he found that a shorter period of this type could account for
both the 1843 and 1880 comets. They appeared to converge at a previous
perihelion during the latter fifteenth century, although no historical records
mention any likely candidate around that time. Nevertheless, tracing back over
the subsequent orbit of the Sun would bring this hypothetical 1843/1880 comet
back to around the time calculated as the previous return of Pereyra. The
scenario presented by Marsden is that a Subgroup I comet passed perihelion
sometime around 1100 and split apart, one nucleus maintaining the parent’s
orbital period of around 400 years and the other moving away with an increased
period of around twice that value., The first returned in the late 1400s, when it
also split. Once again, these second-generation fragments essentially held to the
parental period and returned in 1843 and 1880. The second of the first-
generation fragments did not return until 1963, when it was discovered as Comet
Pereyra.
Tracing further back into time, the parents of each subgroup can eventually
be brought together as a progenitor comet appearing in the fourth century BC
most probably, Marsden suggests, as the famous object of 372 BC.
This still leaves White-Ortiz-Bolelli unaccounted for however. On the
assumption that the only significant influences acting on these comets orbits are
the mild gravitational perturbations by the planets (“mild”, because no planet can
be closely approached by comets moving along Kreutz orbits), Marsden could
only account for this comet by assuming that it has spent a long time in an
independent orbit. On his model, it probably existed as a separate object since
splitting away from the original progenitor some 2,400 years ago. In terms of the
classification into subgroups, Marsden assigned this to a separate subgroup (or,
maybe, sub-subgroup) which he designated as Subgroup IIa. This classification
recognises its closer resemblance to comets such as 1882 and 1965 rather than to
those of 1843 and 1963, while still maintaining its clear separation from all
former objects. At the time of Marsden’s writing, the only other possible
member of this subgroup was C/1984 O2 (SOLWIND).

CASCADING FRAGMENTATION – BREAKING UP IS
NOT HARD TO DO!
Another astronomer who became interested in the evolution of the Kreutz
group is Z. Sekanina. He has long been interested in the various disintegration
phenomena displayed by comets; a subject which covers a wide range of
processes from the various phenomena associated with dust tails, through
nucleus splitting to the total dissolution of the solid nucleus. An expert on the
development of dust tails, Sekanina was the first to provide theoretical
predictions as to how dust tails should develop, based upon computations of
trajectories of particles of different sizes released simultaneously from a nucleus
as well as the theoretical paths of particles having similar sizes but released over
an extended period of time. This mechanical theory of dust tails goes back a long
time and was employed before Sekanina to explain various observed features in
comets, but he was the first to use it to predict tail development ahead of time by
successfully applying it to the development of Comet Kohoutek back in 1973.
Sekanina’s interest in the behaviour of the solid component of cometary
matter did not stop at tail particles but also involved investigation of the
trajectories of more massive bodies such as fragments broken away from
cometary nuclei. The sungrazing family represents the most extensive instance
of this, and naturally became a principal focus of his investigation.
Sekanina, unlike Marsden, did not confine nucleus disruptions to the near-
perihelion sections of a comet’s orbit. Tidal splits due to close approaches to the
Sun or to a major planet are only part of the story, as many comets have been
seen to split – apparently spontaneously – far from any large body that could be
affecting them tidally. Sekanina went even further and postulated the splitting of
comets even in parts of their orbits so far from the Sun that they were not in an
active state. Partially through the study of SOHO sungrazers and partially by
considering the continuing breakup of Comet Shoemaker-Levy 9 (the object that
initially broke up very close to Jupiter and subsequently collided with that planet
in the early 1990s) Sekanina proposed that the phenomenon of “cascading
fragmentation” is an important but largely overlooked process affecting comets.
Briefly stated, this process means that disruption of comet nuclei often happen,
not in a single splitting, but over extended periods of time and essentially at any
location within the comet’s orbit. An initial split may occur very close to the
Sun, but subsequent disruptions can continue to occur even at, and following,
aphelion. Sekanina and his colleague P. Chodas found that the small-scale
clustering of SOHO sungrazers could be best explained by fragmentation events
far from the Sun and saw something similar at work in Shoemaker-Levy 9 and,
later, in the short-period comet Schwassmann-Wachmann 3. This object
unexpectedly surged mightily in brightness during its 1995 return, brightening
suddenly from a telescopic object of magnitude 12 to a faint naked-eye comet
sporting a beautiful dust tail as observed through binoculars. During the latter
phase of the 1995 apparition, observations of the nuclear region found that
several fragments had broken away from the main mass, presumably triggering
the brightness outburst. The following return was a very unfavourable one, but
several observations of the comet were nevertheless made and several fragments
were visible as small comets travelling close to the main one. The subsequent
return in 2006, however, was an exceptionally favourable one during which the
comet passed close to Earth. Over 100 small comets were found, accompanying
the main one, at this apparition. Moreover, the major secondary fragment was
clearly in the process of further fragmentation, shedding tiny comet-like
condensations down its tail as its nucleus flared and progressively broke apart.
Most of the smaller secondary comets were the products of fragmentation events
occurring far from the Sun between successive returns of this comet.
Further evidence of fragmentation far from the Sun is supplied, so far at
least as the Kreutz group is concerned, by the very small dimensions of the
SOHO discoveries. These fragments must have been spawned since the previous
perihelion passage when at relatively large distances from the Sun for them even
to have survived!
Referring specifically to the Kreutz group, Sekanina and Chodas found that
fragmentation events far from the Sun gave very different results to those
occurring at or very near perihelion. The typical perihelic split resulted in comets
following in what is essentially the same orbit but with periods differing by
about 80 – 100 years. This means that at the return to perihelion following the
one during which the split occurred, 2 or more comets travelling in orbits having
almost identical orbital elements will appear around 80 – 100 years apart. The
Kreutz pair of 1882 and 1965 provides a classic example of this.
On the other hand, a fragmentation event taking place near the aphelion
point of the orbit has the opposite result. In this instance, the orbital elements can
change dramatically, but the periods of the different fragments remain fairly
constant. In the most extreme instances, when splitting takes place essentially at
or after aphelion, it is possible for one fragment to assume a Subgroup I orbit
and the other a Subgroup II, yet to arrive at their next perihelia just a couple of
months apart! The velocity attained by the fragments appears to be governed by
the rotation of the original body and to be independent of the comet’s location in
its orbit at the time of the split. From their study of the sungrazers, Sekanina and
Chodas found that velocities up to about 5 – 5.5 metres per second seemed to
represent a maximum for the members of this group, but that the velocities of
separation attained were nevertheless entirely sufficient to explain, at least in a
general sense, the evolution of the Kreutz family.
The importance of this theoretical development lay in the fact that the slow
dynamical evolution through the gravitational perturbations of planets ceased to
be the principal factor determining the nature of the Kreutz group. Moreover,
fragmentation events no longer needed to be confined to splits occurring at
perihelion. A secondary effect of these innovations to the basic model was a
potential change in the relative importance of the phenomena of subgroups vs.
temporal clusters. As we saw, Marsden felt compelled to come down on the
subgroup horn of this particular dilemma. Sekanina increasingly favours the
cluster horn, but actually hopes that the dilemma itself will eventually disappear.
The Sekanina/Chodas model is still a work in progress, but these authors have
expressed the thought that the subgroups will probably be eventually dispensed
with as fundamental structures of the Kreutz family. In broad terms, this model
already can account for clusters of sungrazers comprised of both Subgroup I and
Subgroup II objects; each cluster spread over several years and separated from
previous and subsequent ones by intervals of 80 – 100 years. Each cluster
represents the fragments of a proto-fragment that experienced one or more
secondary fragmentation events at large distances from the Sun. This accounts
for the small time interval between cluster members, but the proto-fragments
themselves are products of the splitting of a parent comet close to its previous
perihelion passage. It is the separation of the various proto-fragments that is
reflected in the far longer time between the appearances of the clusters.
Applying this general theory to the actual family tree of the Kreutz group is,
however, not at all straightforward. This is still a work in progress and several
revisions have already been made as increasing observational and theoretical
knowledge of the group forces modifications and corrections to earlier
proposals.
The initial paper by these authors, published in 2004, attempted to explain
the sungrazers since 1843 in terms of a progenitor comet that first split far from
the Sun early in the first millennium. The major fragments returned early in the
twelfth century, where they tidally split near perihelion into proto-fragments,
which subsequently fragmented far from the Sun into the objects that passed
perihelion between 1843 and the present day. A revised version of this model
was published in 2007, but before looking at this, some very important advances
made by these authors in the 2004 and other early papers must first be noted.
For a start, Sekenina’s re-examination of the published observations of the
great Comet of 1843 enabled him to show that the relatively short period
traditionally given for this comet was not accurate. Not all of the positional
measurements were trustworthy, and when a small number of doubtful ones were
eliminated, the period of this comet came out similar to those of the other well
observed sungrazers. The need for an unobserved comet in the late 1400s was no
longer required. The comet of 1880 could be explained if it and the 1843 one had
split away from each other within the sunward part of their common orbit, albeit
some 100 – 150 days after perihelion passage itself, which presumably took
place sometime close to 1100. A later piece breaking off the 1880 object whilst
far from the Sun explained the appearance of the other bright sungrazer in 1887.
Even the troublesome White-Ortiz-Bolelli could be brought to heel within
the new model, although the situation in that instance was rather more
complicated. Sekanina and Chodas hypothesised that the splitting of the major
fragment that produced the 1882 and 1965 comets also resulted in a third proto-
fragment, which later split apart far from the Sun during the mid 1700s. The
fragment produced by this later event is hypothesised to have moved away from
the parent proto-fragment at a relative velocity of between 3 and 5 metres per
second in the general direction of the Sun and to the north of the orbital plane.
This enabled it to assume the type of orbit characteristic of Subgroup IIa
sungrazers. It was this fragment that arrived in May 1970 as White-Ortiz-Bolelli.
The rest of the proto-fragment from which it split away presumably arrived a
couple of months later in July of the same year, however at that time of year it
was so poorly placed as to pass by unnoticed.
It is possible that Comet Pereyra, despite its typical Subgroup I orbit, is also
more closely related to the 1882/1965 pair than to 1843 and 1880. Initially,
Sekanina and Chodas had it breaking away from the same proto-fragment as
White-Ortiz-Bolelli, however in their 2007 work, which placed more stringent
limits on the relative velocities of separation of fragments, they questioned this
earlier scenario in view of the rather high separation velocity required. They
suggest an even more complex fragmentation scenario might be required in
which Pereyra becomes the end product of a fragmentation process involving
further steps than initially hypothesised. This further scenario has not yet been
worked out in detail.
A problem with the first draft of the Sekanina/Chodas theory is that it was
oversimplified in so far as only the undisputed Kreutz members from 1843
onward were considered. However, once the basic thesis was presented, the
authors extended their model to include probable Kreutz comets back to the
sixteenth century and, where necessary, to adjust the model accordingly. The
effect of adding these earlier comets was to take the centre of mass of the system
(in effect, the hypothetical perihelion date that the proto-fragment would have if
it had not suffered further fragmentation) back further in time. Because the
assumption that one of the two original fragments was observed as the comet of
1106, this had the effect of fixing the time of the previous perihelion passage and
therefore placing the centre of mass further back in time effectively reduced the
length of the parent comet’s orbital period. We will return to this in a moment.
The authors also placed a tighter limit on the velocities of separation
allowed for the fragments. Observationally, velocities up to 5 or at the most, 5.5
metres per second appear to be the highest attainable by the separating fragments
of Kreutz comets. This value is probably dependent upon the rotational
velocities of the fragments themselves.
Perhaps the biggest alteration made to their original model however, was the
abandonment of the assumption that the 1106 comet moved in a Subgroup II
orbit. The scenario worked better if this comet had a Subgroup I orbit. The
unobserved comet that presumably passed perihelion around the same time was
the one that moved in a Subgroup II orbit and it was the one that split into the
1882/1965 pair. Taking into consideration the obviously large dimensions of the
nucleus of the 1882 object, in comparison with the other members of the group,
it would appear that this second comet experienced little fragmentation before its
twelfth-century tidal split and, even following this, the great bulk of it arrived
still intact in 1882. Somewhat ironically, the brightest and most massive member
of the 1880s cluster appears to have been an interloper. With both parent comets
passing perihelion close to each other in time however, this sort of composite
group membership is not too surprising.
If only the definitive major sungrazers from 1843 to 1970 are taken into
account, the centre of mass of the entire system would have passed close in time
to 1882. There is not the slightest reason to think that the centre of mass of this
system of fragments would have been occupied by a comet, but if the original
comet had not broken up, we might expect that it would have returned to
perihelion at that time. Taking the earlier suspected Kreutz objects into
consideration, as we have seen, moves the centre of mass back in time, probably
back by at least a century. The original period of Comet Kreutz would then turn
out to be between 600 and 700 years and, working from this and assuming that
the 1106 comet was at least the main nucleus of Comet Kreutz, an earlier
apparition should be sought sometime during the fifth century.
The list drawn up by England includes 2 reasonable suspects during that
century, as we earlier saw. The first of these appeared in February 423 and the
second in February 467. Although England ranked both of these an equal 5, the
available description of the 423 comet’s path through the sky may not have been
entirely consistent with a Kreutz object. Morevoer, the 467 one appears to have
been the brighter and more spectacular of the pair. Perhaps significantly,
Sekanina and Chodas found that linking the earlier object with that of 1106 gave
unrealistically high separation velocities for the suspected Kreutz objects of the
1500s and, tracing the orbit back in the other direction, failed to yield an earlier
suspect. Linking 1106 with 467 on the other hand, enabled all of the suspected
Kreutz comets from 1500 to be explained in terms of cascading fragmentation of
(initially) tidally-separated fragments of 1106. Furthermore, this link yielded a
previous perihelion passage in January of 214 BC; a year in which a comet was
indeed recorded in China. Unfortunately, this object is only very briefly
mentioned and neither its location in the sky nor the month of its appearance is
recorded, making any confirming Kreutz credentials impossible to determine.
The sole account is a couple of lines in the Shih chi to the effect that “a bright
star appeared in the west” sometime during the year in question. The
commentary on this chronicle refers to it as a “broom star’, providing the only
direct evidence that this object was even a comet! Incidentally, neither scenario
provides any possible identification of Comet Kreutz with Aristotle’s spectacular
object of 372 BC, whose Kreutz credentials must now be called into serious
question.
It is rather ironic, in view of some of the wilder speculation over the years
as to what “the original sungrazer” may have looked like, that the most probable
suspect should be an obscure object that gets the briefest of mentions in a single
chronicle. If this object had appeared very early in the year, which would agree
well with the predicted January perihelion passage of Comet Kreutz according to
the Sekanina/Chodas model, it would not have been well placed from China.
Moreover, if it had not previously experienced any serious splitting, at least not
for several revolutions, it’s nucleus may have been coated by a thick layer of
insulating material which probably damped down its level of activity somewhat.
It is conceivable that the “original Sungrazer” may have been intrinsically fainter
than the brightest of its subsequent fragments. This would not be the only time
that fragments of a split comet rivalled or even outshone the parent!
Sekanina and Chodas also draw attention to the fact that, assuming the
467/1106 connection to be the correct one, Comet Kreutz would have passed
closer to the Sun at its 214 BC perihelion passage that at any of the earlier ones
computed back to 2086 BC and closer than any of those computed on the
assumption of a 423/1106 identity. Although not strong evidence in favour of the
identification, it could be argued that the extra tidal stress placed upon it at that
perihelion meant that if it was destined to “crack”, that was the most likely time
for this to happen. That it did “crack” – or at least suffer some weakening – is
implied by the second comet of the early 1100s. This, the immediate parent of
1882 and Ikeya-Seki, is best explained if it broke away from the main object far
from the Sun on its way toward the fifth century perihelion passage. Apparently
it passed unseen than as well, but must have appeared around the time of the
main comet – presumably close to the year 467.
According to this revised model, clusters of bright sungrazers – fragmented
nuclei of the 1106 comet – may continue to arrive into the future. Whether any
remaining proto-fragments are massive enough to non-tidally fragment into
pieces large enough to be visible as bright sungrazers is another matter (and may
be doubted if only the 2 largest fragments of the 1882 comet are massive enough
to return as clusters, as we suggested earlier). Nevertheless, the model suggests
that given realistic velocities of separation, clusters can theoretically keep
arriving every 80 – 100 years until around 2120 (on the assumption that the
comet of 467 is the same as that of 1106) or as late as 2200 on the less likely
scenario that the 423 comet was the earlier return of 1106.
Although the possibility of further clusters does not necessarily mean that
these will happen, Sekanina and Chodas provide a piece of evidence that
appeared to support just such a prediction. This evidence came in the form of the
continuing, and increasing, stream of minisungrazers being found in SOHO and
STEREO data. As already remarked upon, their annual numbers increased from
about 50 in the mid 1990s to over 180 by 2010. These were not, according to
Sekanina and Chodas, directly associated with the cluster of bright sungrazers of
the 1960s, but more likely represented a string of debris connecting adjacent
nuclei of the broken 1106 comet. Like the 1882 object, this comet would
probably have resembled a string of beads, had telescopes been available in
those days. The resent SOHO sungrazers may represent the small pieces (and the
cascading fragments of these) comprising a section of the “string” between 2
proto-fragments. One of these has passed already as the cluster of the 1880s. The
other is still to come and, considering the increase in the numbers of SOHO
comets in recent years, may not be too far away. The authors opined in their
2007 paper that the leading member may only be several years in the future.
While on the subject of SOHO sungrazers, it is interesting to note that the
authors plotted the orbital elements of 1,000 of these observed until 2006 and
found that the main concentration lay along the Subgroup I region. Nothing
represented Subgroup II! That is not to say that there were no SOHO comets
with orbits similar to 1882 and Ikeya-Seki, but there was no specific feature of
the graph there; Subgroup II comets were just part of the random scatter of the
“outliers” or the comets that did not closely conform to the main core of the
distribution. The subgroups effectively vanished. What had been called
“Subgroup I” was apparently the main core of the Kreutz system.
That is how the picture emerged in 2007. But then ...

ALONG CAME COMET LOVEJOY!
When this comet developed into a spectacular southern object during the
festive season of 2011, it was widely assumed by comet enthusiasts that we were
seeing the predicted forerunner of the twenty first century sungrazer cluster.
Nevertheless, not everything was straightforward. Although a fragment of
the 1106 comet may have assumed an orbit quite unlike the core Subgroup I
type, the extent of Comet Lovejoy’s divergence from this type of orbit came as a
surprise to many. The comet’s discoverer noted that the orbit had more in
common with that of White-Ortiz-Bolelli than with any of the other previous
bright Kreutz objects, although the differences were still considerable and the
very small perihelion distance of Lovejoy was definitely more in line with a
Subgroup I orbit. One suggestion placed the comet in a new Subgroup III, but it
could just as easily have been called “Subgroup IIb”, “Subgroup Ia” or even, at
the risk of becoming ridiculous, “Sub-subgroup IIa–a”! Maybe this shows that
the whole subgroup classification is starting to break down.
A potentially more serious objection however, was the relatively short
period implied by the early elliptical solutions. Various attempts to compute an
elliptical orbit gave some widely discrepant results, but the initial ones suggested
a period between 300 and 400 years and the suggestion was put forward that this
comet may be one of the suspected sungrazers of the 1600s (or a fragment of one
of these) returning to perihelion on a moderately short-period orbit. If that were
so, Comet Lovejoy was not the expected leading member of the predicted new
cluster and its apparent fulfilment of the Sekanina/Chodas prediction simply
fortuitous.
Because sungrazers reach such high velocity around the time of perihelion,
slight changes in eccentricity result in large changes in orbital period and slight
errors in the determination of eccentricity values can lead to large errors in
computed periods. In the case of Comet Lovejoy, this was exacerbated by the
early loss of a discrete nuclear condensation. Prior to perihelion, when a true
condensation was present, the comet quickly plunged into twilight and the
observational arc was rather brief. Then, by the time it emerged again into a dark
sky, the starlike condensation had already elongated into a bright streak directed
down the centre of the dust tail. As we saw earlier, this feature has been
explained as a debris plume following the cataclysmic fragmentation of the
nucleus and is believed to consist entirely of small solid particles. The more
massive particles remained near the location of the destroyed nucleus while the
smallest ones travelled furthest in the direction opposite that of the Sun. In a very
extended sense, the largest particle(s) could be said to represent the nucleus, but
even these were small enough to be affected by solar radiation pressure in a way
that the original nucleus was not. Nevertheless, plotting the point at which the
bright streak intersected the pseudo-parabolic envelope of the comet’s head
provided the nearest thing to measuring the theoretical position of the (no longer
existent) nucleus. A series of just such precise measurements was secured by
Rob McNaught of Siding Spring Observatory in New South Wales, Australia,
and these proved invaluable to Sekanina and Chodas in their calculation of a
definitive elliptical orbit for this comet. It was, of course, necessary to derive as
accurate an orbit as possible before one could decide whether this comet really
was the first bright member of the expected new cluster or whether it was a
member of one of the clusters of the 1600s making an encore appearance. If a
sufficiently reliable orbit suggested a previous perihelion passage around 1100,
then it was a member of a new cluster, but if it had returned as recently as 1600 –
1700, then it clearly was not.
In order to perform the required calculations, the authors computed various
orbital periods using high quality pre-perihelion measurements of the (at that
time, still intact) nucleus. Analysing the bright streak as a synchrone (i.e. a
feature formed by the simultaneous release of particles of varying sizes), it is
obvious that the hypothetical position of the missing nucleus must have been
located somewhere on this synchrone, simply because it was the source of the
particles comprising the elongated feature. McNaught’s measurements of the
streak’s tip and orientation enabled the authors to compute the position that the
nucleus would have occupied, had it still existed, on the dates at which these
measurements were taken. The hypothetical positions of the nucleus were fixed
as the co-ordinates of the points of intersection of the streak’s axis with the lines
of forced orbital-period variation which the authors derived from orbital
solutions using the positional measurements obtained prior to perihelion. From
this unusual method, a definitive elliptical orbit was computed.
So, did the resulting orbit indicate the time of the previous perihelion
passage as the early 1100s or the 1600s?
Actually, neither!
The final elliptical orbit solution had the comet returning around the year
1329!
Nevertheless, this latest bright sungrazer could still be accommodated with
only a slight extension of the 2007 version of the model. In their 2012 paper on
the comet, Sekanina and Chodas proposed the following scenario.
During its return in 467, Comet Kreutz split tidally at perihelion. This is a
new consideration, although it is an entirely reasonable one. We might go so far
as to say that it was expected, in view of the presumed weakening of the comet
in 214 BC and the subsequent breaking away of the 1882/1965 progenitor, far
from the Sun, on the way to its fifth century perihelion passage. One of the
fragments of the fifth century split retained the original 639-year period and
returned as the comet of 1106, which then fragmented into most of the more
recent sungrazer clusters as described by the authors’ 2007 paper. However, the
second fragment increased its orbital period to about 862 years. This could be
accomplished if it is supposed that this fragment broke away close to perihelion
with a velocity of separation of about 2.2 metres per second in a direction
essentially matching the comet’s direction of motion. It then returned around the
year 1329, plus or minus several years or even decades, as these values cannot
be determined precisely (nor do they need to be for the overall validity of the
model to hold). In any event, this fragment split again at the fourteenth century
perihelion passage. This time as before, one piece maintained the orbital period
of the parent fragment (about 862 years in this instance) but the second separated
in a direction opposite that of the comet’s motion at a separation velocity of
around 3.6 metres per second. This had the effect of reducing the period of the
second fragment, almost back to that of the original Comet Kreutz!
 This second, shorter-period, fragment was not Comet Lovejoy, but became
the proto-fragment of a new cluster of non-tidally separated fragments which
included Comet Lovejoy. According to the model, the fragmentation of this
proto-fragment would have taken place far from the Sun, in the general region of
aphelion. Assuming the original orbit of the proto-fragment to be fairly close to
that of the core members of the Kreutz group (i.e. a “Subgroup I” type orbit)
several cascading fragmentation episodes, with velocities of separation
averaging approximately 2 – 3 metres per second, must have taken place close to
aphelion for a fragment having the orbit of Comet Lovejoy to be produced.
Although not immediately relevant to the story of Comet Lovejoy, nor of
any practical interest to anyone reading these words, this scenario also leaves us
with a second fragment from the fourteenth century fragmentation event due to
arrive sometime around the year 2200. Presumably it has also experienced
further fragmentation events far from perihelion and will most likely return as a
new cluster welcoming in the twenty third century. Of more immediate interest
though, because fragmentation events in the region of aphelion cause only slight
alterations in the orbital periods of the resulting fragments, there are grounds for
thinking that further Lovejoy-sized pieces are on their way and that this comet
really was the leading member of a new cluster. That was the conclusion of
Sekanina and Chodas, even though their earlier identification of the predicted
cluster with a fragment of the 1106 comet would then need to be abandoned.
Although it is not set in concrete that other bright sungrazing comets will
appear in the coming years, the chances that they will look pretty good.
Clustering appears to be the rule rather than the exception, and there seems no
reason to think that the present case will be any different. The chance that
Lovejoy was the only relatively sizable fragment to survive the relentless
process of cascading fragmentation, although clearly possible, nevertheless
seems rather unlikely.
It would be helpful in refining the model further if the hypothesised
fourteenth century comet could be identified in the records, but attempts thus far
have drawn a blank. The only tolerably suitable candidate for Kreutz
membership that century appeared in November 1381. It was noted as a possible
sungrazer by Hasegawa but only given a probability score of 4 by England.
From the description by Korean astronomers, it does not seem to have been very
large and, if a sungrazer, may not have been robust enough to bequeath
fragments large enough to become naked-eye objects at the subsequent
perihelion passage. In any case, Sekanina and Chodas consider the date to be too
late in the century for this to have been the Lovejoy progenitor. Several years or
even a couple of decades either side of the theoretical 1329 date would be
acceptable considering the uncertainties involved, but 52 years is considered a
little too wide of the mark. At most, we might suggest that if the 1381 comet
really was a Kreutz sungrazer and if the Lovejoy progenitor arrived somewhat
later in the century than the formal date suggests, then it may have been a fellow
member of the same cluster.
While we are speculating, we might just wonder aloud if the tidal
fragmentation of 467 produced any other fragments, in the manner of those
observed in 1882 and hypothesised in 1106. It is quite likely that this did happen,
but there is also a pretty good chance that these were smaller and broke up
completely before returning to perihelion even as clusters of bright objects. Yet,
suppose that another was large enough to survive. Maybe some of the “possible”
sungrazers appearing between the eleventh and fourteenth centuries were their
fragments and therefore true members of the Kreutz group, albeit ones not
directly related to the comets of more recent years and centuries.
The comet of 1232 is one that immediately springs to mind, being located
between the observed comet of 1106 and the hypothesised one of circa 1329.
This is the “elephant’s trunk comet”; a description given by a Chinese
astronomer that was remarkably echoed by an Indian observer of the Great
Comet of 1882. Hasegawa suspected that this comet might be a member of the
Kreutz group and suggested perihelion on October 13. England included it in his
catalogue, but gave it a probability score of just 4.
The present writer would probably rate this comet’s score somewhat higher
if observations ceased at the end of October. At that time, the tail was over 30
degrees long and both the comet’s path through the constellations and the
development of its tail seemed quite consistent with that expected for a Kreutz
comet. The comet was drowned out by moonlight in early November and picked
up again after the Moon began to wane. By November 11 it was said to have
sported a tail measuring 60 degrees. This is rather long for a Kreutz comet at that
time of year, although not sufficient to rule it out as a sungrazing candidate. The
tail in early November was apparently not very intense however, if it became lost
in moonlight. Sungrazer tails tend to be unusually intense and, although about 3
weeks past perihelion, this one should still have been intense. The comet must
have been intrinsically bright, as observations continued until December 14, so
the tail would probably have been very intense for several weeks following
perihelion if the comet was truly a Kreutz member.
 If the 1232 comet was a Kreutz and if it originated as a fragment split away
from Comet Kreutz during perihelion passage in 467, its period would have been
about 765 years. If the comet, or a fragment thereof, maintained the same period,
other things being equal it should have returned around 1997. Assuming that it
fragmented non-tidally far from the Sun, this may have given rise to a cluster of
bright sungrazers during the 1990s and early 2000s or thereabouts. To date, no
bright sungrazer having compatible period has been found.
Returning now from these speculations and coming back to Comet Lovejoy
and its place in the Kreutz scheme of things, we suspect that the conclusion in
the 2007 paper as to the significance of the SOHO minisungrazers may need
changing. We recall that these were explained as the products of small fragments
forming a filament of debris between the nuclei of the 1106 comet – the “string”
along which the “beads” (major fragments, each of which subsequently became
a proto-fragment of a cluster) were threaded. However, if Lovejoy and the
predicted cluster of which it is believed to be a part is not directly related to the
1106 comet and if the increasing annual numbers of SOHO comets is directly
associated with the approaching cluster, then it follows that most or all of these
comets are related, not to the 1106 comet, but to the fourteenth century Lovejoy
progenitor. It would be beneficial if some of these minisungrazers could be
followed over a long enough observational arc for a good elliptical orbit to be
derived, but that seems a vain hope. The small sungrazer C/2008 D3 (STEREO)
was followed for 2 days in late February, 2008 (perihelion occurring on February
22), and an elliptical orbit having a period of between 120 and 180 years
computed. At face value, this might suggest association with fragments released
by the comets of the nineteenth century, but the velocities of separation would
surely be too high for this to be taken seriously. The short period is almost
certainly incorrect, but there is no hint as to what the real period may be.
Another minisungrazer followed over a longer than usual arc was C/2012 E2
(SWAN), but the orbit is not sufficiently well determined to provide a realistic
estimate of its orbital period. More reliable evidence, I think, is the observation
that the number of SOHO comets discovered each year has increased
dramatically since 1996. It would surely be an odd coincidence if this apparent
build up to the arrival of a bright sungrazer should have occurred without the
small comets and the bright one being part of the same clustering and having the
same immediate origin!
In their 2012 paper, Sekanina and Chodas plotted the position of 1565
SOHO sungrazers that appeared between January 1996 and June 2010 according
to their longitudes of ascending node and inclination. The general conclusion
reached from a similar plot of a smaller number in 2007 remained the same, i.e.
the main trunk of the population belongs to Subgroup I with no special
secondary feature around the Subgroup II region. There was, however, a pair of
surprises thrown up by the new analysis. A secondary feature did emerge; a
“branch” striking out from the main “trunk” and comprising about 100 comets.
Moreover, this “branch” split toward the end as a thin but clearly defined “twig”
separated away from it. The “twig” was comprised of around 30 individual
comets. Neither of these represented Subgroup II however. Perched securely on
the “branch” was White-Ortiz-Bolelli and on the “twig” Lovejoy! The 2
“subgroups” represented by just one bright comet apiece are quite conspicuous
amongst the groups smaller members while Sungroup II, which includes 2 of the
brightest major members of the sungrazing group, vanishes into the random
scatter once the smaller objects are included. What significance this may or may
not have awaits future research. By the way, because no comets found after June
2010 were included, it is not presently known whether any of the other
sungrazers of December 2011 were located along this “twig”. It is very likely
that at least some of these, especially the pair that shepherded the bright comet
toward perihelion on December 16, were fragments that had come adrift from
the main mass on its way toward the Sun.

WHAT IS “TYPICAL” BEHAVIOUR FOR A
SUNGRAZING COMET?
We have already remarked how Comet Lovejoy did not behave in the
manner that most astronomers expected. The steep rise in brightness on the way
to perihelion and the fading on the final pre-perihelic leg of its journey
conflicted with the other sungrazer that had been well observed prior to
perihelion – Ikeya-Seki. Yet, it behaved much as the smaller members of the
group behave and in that sense should be regarded as typical. This is also
supported, as we saw, by the observation that several bright sungrazers of earlier
years were well placed prior to perihelion and, based on their post-perihelic
performance, should have been pretty bright, yet remained undiscovered until
after perihelion. White-Ortiz-Bolelli was named as the classic example.
If a sungrazing comet continued to brighten all the way to perihelion at the
same rate as the average comet, the brightness briefly attained near the Sun
would be tremendous. Without entering into the debate about what the “average”
rate is – or even if, strictly speaking, there is such a thing – if we simply assume
that most comets increase in brightness at something like the inverse fourth
power of their distance from the Sun, we can see that even a comet of quite
modest absolute magnitude can far outshine the full Moon when within 0.01 AU
of the Sun. Take Ikeya-Seki for example. This comet had an absolute magnitude
of approximately 6 (bright, but not exceptionally luminous) and increased in
brightness as it approached perihelion at approximately the inverse fourth power
of its distance from the Sun. At perihelion, if this trend held, it should have
reached magnitude -15 or around 10 times the brightness of the full Moon.
Assuming a similar trend for the Great Comet of 1882, which seems to have
been about 5 magnitudes brighter than Ikeya-Seki in terms of absolute
magnitude, yields a peak brightness of about -20, less than 7 magnitudes fainter
than the Sun itself! Comet Lovejoy, which was about 9 magnitudes fainter than
Ikeya-Seki prior to perihelion, would have reached about -9 if the steeper
increase held or around -6 if it had increased at a similar rate to Ikeya-Seki.
Are magnitudes of this order truly attained by sungrazers? Certainly, they
were not attained by Comet Lovejoy and the fact remains that of the major and
undoubted sungrazers thus far observed since 1843, only one third were
observed with the naked eye in daylight. Those of 1843, 1882 and 1965 were;
the ones observed in 1880, 1887, 1945, 1963, 1970 and 2011 were not although
the latter was observed telescopically in full daylight. Moreover, if we include
sunskirting comets as well, the situation becomes even odder. Although a
sunskirter of equal absolute magnitude would not become as bright as a true
sungrazer, very high levels of brightness might still be expected. For instance,
something of the same absolute magnitude as Ikeya-Seki but coming to
perihelion at 0.03 AU could still reach around -9 or even greater if forward
scattering is taken into account. Yet, strangely, no known naked-eye daylight
comet has thus far been a sunskirter. Indeed, of the 6 sunskirters that reached
perihelion between 1900 and the end of 2011 (excluding the very small ones
seen only in SOHO images), not one was observed in daylight – not even
through a telescope. Admittedly, one was found after perihelion and another
faded out on its way to the Sun, but 3 were actively sought telescopically during
the daytime when they were close to perihelion, but without success.
Possibly the most telling of these failed attempts concerned the sunskirting
Seki-Lines of 1962. This was an intrinsically rather bright comet that passed
perihelion at 0.03 AU from the Sun on April 1 of that year. At that time, the
comet should have been well clear of the Sun’s limb with a predicted brightness
of around -7. This should have made it an easy naked-eye object given clear
skies without too much glare surrounding the Sun, yet it turned out to be
something of an April fool’s day joke! Searches were made, but nothing was
seen, not even telescopically. Strangely however, the final pre-perihelic
observations and the first post-perihelic ones, in both instances made when the
comet was deep in twilight, found it a brilliant object of magnitude -1 or
brighter, just as predicted!
An interesting study of this comet was made by B. Jambor in 1973. This
author was not especially interested in its brightness, but rather in the evolution
of its impressive dust tail. Following perihelion, this comet displayed a broad tail
divided by a dark lane running down its centre, giving it a somewhat “hollow”
appearance. What Jambor found was that the dark lane was in the position where
particles released during the several hours around perihelion should have been
located. In other words, although the particles released prior to and those
following perihelion were clearly present and constituted both segments of the
divided tail, those that should have been released at perihelion were missing.
Jambor hypothesised that a dense cloud of particles was probably released just
prior to perihelion and acted like a sun-shade until finally dispersing several
hours later. While the “sun-shade” was present however, the comet’s activity was
effectively turned off, although it resumed at full strength shortly after perihelion
passage. Although no mention was made of the lack of daytime observations, the
missing tail particles and the failure of observers to see the comet at perihelion
must surely be alternate faces of the same coin.
We will come back to this later. First however, let’s keep things in
perspective by recalling that the sungrazers that did become naked-eye daylight
comets were undoubtedly the brightest objects of their kind ever recorded. The 3
since 1843, plus that of 1106 surely outshone all other comets because of their
proximity to the Sun. Unless we imagine a super comet of unprecedented
absolute magnitude, the only real chance that a non-sungrazer has of competing
with these brightness levels is to make a very close pass of Earth almost in line
with the Sun and let forward scattering of sunlight send the brightness through
the roof. Even here however, the coma would attain such large size that it is
unlikely to be seen again the daytime sky, leaving the fainter (albeit still
brilliant) central condensation visible as an almost star-like dot of light. The
classic example of this was Halley in 1910 which almost certainly had a coma
over a degree across when it passed near Earth and transited the Sun. The
integrated light from forward scattering of coma dust must have been very
considerable, yet all but the central condensation was lost against the sky. A
handful of reliable naked-eye sightings of the central condensation (described as
a small bright spot) were made from Tasmania, where it was tracked almost into
the limb of the Sun. Tasmania, by the way, still has the atmospheric station from
where the standard of Earth’s atmosphere is measured, so perhaps the extra clean
air of this island contributed to these observations!
Possibly, the brightest known comet other than a sungrazer was Skjellerup-
Maristany in 1927. At perihelion, this was “only” about magnitude -2, but in the
following days passed through such large phase angles that forward scattering
increased its apparent brightness to possibly as high as -9, rendering it a
conspicuous naked-eye object just 1 – 1.5 degrees from the Sun. It was so bright
that a number of independent discoveries were actually made in broad daylight
at that time. This comet was rather typical of the majority of naked-eye daylight
comets, namely, it had a small perihelion distance (though certainly well outside
sungrazing or sunskirting distances) and was in strong forward scattering
geometry at the time of the daytime sightings. Comets in this category, with
perihelion distances ranging from a little more than 0.1 AU to a little less than 1
AU, which are intrinsically rather bright and which pass through large phase
angles where the effects of forward scattering are very pronounced, comprise
most of the daylight comets reported over the years. The second largest group
are the bright sungrazers, but sunskirters have remained conspicuous only by
their absence, at least when naked-eye daylight comets are considered.
Two telescopic daylight comets of the nineteenth century were, however,
sunskirters. That century saw the discovery of 7 sunskirting comets, some of
which became spectacular objects in the night sky (especially the Great Comet
of 1865; C/1865 B1) but were not observed in daylight. Telescopic daylight
observations were, however, made of C/1847 C1 (Hind) on the day of perihelion
(when it was 0.04 AU from the Sun) by the comet’s discoverer, J. Hind, and his
colleagues shortly before perihelion. Hind also mentioned that it was seen on
that same day by observers in Truro and Ynys Mon and “about noon” by a
clergyman on the Isle of Anglesey. It appears that these observations were also
telescopic. Brightness prediction based on the comet’s magnitude earlier in the
month suggested that it should have been about -3 at that time, perhaps even
reaching -6 closer to perihelion when, however, it would have been closer to the
Sun’s limb. Just before perihelion when these observations were made, the
comet was in good forward-scattering geometry, so we might have expected
even brighter magnitudes, but -3 or -4 appears to have been about the peak of the
comet’s performance, judging by the available reports.
 The second telescopic daytime sunskirter was C/1882 F1 (Wells). This
comet reached perihelion at 0.06 AU on June 11 of that year and became the
subject of several daytime telescopic sightings between June 6 and 12. Its
brightness behaviour prior to perihelion suggested that it may have reached
about -8 magnitude near perihelion passage, however the available observations
do not indicate this level of brightness. One observer on June 11 remarked that
the comet was not as bright as Capella, which was also visible in his telescope
during daylight at the same time. This suggests a brightness of between 0 and
first magnitude, even though the brightness parameters based on earlier
observations suggested about -7 at that time. The observational geometry of this
comet did not favour brightness enhancement through the forward scattering of
sunlight.
From what we have been saying, it appears that most comets that come to
perihelion at sungrazing or sunskirting distances do not become as bright as their
behaviour while further from the Sun would suggest. Yet, what can we say about
the brightest of the Kreutz comets? These certainly became very bright, but were
they as brilliant as a brightness forecast based upon their absolute magnitude and
rate of brightness increase days or weeks prior to perihelion would reasonably
lead us to expect?
Maybe not. The Great Comet of 1843 was not seen in the immediate
vicinity of perihelion and the early sightings of this comet give only the very
crudest estimate of its intrinsic brightness. One daytime sighting was reported
prior to perihelion, but most occurred the following day when it may have
experienced a post-perihelic brightness surge similar to that of Comet Lovejoy.
The most thoroughly observed sungrazers before and after perihelion prior to
Lovejoy were the Great Comet of 1882 and (especially) Ikeya-Seki. As we have
seen, the first of these was amongst the intrinsically brightest comets on record
whereas the second was relatively average in respect to its absolute magnitude
and would have been nothing special were it not for its sungrazing encounter.
Yet, each appeared to attain remarkably similar brightness during the several
hours spent in the close neighbourhood of the Sun. That, at least, is the
conclusion drawn from a comparison of the Finlay and Elkin observations of
1882 just prior to its transit of the solar disc (when it was located just 2 hours
from perihelion with forward scattering was at maximum and the comet just 0.02
AU from the Sun) with the Ikeya-Seki observations of Roemer 8 hours before its
perihelion passage. That comet was then about 0.05 AU from the Sun and also in
strong forward-scattering geometry, albeit not quite as strong as the former
comet. Both objects seem to have been about magnitude -10 or -11 at the time of
these observations. The 1882 comet was not seen at perihelion and, indeed, there
may have been no attempts made to observe it then. However, it was clear of the
limb of the solar disc at the time and if it truly did become as bright as the light
curve based upon its absolute magnitude implied (about -20!) somebody
probably would have spied it. Taking into consideration the earlier daylight
observations, it seems very unlikely that the comet even came close to attaining
that level of brightness.
The situation with Ikeya-Seki is interesting. Just 10 hours before perihelion,
de Vaucouleurs estimated its brightness as -10 and commented that it was then
about a magnitude brighter than predicted. The forward scattering effect may
have contributed to this, although dust particles responsible for this effect would
have short lifetimes so close to the Sun. It is interesting to note that the comet
was then just 0.06 AU from the Sun and entering the region where the brightness
of Lovejoy and most of the SOHO comets had already passed their brightness
peak. Nevertheless, there seems to have been a continued increase in brightness,
with Roemer estimating the brightness to be in the range of -10 or -11, 2 hours
later and astronomers at Tokyo Observatory’s Mount Norikura Coronagraph
Station described it at the time of perihelion as being “10 times brighter than the
full Moon”. This description, if taken as a serious estimate and not simply as an
impression of the comet’s brightness, implies a magnitude of almost -15, very
close to that actually predicted by the comet’s light curve. This estimate, and its
apparent agreement with the light curve, is probably not to be taken too
seriously, although the perihelion brightness does appear to have been greater
than -11 or thereabouts.
The apparent convergence of the peak brightness of the intrinsically very
bright 1882 comet and the intrinsically rather mediocre 1965 one probably hints
that these levels of brightness hit the ceiling of what is attainable. Moreover,
each comet was in relatively strong forward-scattering geometry in the hours
when most of these estimates were made. Forward scattering would not,
however, have been a factor of the Ikeya-Seki observations at the time of
perihelion. Nevertheless, the loss of any enhancement from this effect may have
been compensated for if the partial disruption of the nucleus beginning about 30
minutes before perihelion (or, at least, first noticed at that time) resulted in an
extra surge of dust production. What might have been described as a “flare” if
occurring at larger distances from the Sun may have accompanied the nucleus
disruption.
 All comets depend for their brightness on light from the Sun, either
reflected from dust particles or absorbed and re-emitted by gas molecules and
ions. A comet almost grazing the photosphere might become almost (although
not quite) as intense as this, but because of the extreme conditions prevailing at
these distances, released dust would very quickly evaporate and gas would be
quickly ionised and swept away. Moreover, as a comet passes through the solar
corona it is effectively met by a million-degree, million-plus kilometre per hour
headwind. The corona is, admittedly, very tenuous, but denser coronal streamers
and even prominences must sometimes be encountered by sungrazing comets
and we could imagine the very tenuous outer regions of the head being stripped
away. Deeper down in the corona, things become a lot more serious. Research by
J. Brown et. al. has shown that the inner corona of the Sun, below 7,000
kilometres from the photosphere, is dense enough to turn the nucleus of a comet
into a “solar meteor”! The head of a comet must therefore be very small at
distances of just tens of thousands of kilometres from the photosphere, so even if
it could become as intense as the surface of the Sun itself, the small size must
severely limit its total magnitude. Actually, an empirical relationship has been
determined relating the diameters of comet heads and solar distance. The
diameters of the head varies as the inverse square of its distance from the Sun.
Taking the average diameter of a comet at 1 AU as about 300,000 kilometres,
this implieds a diameter of about 3,000 kilometres at 0.1 AU, and just 30 at 0.01.
Too much stress should not be placed on these figures as they are pushing a
purely empirical – and approximate – relationship to extremes which it was not
intended to deal. The empirical relationship probably levels off somewhere
within a few tens of millions of kilometres of the Sun and sungrazers or deep
sunskirters large enough to become bright naked-eye objects are likely to have
diameters in the low thousands of kilometres at perihelion. The Elkin/Finlay
estimate of the 1882 comet as 4 seconds of arc just prior to its transit implies a
diameter of around 3,000 kilometres. Observations of Ikeya-Seki not long before
perihelion indicated a diameter of around 5,000 kilometres. The measured
diameter of the 1882 comet is probably about as large as we can expect so close
to the Sun and the intensity must have been near maximum as well,
remembering that this was an unusually bright comet in terms of its absolute
magnitude. It may have become a little brighter at perihelion, though probably
not by much. I suspect that the brightness ceiling for any sungrazing comet (and
I think that means for any comet, at least any comet belonging to the Sun’s
family) is probably between magnitude -11 and -13 or thereabouts. Ikeya-Seki
was probably seen to reach it when it was tracked through perihelion. The 1882
comet was missed at perihelion, but probably was about the same despite its far
brighter absolute magnitude. That of 1843 may have, but it passed a lot closer to
the Sun than the later objects and may have experienced a perihelic dip in the
manner of Comet Lovejoy with its similar perihelion distance. We simply do not
know.
The brightness of a sungrazing comet may to some degree be influenced by
the fact that at these very small distances, more and more of the “visible” Sun
slips below the horizon. At the perihelion of Lovejoy (just 0.2 of a solar radius)
the appearance would have been that of flying headlong over a roiling, boiling
photospheric landscape, with much of the Sun below the horizon (remember our
imaginary trip on the nucleus of this comet?). Comets passing at distances as
close as this are being “activated” from the region of the Sun immediately
beneath them, not from the disc of the Sun as a whole.
Nevertheless, the big check on a sungrazer’s brightness as well as on the
size of its head at close solar distances is surely the rapid evaporation of dust
particles released from the nucleus. This was most impressively demonstrated by
Comet Lovejoy which, as noted earlier, faded, lost its tail and reduced to a tiny
point of light around the time of perihelion. At the actual moment of perihelion
passage itself, the comet was unobservable through being occulted by the disc of
the Sun. We do not really know how faint it became then, but it would not be
surprising, had it remained clear of the solar disc, if it would have faded from the
view of the LASCO coronagraph altogether. This is hinted by the very faint
appearance as it came back into view and the rapidity at which it was then re-
brightening. In any event, the comet clearly appeared to “turn off” around the
time of perihelion, only to surge back with a vengeance during the following
hours. At the time, and recalling Jambor’s analysis of Seki-Lines, I suggested
that something similar may have happened to Comet Lovejoy – a cloud of
particles released from the nucleus that remained sufficiently dense to shield the
nucleus surface from the direct heat of the Sun.
Needless to say, that rather off-the-top suggestion did not win much support.
And rightly so! I place absolutely no faith in it myself anymore.
A far more credible hypothesis was put forward by Sekanina and Chodas in
their 2012 paper on this comet. They found from their analysis of the tail that the
trajectories of all particles released within 1.8 solar radii, or 0.00837 AU, were
missing. They were missing, not because the comet was screened through
perihelion by a dense cloud of particles but simply because they evaporated
almost as soon as they were released from the nucleus. The comet temporally
lost its tail and faded out because the matter released was almost instantaneously
destroyed so close to the Sun. Later, once the comet had withdrawn beyond
0.00837 AU, the survival time of dust particles was again long enough for a
coma and tail to re-form and the comet brightened rapidly, actually surpassing
the lustre attained prior to perihelion. Like the sungrazers discovered in SOHO
images, Lovejoy’s peak brightness prior to perihelion occurred near 10 solar
radii. Several decades ago, L. Jacchia figured that a distance of 0.05 AU –
approximately 10 solar radii – is about the minimum distance that particles of
the type responsible for photographic meteors can approach the Sun. Seki-Lines,
as mentioned, approached the Sun more closely than this at around 0.03 AU.
Perhaps the apparent dip in this comet’s activity around the time of perihelion is
also better explained, not in terms of a screening cloud of particles, but as due to
rapid evaporation of released dust. Lovejoy still possessed a strong dust tail at
that location, however it passed quickly through the 0.05 – 0.03 AU zone on its
way to a far closer encounter with the Sun. Having a larger perihelion distance,
Seki-Lines spent more time in that region during which it’s shed dust may have
by and large evaporated. If this, rather than the hypothetical shielding dust cloud,
was the real reason for the missing tail segment and lack of daylight visibility of
Seki-Lines, it may be quite a common phenomenon amongst sunskirting comets,
especially those rare examples that venture within 0.05 AU of the Sun, and may
explain their absence in lists of naked-eye daylight comets.

HOW LARGE WAS “COMET KREUTZ”?
Many accounts of the Kreutz sungrazing group state that the original non-
fragmented progenitor of the family – what we have for convenience of
expression been calling “Comet Kreutz” – must have been one of the largest
comets to pass perihelion within Earth’s orbit. Some estimates suggest that it
was as large as 100 kilometres across! That would make it a truly enormous
object by cometary standards, but is there any compelling reason to believe that
it was really as large as this?
It must, of course, have been a big comet. The number of fragments alone
does not necessarily require an extra large mass, except that one of those
fragments (the Great Comet of 1882) has also been ranked amongst the largest
comets of the last couple of centuries and is often estimated to have possessed a
diameter of around 50 kilometres, similar to Comet Hale-Bopp of 1997.
Moreover, the rarity of sungrazing comets that are not members of the Kreutz
group, in spite of research suggesting that sungrazing orbits should be not too
uncommon as an end state of dynamical evolution, hints that only objects of
fairly large initial mass can last the distance.
Most sungrazers pass within the Roche Limit of the Sun (about 0.008 AU).
Broadly speaking, this is the distance from the Sun where gravitational attraction
increases so rapidly with decreasing distance that a body lacking sufficient
internal cohesion will simply be pulled apart by the tidal effects if it has a
diameter greater than the critical size. This critical size is given by the formula;

R(max) = Ft1/2 x 1.56x10-2
Where R(max) is the maximum nucleus radius in kilometres and Ft is
tensile strength in dynes/square centimetre.

The tensile strength of a comet nucleus is estimated to be about 1,000 dynes
per square centimetre.
If the tensile strength of the material comprising the nucleus was the only
factor, Comet Kreutz would have been little more than 0.6 kilometres in
diameter for it to have survived its repeated sungrazing passages!
That cannot be correct. Other factors do, however, enter the picture. Over 50
years ago, F. Whipple showed that even at sungrazing distances, a comet’s own
gravity does have a certain stabilising influence and recently B. Gundlach and
colleagues showed that the outgazing of an active comet exerts sufficient force
toward the centre of the nucleus to help hold the body together against disrupting
tidal stresses. They estimated that a comet as large as 27 kilometres diameter
could survive a sungrazing encounter. On the other hand, unknown factors such
as lines of weakness within the nucleus or departure from sphericity make a
comet more vulnerable to tidal stress and work to reduce the survivable
diameter. With respect to the latter, we have no reason to think that Comet
Kreutz had a spherical nucleus and if it was significantly elongated (like Comets
Halley and Borrelly for example) this may have placed severe limitations upon
its size. Just as it is easier to break a banana than an apple, so an elongated
nucleus can more readily be broken apart by tidal forces. This applies even
beyond of the Roche Limit as well as for nuclei too small to be theoretically
affected by tidal stress within the Limit.
The more modest dimensions implied for Comet Kreutz must also mean
that the 1882 comet (as just one of many fragments of Comet Kreutz) cannot be
too large, in spite of its high intrinsic luminosity. As a fairly fresh fragment,
much of its surface was probably exposed ice unlike, say, Hale-Bopp which
probably had a relatively large portion of its surface coated with an insulating
layer build up over many passages through perihelion.
An example of a comet that appeared to have a nucleus free from insulating
crust was C/1996 B2 (Hyakutake), the Great Comet of 1996. Although this
comet was intrinsically quite bright with an absolute magnitude of around 5 at
Earth’s distance from the Sun, radar observations of the solid nucleus during the
comet’s spectacular close approach to our planet revealed a surprisingly small
object. The size could not be estimated with a high degree of accuracy, but a
diameter of around 2.4 kilometres appears to be fairly close to the mark.
We assume that the broken fragments that constitute the sungrazers of today
are, on their inward journey toward perihelion, partially covered by a rather thin
layer of insulating crust, possibly formed during the late stages of the former
apparition of their parent proto-fragments. Much of this is probably stripped
away as they come close to the Sun and the underlying ice starts to sublimate
and blow away the fragile crust. This, as we speculated earlier with special focus
upon Comet Lovejoy, may explain the rapid activation of most of the SOHO
comets and at least some of the larger members of the family. We know that
Comet Lovejoy brightened sharply on the way to perihelion; we strongly suspect
similar behaviour for Comet White-Ortiz-Bolelli and believe it highly probable
for the comets of 1880 and 1887 as well. It may be relevant that nearly all of the
historical “Kreutz suspects” were discovered only after passing perihelion. This
fact may not be entirely explained by the dramatic tail development then. If this
speculation is correct, most sungrazers after perihelion may resemble C/1996 B2
in being active over most of their surface area (it is interesting to note in this
respect that the apparent lack of insulating crust on 1996 B2 was put forward as
possible evidence that this comet is also a fragment of a larger object that
disrupted long ago). Although not pretending to any reasonable degree of
precision here, if we take the roughly estimated 2.4 kilometre diameter of
Hyakutake as the measure for the size of an 100% active comet of absolute
magnitude 5 and assume that the post-perihelic absolute magnitudes of
sungrazers are also those of 100% active comets, the diameters of Kreutz comets
can be estimated accordingly, assuming that the absolute magnitude is
proportional to the (active) surface area. Assuming that these objects are at least
approximately spherical their diameters turn out to be surprisingly small. Except
for 1882, all were either equal to or (mostly) fainter than Hyakutake. The large
objects of 1843 and 1963 turn out to be similar to this comet (say, 2 – 3
kilometres in diameter) while the others are smaller, going down to just 300 –
400 metres for Lovejoy just prior to its cataclysmic fragmentation following
perihelion passage. Even the 1882 comet comes out at having a nuclear diameter
of the order of 16 kilometres.
Given the speculative nature of much of this and the uncertainties that are
certainly involved, too much faith should not be placed in the accuracy of these
figures, but they are suggested here at least as ballpark estimates. In reality, the
fragments are hardly likely to be spherical. This is especially true of the smaller
ones that constitute the end products of many fragmentation events. They could
have just about any shape. Nevertheless, when all the “ifs”, “buts” and other
sundry caveats are taken into consideration, as intended ballpark estimates they
are probably not too far out. It is also encouraging to note that a diameter of 27
kilometres for Comet Kreutz itself, as estimated by Gundlach and colleagues as
around the maximum diameter capable of surviving within the Roche limit if
sufficiently active, is more than large enough to accommodate the entire Kreutz
family observed to the present, if these size estimates are more or less correct.
While we are in the speculating mood, if we assume that the diameter of
Comet Kreutz really was 27 kilometres (it may have been several kilometres
smaller and still remained adequate to account for the sungrazing group) we can
at least estimate its approximate absolute magnitude assuming that it, too, was
100% active following perihelion. Using Hyakutake once more as our standard,
we find that Comet Kreutz would have been about 5.2 magnitudes brighter in
absolute terms, i.e. its absolute magnitude would have been in the vicinity of
-0.2, or roughly a magnitude brighter than the comet of 1882. However, if the
original sungrazer had not suffered any serious fragmentation events, the surface
may have accumulated a more significant insulating crust. If the crust was not
entirely purged at each return and if there were no fragmentation events to help
dislodge or even loosen it, it may have become sufficiently thick after a good
number of perihelion passages to significantly dampen down the comet’s
activity. It may still have been an intrinsically bright comet, though perhaps
more like 1843 or Ikeya-Seki than 1882, which might be why (coupled with poor
placement) it was only given a passing mention when it appeared in 214 BC –
assuming that this was the comet seen by the Chinese that year.
One final speculation. If the comet did build up a significant crust and if its
activity was dampened down, could this have been a contributing factor to
weakening it in 214 BC? If, as Gundlach argues, reduction of activity lessens the
force that helps hold the nucleus together against tidal stress, perhaps a partial
decrease in activity, while still exercising enough force to withstand
fragmentation, may nevertheless have resulted in a degree of weakening
(cracking?) within the Roche Limit. Once begun, the process continued and
apparently (as we have seen) led to the splitting of the proto-fragment of the
1882/1965 pair far from the Sun prior to the comet’s next perihelion passage in
467 AD. Whether true or not, this suggestion is at least food for thought!

THE FUTURE OF THE SUNGRAZING COMET GROUP
Having looked at the likely past history of the Kreutz group, the question
that now must be asked is “How will the group evolve in the future?” What can
be expected from this comet family in the years and centuries to come?
If the past has its uncertainties, the future must be at least as uncertain;
however a few tentative predictions can be made on the proviso that they are not
taken as definitive forecasts.
First, it appears that we are currently experiencing a cluster of fragments,
mostly in the form of minisungrazers of the type being picked up in SOHO and
STEREO data. Thus far, only one moderately large fragment has arrived in the
form of Comet Lovejoy, but there are good reasons for thinking that a small
number of similar bodies will arrive during the coming years or decades. This
prediction is certainly not set in concrete, but the chances that it will be fulfilled
seem quite good.
Looking further into the future, the Sekanina/Chodas analysis of Comet
Lovejoy’s genesis implies that a second fragment of the parent fourteenth-
century comet should return as a cluster of fragments welcoming in the twenty
third century. Whether any other fragments of this comet exist and whether, if
they do, they are massive enough to give rise to other clusters is unknown.
Likewise, we do not know whether other fragments of the presumed perihelic
fragmentation of the 467 comet returned several hundreds of years ago and
whether any of these will return in future centuries as clusters of sungrazing
comets. This possibility can neither be ruled in nor out on present knowledge.
Still further ahead, the latter centuries of the present millennium should
theoretically witness the return of the sungrazers of recent times. However, it is
probable that most of these were small, albeit highly active (at least for a time
following perihelion) and have since progressively fragmented into pieces too
small to reappear as anything more than minisungrazers, at best. The principal
exception is likely to be the main nucleus of the 1882 comet (now officially
designated 1882 R1-b) and, secondly, the largest of this comet’s secondary
nuclei, 1882 R1-c. These may give rise to further clusters around the years 2571
and 2656 respectively. Comet Pereyra is set to come back in 2753 and, because
its nucleus was still active some 4 months after perihelion in 1963, enough mass
may remain to give rise to a cluster of bright objects around that time.
If a cluster of Pereyra fragments does appear during the 2750s, it may
overlap with the fragments of Ikeya-Seki–a, due to return around the year 2764.
Nevertheless, this nucleus was observed to become very diffuse in appearance
late in 1965, casting doubt upon the survival of any sizable fragments. The same
is true of Ikeya-Seki–b which is not due to return until sometime around 3021.
As this nucleus was apparently triple within weeks of perihelion in 1965, it
might be assumed that cascading fragmentation continued apace and that nothing
more than pebbles will remain by the time of its next return.
It is unlikely, in my opinion, that any of the future fragments will match the
largest sungrazers of the past. Another comet the size of the 1882 object is
improbable and I suspect that those of the future will look more like the 1880,
1887 and 2011 comets than those of 1843 and 1965. If any larger ones do appear,
my guess is that the predicted 2571 cluster will be the one most likely to include
them.
After the turn of the fourth millennium, it is likely that the Kreutz
sungrazing group will be consigned to history. Yet, one historic sungrazer
remains; the Great Comet of 1680. Computed periods vary greatly for this
comet, but it seems that they are unlikely to be shorter than about 8,000 years.
Maybe this object has also fragmented and will return thousands of years from
today as another family of sungrazing comets, spaced out over far greater
periods of time than the relatively compact Kreutz group of long before.
Then, looking even further into the mists of the distant future, there is a
small possibility that Comet Hale-Bopp will evolve into a relatively short-period
sungrazing orbit. Following its most recent return in 1997, the period of this
comet was reduced from some 4,200 years to just 2,380 years. It is conceivable
that it could reduce down to Kreutz-like levels and there is also a 15% chance
that the perihelion distance could be reduced from its present position just within
Earth’s orbit right down to sungrazing values. Eventually fragmenting, this
comet could give birth to a very large sungrazing group in the distant future. The
chance of this actually happening is, however, small but the theoretical
possibility may throw some light on the early stages of the dynamical evolution
of Comet Kreutz itself. Perhaps Comet Kreutz once had an orbit not unlike that
of Hale-Bopp, before its perihelion distance shrank to its later very small values
and its orbital period shortened from millennia to centuries.
Just a passing thought here. Some astronomers have drawn attention to
alleged similarities in the orbital elements of the Kreutz group and those of other
comets having small (though certainly not sungrazing) perihelion distances and
the suggestion has been made that some of these may have begun life as
fragments of Comet Kreutz in the relatively distant past. This, however, would
surely imply unrealistic velocities of separation; something far more dramatic
than the fragmentation events responsible for the separation of the Kreutz
members themselves. Although all such apparent relationships are most probably
nothing more than coincidence, it is just possible that some fragments came
away from Comet Kreutz before it evolved into a sungrazer, when its orbit may
have differed significantly in elements other than perihelion distance, and that
some small-perihelion comets may have originated in this manner. Personally, I
have grave doubts as to the validity of this and find simple coincidence more
credible, but the suggestion is nevertheless made for what it is worth.
Somewhat ironically then, in considering the future of the Kreutz group, we
come back again to its past. And there we shall leave our discussion of this
family of fascinating comets. Hopefully though, a new chapter in the life of this
comet family will soon be written with the appearance of another bright member
in our skies. Who knows what new insights will follow from its observation?