Iowa State University

Digital Repository @ Iowa State University

Graduate Theses and Dissertations Graduate College

2013

Objective analysis of toolmarks in forensics

Taylor Nicole Grieve
Iowa State University, [email protected]

Follow this and additional works at: http://lib.dr.iastate.edu/etd

Part of the Mechanics of Materials Commons

Recommended Citation
Grieve, Taylor Nicole, "Objective analysis of toolmarks in forensics" (2013). Graduate Theses and Dissertations. Paper 13014.

This Thesis is brought to you for free and open access by the Graduate College at Digital Repository @ Iowa State University. It has been accepted for
inclusion in Graduate Theses and Dissertations by an authorized administrator of Digital Repository @ Iowa State University. For more information,
please contact [email protected].
 Objective analysis of toolmarks in forensics

by

Taylor N. Grieve

A thesis submitted to the graduate faculty

in partial fulfillment of the requirements for the degree of

MASTER OF SCIENCE

Major: Materials Science and Engineering

Program of Study Committee:

Scott Chumbley, Major Professor
Max Morris
Lawrence Genalo

Iowa State University

Ames, Iowa

2013
 ii

TABLE OF CONTENTS

ABSTRACT ......................................................................................................................................... iv

CHAPTER 1. BACKGROUND .......................................................................................................... 1

A Brief History of Toolmarks ............................................................................................................ 1
Use of Technology for Toolmark Examination .................................................................................. 3
Tools and Their Marks ....................................................................................................................... 4
Toolmark Characteristics.................................................................................................................... 6
Toolmark Comparison Techniques .................................................................................................... 7
Theory of Toolmark Identification ..................................................................................................... 8
Consecutive Matching Striae .............................................................................................................. 9
The Daubert Criteria ........................................................................................................................... 9
Research Related to Toolmarks ........................................................................................................ 10
Statistical Algorithm for Toolmark Analysis ................................................................................... 14
Research Related to Firearms ........................................................................................................... 16
References ........................................................................................................................................ 18

CHAPTER 2. OBJECTIVE COMPARISON OF MARKS FROM SLIP-JOINT PLIERS ....... 22

Introduction ...................................................................................................................................... 22
Experimental..................................................................................................................................... 23
Results .............................................................................................................................................. 28
Discussion......................................................................................................................................... 38
Summary and Conclusions ............................................................................................................... 42
Acknowledgments ............................................................................................................................ 42
References ........................................................................................................................................ 42

CHAPTER 3. CLARITY OF MICROSTAMPED IDENTIFIERS AS A FUNCTION OF

PRIMER HARDNESS AND TYPE OF FIREARM ACTION ...................................................... 44
Introduction ...................................................................................................................................... 44
Experimental..................................................................................................................................... 49
Results .............................................................................................................................................. 54
Microstamp Evaluation ........................................................................................................... 54
SEM Evaluation ...................................................................................................................... 58
Hardness Evaluation ................................................................................................................ 60
Discussion ........................................................................................................................................ 61
Summary and Conclusions ............................................................................................................... 64
Acknowledgments ............................................................................................................................ 65
References ........................................................................................................................................ 65
 iii

CHAPTER 4. GEAR CODE EXTRACTION FROM MICROSTAMPED CARTRIDGES ...... 67

Introduction ...................................................................................................................................... 67
Experimental..................................................................................................................................... 70
Results .............................................................................................................................................. 71
Sig Sauer ................................................................................................................................. 72
Taurus ...................................................................................................................................... 73
Hi-Point ................................................................................................................................... 75
Lacquered Cartridges .............................................................................................................. 77
Gear Code Analysis by Magazine ........................................................................................... 79
Discussion......................................................................................................................................... 82
Conclusions ...................................................................................................................................... 84
References ........................................................................................................................................ 85

CHAPTER 5. SUMMARY AND CONCLUSIONS ........................................................................ 87

ACKNOWLEDGMENTS.................................................................................................................. 89
 iv

ABSTRACT

Since the 1993 court case of Daubert v. Merrell Dow Pharmaceuticals, Inc. the subjective

nature of toolmark comparison has been questioned by attorneys and law enforcement agencies alike.

This has led to an increased drive to establish objective comparison techniques with known error

rates, much like those that DNA analysis is able to provide. This push has created research in which

the 3-D surface profile of two different marks are characterized and the marks’ cross-sections are run

through a comparative statistical algorithm to acquire a value that is intended to indicate the

likelihood of a match between the marks. The aforementioned algorithm has been developed and

extensively tested through comparison of evenly striated marks made by screwdrivers. However, this

algorithm has yet to be applied to quasi-striated marks such as those made by the shear edge of slip-

joint pliers. The results of this algorithm’s application to the surface of copper wire will be presented.

Objective mark comparison also extends to comparison of toolmarks made by firearms. In an

effort to create objective comparisons, microstamping of firing pins and breech faces has been

introduced. This process involves placing unique alphanumeric identifiers surrounded by a radial

code on the surface of firing pins, which transfer to the cartridge’s primer upon firing. Three different

guns equipped with microstamped firing pins were used to fire 3000 cartridges. These cartridges are

evaluated based on the clarity of their alphanumeric transfers and the clarity of the radial code

surrounding the alphanumerics.

 1

CHAPTER 1. BACKGROUND

A Brief History of Toolmarks

The history of toolmarks and firearms stretches back nearly 180 years to the first documented

case of firearms identification in 1835 [1]. Early firearms identification relied primarily on the

identification of the caliber, any macroscopic imperfections of the bullet, and the shape and type of

bullet used in the crime [2]. The first recognized case of this occurred in the City of London, England

in 1835. A homeowner was shot and killed, with the servant as the suspected killer. Henry Goddard, a

part of the police force at the time, investigated the case and was able to identify the mold mark on

the fired lead ball in addition to identifying the paper patch used in firing the black powder weapon.

From these clues, Goddard was able to deduce the guilty party and bring him to justice [1].

One of the earliest cases of firearms identification in the United States occurred during the

Civil War in 1863. Confederate General Stonewall Jackson was fatally wounded in battle and the

bullet that killed him was used to identify the type of firearm used. It was determined the bullet could

have only been fired by one of his Confederate soldiers. Union forces at that time were known to use

a 58 caliber ball, while the bullet that finished Jackson was a 67 caliber ball; the same caliber used by

Confederate forces. Similarly, a year later in 1864, Union General John Sedgwick was killed in battle

by a single bullet. After his death, it was determined the shape and caliber of the fatal projectile were

in agreement with those used in Confederate sniper rifles [1].

The late 1800s and early 1900s saw an increased interest in firearm identification. This

interest included several court cases within the United States, and promoted research conducted

throughout the U.S. and Europe. Published works included titles such as, “La Deformation Des Balles

de Revolver” (Deformation of Revolver Bullets, 1889), “The Missile and the Weapon” (1900), “Zur

Sachverstandign Beurteilung Von Geschossen” (The Expert Examination of Fired Bullets, 1905)
 2

written by A. Lacassogne of Lyon, France, Dr. Albert Llewellyn Hall of Buffalo, New York and Dr.

R. Kockel of Leipzig, Germany, respectively [1]. Some credit Dr. Kockel with the first use of

striation matching of toolmarks, which occurred around 1900. In his first paper, Kockel identified

knife cuts made in wood through oblique lighting and photography. In a later notable paper, he

described the examination of marks through magnification and measured the relative spacing with

calipers. Additionally, this paper noted the change in geometry of the toolmark with different

attacking angles of the knife blade [2].

In 1915, the State of New York saw a great mishandling of a murder case. Charles Stielow

was accused of shooting and killing his employer and the employer’s housekeeper. After being shot,

the housekeeper ran and was found at Stielow’s door. An alleged firearms examiner was hired to

examine the evidence and determined that the revolver at Stielow’s residence had fired the bullets in

question. As a result, Stielow was sentenced to death. However, upon reexamination of the evidence,

Charles E. Waite of the New York Attorney General’s office and Dr. Max Poser of Bausch & Lomb

were able to determine that Stielow’s revolver was not involved in the crime in question. Stielow was

subsequently pardoned. As a result of this case, Waite, Phillip O. Gravelle, John H. Fisher and Calvin

H. Goddard gathered together to investigate “forensic ballistics.” Consequently, the group adapted the

comparison microscope to firearms identification, a vital tool still used in today’s forensic

laboratories [1, 2].

The next significant court case for firearms and toolmarks was the case of Paul V. Hadley in

1921. In Tucson, Arizona, Hadley accepted a ride from an elderly couple, who he later shot. The

woman later died as a result of her injuries. Upon Hadley’s arrest, a 32 caliber pistol and several

cartridges were found on his person. A practicing attorney, A. J. Eddy, was asked to examine the

bullets from the couple and determine if they were fired by the pistol carried by Hadley. Eddy

performed three months of experiments and concluded that yes, the bullets had come from Hadley’s

pistol. As a result of Eddy’s testimony, Hadley was convicted of the shootings. This ruling was
 3

appealed, only to have the lower court’s ruling upheld, thereby recognizing firearms and ballistics

evidence as admissible in court [1].

The 1930s, 1940s and 1950s saw continued growth of forensic toolmark and firearms

analysis. By 1930 the Scientific Crime Detection Laboratory was operational at Northwestern

University in Chicago, soon followed by the Federal Bureau of Identification Laboratory in 1932.

Other crime laboratories popped up across the country to assist police forces in firearms and toolmark

identification [1].

Most early studies and cases largely focused on ballistic toolmarks, with the exception of a

few studies including Dr. Kockel’s work as previously described. In 1948, Dr. Thomas of the

University of Ghent added to the toolmark references by publishing a paper describing the toolmarks

left on a skull by an axe. Since then, many different types of toolmarks have been characterized [2].

The Association of Firearm and Tool Mark Examiners (AFTE) was formed in 1969. Its

original members were comprised of specialists from the United States and Canada. AFTE has since

become an essential resource for firearm and toolmark examiners throughout the United States and

abroad by providing training, access to journal articles and other resources. In 1980 the AFTE

Glossary was published- complete with definitions, illustrations, formulas for bullet energies and

various chemical formulas as a reference for examiners. Since then, the Glossary continues to be

updated as the organization sees fit [1].

Use of Technology for Toolmark Examination

In 1958 John E. Davis wrote the book, “An Introduction to Tool Marks, Firearms and the

Striagraph.” In his book, Davis introduces the striagraph, a specialized instrument he describes as,

“primarily a measuring, tracing and recording device suited to the analysis of micro surface-contours,

that is, to the detection of microscopic irregularities in surface smoothness” [1, p.276]. Davis’s
 4

methods presented a new way of objectively comparing toolmarks since the contours of a mark could

be quantified. Unfortunately, Davis’s work was largely ignored and the striagraph was considered

primarily a research curiosity [3]. Arguably, this is the predecessor to more modern technology for

recording the surfaces of bullets and toolmarks, such as laser and digital imaging used today.

Technology has greatly advanced in the past twenty to thirty years and this advancement has

significantly aided the toolmark examiner. In 1999 the Bureau of Alcohol, Tobacco, Firearms and

Explosives implemented the National Integrated Ballistic Information Network (NIBIN). This

network enables law enforcement agencies to use 2-D digital imaging to acquire and compare the

ballistics markings on bullets and cartridge cases already recorded nationwide from over 200 different

sites [4]. This database utilizes the automated integrated ballistic imaging system (IBIS) to acquire

digital images of markings on fired ammunition from crime scenes and compares these marks with

those already registered in the database. Since NIBIN has been implemented, over 1.2 million pieces

of evidence have been entered and over 47,000 hits have been recorded, greatly assisting forensic

examiners with identification of ballistics evidence [4, 5].

The past nearly 180 years of toolmark identification has yielded a great body of research and

reference works. Many are still as useful and relevant today as they were when first published. The

basic assumption behind these works- each tool makes its own unique mark- has not changed. With

the advancement of technology in recent decades, more research substantiating this idea continues to

be published. Toolmark examination has remained essentially unchanged in the last half century and

the ideas behind it will be discussed in the following section.

Tools and Their Marks

A tool, as defined by the Association of Firearms and Tool Mark Examiners is, “An object

used to gain mechanical advantage. Also thought of as the harder of two objects which when brought
 5

into contact with each other, results in the softer one being marked” [6, p.176]. Tools can be thought

of as typical instruments such as a screwdriver, hammer, pry bar, drill bit, punch, or possibly

something else, such as a car bumper or a rock. From this definition, tools can be a wide variety of

objects and create any number of different marks, though each mark is thought to be unique to the

tool that made it.

Tools and their marks are significantly affected by the manufacturing processes used in

production. For this reason, examiners are expected to be familiar with various manufacturing

processes such as forging, casting, machining, extrusion, etc. Some manufacturing processes produce

marks that evolve over time as a result of the cutting tool in contact with the workpiece, such as

broaching or machining. The wear on the cutting tool’s surface is well documented [7, 8], especially

in a machining operation. A built up edge can occur during machining, especially if a ductile material

is cut slowly. This built up edge occurs when the material begins to cold weld onto the cutting edge of

the tool due to the high pressures associated with the process. As the material builds up, it breaks off

and new material begins to replace the previous built up edge. This ever changing cutting edge is

reflected in the finished surface of the work piece. In the case of tools, this subtle change in the

cutting surface has been used to examine sequentially manufactured tools and their marks [7, 8].

In addition to manufacturing, toolmarks are also affected by circumstances and the

environment after the tools’ production such as tool wear or corrosion. The working surface of a new

tool will change rapidly during its initial use until the “break-in period” is over. Wear rate then slows

and becomes more uniform. Furthermore, tool misuse and abuse will result in a more unique working

surface and therefore a more unique toolmark. This change in working surface due to wear can

greatly assist an examiner when determining the tool used in a crime [8].
 6

Toolmark Characteristics

Different types of tools leave different types of marks. The two main types of marks are

impressed marks, such as a hammer strike, and striated marks, such as those made as a screwdriver

slides across a softer surface [6]. These two types of marks can be found together at times. For

example, firearms produce impressed marks on the cartridge primer when the firing pin strikes it and

striated marks on a bullet as it passes through the rifled barrel.

In addition to the broad classifications of impressed and striated marks, toolmarks are also

characterized by individual, class and subclass characteristics. Individual characteristics are random

imperfections, which are produced during manufacture or caused by use, corrosion, or damage [6].

Individual characteristics are what make a tool unique amongst other tools of its type and are

produced by accident. An example of an individual characteristic might be a screwdriver that has a

chip missing from one edge of the blade. Marks then made from this edge will have an individual

characteristic that separates them from other screwdriver marks.

Class characteristics are features determined prior to manufacture; this includes size and

shape of the tool. Examples of these characteristics might be the caliber of a firearm or size of a

hammer’s head. Also included in class characteristics is the type of action imparted by the tool:

compression, crimping, shearing, slicing, etc. [6, 8].

Subclass characteristics are somewhat less clear and more elusive than individual or class

characteristics. They can be mistaken for individual characteristics, though trained examiners are able

to distinguish between the two. Subclass characteristics, as defined by AFTE are, “discernible surface

features of an object which are more restrictive than class characteristics in that they are produced

incidental to manufacture; are significant in that they relate to a smaller group source (a subset of the

class to which they belong); can arise from a source which changes over time” [6, p.175]. Examples

of these marks include broaching marks or mold marks on a part from a master pattern. With the case

of broaching marks, due to the contact of the cutting surface with the workpiece during manufacture,
 7

the cutting tool is constantly undergoing change due to abrasion and built up edge, as previously

discussed. This changing cutting surface is reflected in the workpiece after the broaching operation is

complete [8].

Toolmark Comparison Techniques

The objective of a toolmark examination is to determine if a suspect tool made an evidence

mark. Generally speaking, the comparison of tools and their marks can be classified into two general

categories: pattern fit and pattern transfer. Pattern fit is easily understood and can be likened to a

puzzle piece fitting into the missing part of a jigsaw puzzle. Each piece of the puzzle is assumed to be

unique and fit only in its designated place. Pattern fit can also be described as a physical match or a

fracture match. The more contours the fractured surfaces possess, the higher the likelihood of a true

match. For example, if a ceramic mug is accidentally dropped and breaks, the shards can be pieced

back together. On the other hand, pattern transfer is not quite as straightforward because it involves

impressions and striations of two and three dimensional marks. Consider a screwdriver blade sliding

across a lead surface. The blade of the screwdriver has its own contours and when the action is

performed with adequate force, it will leave a striated mark with the transferred pattern of the blade

on the lead. Hammer impressions, striated chisel marks and the firing pin impression upon a cartridge

primer can all be grouped into the pattern transfer category [9].

Because pattern transfer encompasses such a wide variety of markings, it is the primary focus

of discussion during the examiner’s work and training .When a suspect tool and evidence mark are

submitted for evaluation, the examiner will study the tool and mark in question to determine if the

mark was made by a tool with class characteristics similar to the tool submitted. If so, then the

examination continues and the tool and evidence mark are evaluated for any trace evidence such as

paint or metal transfer. Test toolmarks are made with the suspect tool with the intent to recreate the
 8

evidence mark as closely as possible. This includes accounting for the angle of tilt and angle of

progression used to create the original mark. The test marks are then examined with a comparison

microscope to see if in fact an identification exists between the tool in question and the given

evidence mark. An identification is determined when ‘sufficient agreement’ exists between the test

toolmarks and the evidence mark [8]. The definition of ‘sufficient agreement’ will be further

discussed in the next section. Once a toolmark identification is made, four different statements are

expected to be true: 1) the suspect tool was used to make the evidence mark, 2) the tool’s working

surface has not been significantly damaged since making the evidence mark, 3) the evidence mark has

sufficient unique features for comparison, and 4) the tool’s working surface has an individual surface

finish [9].

Theory of Toolmark Identification

When evidence marks are submitted for examination, a toolmark examiner is presented with

four possible conclusions when evaluating the marks: identification, inconclusive, elimination, or

unsuitable. These four categories are fairly self-explanatory. Examiners often err on the side of

caution and only accept identification when there is overwhelming support for this conclusion.

The Association of Firearms and Toolmark Examiners has accepted a non-quantitative

position on the theory of identification of toolmarks. AFTE’s theory simply states, “The theory of

identification as it pertains to the comparison of toolmarks enables opinions of common origin to be

made when the unique surface contours of two toolmarks are in “sufficient agreement”” [6, p. 175].

This qualification of sufficient agreement is somewhat vague; however AFTE does provide

clarification for this term. Agreement between marks is significant when it “exceeds the best

agreement demonstrated between toolmarks known to have been produced by different tools and is

consistent with agreement demonstrated by toolmarks known to have been produced by the same
 9

tool” [6, p. 175]. By concluding two marks have sufficient agreement, examiners acknowledge that

the likelihood of another tool making these marks is so remote as to be considered a practical

impossibility [6]. Because toolmark examinations and, ultimately, the conclusion of “sufficient

agreement” are subjective in nature, this method has received considerable criticism [10, 11]. While

AFTE acknowledges this subjectivity, it does state that the interpretation of identification is founded

on scientific principle and, in the end, is based on an examiner’s training and experience [6].

Consecutive Matching Striae

The traditional method of identifying toolmarks is pattern matching. However, in an attempt

to quantify a “match,” a method of counting the consecutively matching striae (CMS) has been

suggested. Both pattern matching and CMS employ the same science and techniques, but differ in the

manner in which they describe their results. Nichols acknowledges this by saying, “There is no

difference between a “pattern matcher” and a “line counter” except the manner in which they

document their casework and articulate their conclusions” [12, p. 300]. The CMS method will

describe the best non-match observed and from that experience an examiner can use this information

to determine an identification. Nichols argues, especially in court CMS, appears to hold up better than

the traditional pattern matching, since the CMS method is better able to articulate the reasoning for an

identification in a way that a lay person might understand the result.

The Daubert Criteria

In 1993, the case of Daubert v. Merrell Dow Pharmaceuticals, Inc. changed the admissibility

standards of expert testimony. Previously, the case of Frye in 1923 had been the accepted standard.

Under Frye, the only criterion set forth for the admissibility of expert testimony was that the opinions
 10

expressed by the expert had general acceptance in the field in which it belongs. By being accepted by

its respective field, the testimony was believed to have been thoroughly tested and thus be valid in

court. The Frye test held until 1993 when Daubert sought to define in more specific terms the Frye

principles associated with the description of being “thoroughly tested” by outlining a set of criteria for

expert witness testimony. Under Daubert, four different criteria must now be met: 1) testability of

scientific principle, 2) known or potential error rate, 3) peer review and publication, and 4) general

acceptance in a particular scientific community. Through these criteria, Daubert has essentially

placed the presiding judges into gatekeeper positions, leaving them to decide what is admissible and

what is not. While Daubert is now the controlling standard for all federal cases, not all states have

adopted it and many still use Frye or some modification of it when evaluating admissibility [13].

Toolmark examination can and does meet the criteria set forth by Daubert, but many

attorneys have sought to have the examiner’s testimony omitted from cases claiming the examinations

are not rooted solidly in science or that the examiner’s conclusions are subjective and cannot be

trusted [13, 14]. A scientific foundation and objectivity are found in any experienced toolmark

examiner’s toolmark comparisons. In recent years, to reinforce these ideas, different groups have

sought to make objective toolmark comparisons with the use of comparative statistical algorithms.

Research Related to Toolmarks

Toolmark research, as it relates to tools, clearly has a long history extending to the turn of the

twentieth century with the publication of Dr. Kockel’s papers. In 1942 a notable paper was published

by Burd and Kirk examining the marks made by screwdrivers. In this study [15] the authors addressed

four different points: 1) the effect of varying the angle of application of the screwdriver on a

toolmark, 2) establishing the necessary criteria for identification, 3) assessing the similarity between

tools with identical appearance and manufacturing process, and 4) classifying the different types of
 11

marks that can be encountered. Burd and Kirk pointed out in the study the traditional method of

examining toolmarks with oblique lighting and a comparison microscope will only yield a match if,

and only if, the marks in question have a similar contour, since this is reflected in the “lines” or

striations seen through the microscope. The authors go on to conclude several important points. First,

two marks made with the same tool must be made with a difference in vertical angle of no more than

15 degrees if a match is to be obtained. Similarly, two marks made with the same tool must be made

with a difference in horizontal angle of no more than 20 degrees if a match is to be determined. The

authors also established the maximum percentage of lines that matched in non-match comparisons did

not exceed 25% and when match comparisons were performed this percentage jumped to around

80%. Additionally, examination of “identical” tools produced noticeably unique marks that could not

be matched to another “identical” tool. This paper is very well written and remains valuable and

relevant today.

As summarized by Nichols [16], many other papers have been published by various authors

concerning toolmarks made by other tools since Burd and Kirk’s study. Several studies have since

been published concerning toolmarks made by screwdrivers, as they make the quintessential striated

mark that is easily examined. Nichols specifically mentions those published by Burd and Gilmore

[17] and by Vandiver [18]. Other significant studies concerning knives [19, 20], bolt cutters [21-23],

drill bits [24], rotary glass cutters [25] and cast bullets [26] all reach the same conclusion, namely,

each tool makes its own unique mark.

Tongue and groove pliers were evaluated in 1980 by Cassidy [27]. These pliers are often used

to pry open door handles and their marks are simple striated marks stemming from a plier tooth

sliding across a surface gripped in the pliers’ jaws. For this study Cassidy procured three sets of upper

and lower jaws that were sequentially broached with no further manufacturing processes applied to

preserve any subclass characteristics present from the broaching process. He observed no subclass

characteristics that might be mistaken for individual characteristics. In the study’s discussion, Cassidy
 12

demonstrates that the pliers’ teeth were broached perpendicular to the direction that the marks are

made and would not produce any subclass characteristics in the striated marks. Furthermore, actual

tongue and groove pliers in production go through many processes after broaching; thus, marks

produced by these mass production pliers would produce only marks that have individual

characteristics.

With the availability of inexpensive computing power and increasingly precise metrology

instruments, toolmarks are being reexamined through objective statistical comparison of their 3-D

profiles. In 2007 Faden et al. [28] developed a computer algorithm to compare and match surface data

taken from a stylus profilometer. In the study, 44 sequentially manufactured screwdriver tips were

used to create marks at 30, 60 and 85 degrees from both sides of the screwdriver blade and the

profilometer used to record the surface contours of the mark through 9600 data points. A computer

program was then used to compare the collected profilometer traces. Three different comparison data

sets were generated: 1) true matches, 2) true non-matches, and 3) comparisons between side A and

side B of the screwdriver blades. The Pearson correlation was calculated for all comparisons. Faden et

al. determined that while there is a significant separation in the correlation values between true match

and true non-match marks at the same angle, the Pearson correlation is not effective at determining

when an actual match exists. Moreover, marks made from different sides of the same screwdriver tip

produced a separation of data consistent with that of non-matches.

In 2010 Bachrach et al. [29] expanded the research of statistical comparison of toolmarks by

evaluating screwdriver marks, and tongue and groove plier marks through confocal microscopy. In

this study, Bachrach et al. examined marks made by screwdrivers at different angles in lead and

aluminum. In addition, they examined the marks from tongue and groove plier marks in lead, brass

and galvanized steel. After scanning the marks with a confocal microscope, the mark data were

normalized to level the data, and then put through a signature generation process. This process took

the cross sectional profile of the mark and applied a Gaussian band pass filter to eliminate class
 13

characteristics within the mark. Then, two signatures were run through a correlation component to

evaluate the two signatures’ similarity to each other. From this study, several conclusions were

drawn. First, striated toolmarks in the same medium and produced under the same conditions are

repeatable and sufficiently specific to allow identification. Second, striated toolmarks created with the

same conditions, but different media, have a high reproducibility. Third, screwdriver marks depend

on the angle at which they are made more than the media in which they are created. Fourth, the

probability of two tools displaying similar features is extremely small. Finally, the probability of error

originated from a poor toolmark image, not from the tool’s failure to create an individual toolmark.

Chumbley et al. [30] continued with the work performed by Faden et al. in 2010. In this

study, a statistical algorithm was used to evaluate its effectiveness in comparison to actual toolmark

examiners. Again, data were collected by a stylus profilometer for 50 sequentially manufactured

screwdriver tips. Marks were made at 30, 60, and 85 degrees for both sides of the screwdriver tip, A

and B. The mark profiles collected were then analyzed by a statistical algorithm. These calculated

results were then compared to a double blind study where 50 experienced toolmark examiners

evaluated a given sample set with which the algorithm had difficulty. The results from this study

showed that while the objective algorithm was very effective in discriminating between known

matches and known non-matches, it still did not reach the level of performance of experienced

examiners.

Objective statistical comparison continued through research done by Petraco et al. in 2012

[31]. In research supported by the U.S. Department of Justice, Petraco evaluated striated marks from

screwdrivers and chisels, as well as striated and compressed marks from cartridge cases. Like

Bachrach et al., Petraco et al. also used confocal microscopy when collecting the surface profiles of

the sample marks. The results of this study showed chisel marks were patchy at best and proved too

complicated for the developed software to analyze successfully. Screwdriver and cartridge cases had

much more success in comparisons and had very low error rates. With the successes and the
 14

difficulties associated with this current software, Petraco et al. have made their marks and software

open sourced and accessible to others in the forensic community.

Statistical Algorithm for Toolmark Analysis

Since the algorithm developed by Faden et. al. and Chumbley et. al. forms the basis for part

of the analysis conducted in this thesis, it is suitable to provide a brief description. The algorithm

used divides analysis into two distinct operations: Optimization and Validation. In the first step,

Optimization, the algorithm seeks to identify a region of best agreement between the two chosen

datasets for a user defined window size, the red boxes in Figure 1. This is achieved by calculating the

maximum correlation statistic, described in an earlier paper [28]; this is also referred to as the “R-

value.” Values very close to 1 indicate regions of the datasets which are very similar. As previously

indicated [30], Optimization is not the best tool to use when determining matches and non-matches.

For this reason, the algorithm employs the comparison process called Validation after the

Optimization process. In this step, a series of windows of are randomly chosen and shifted common

distances from the regions defined as areas of best fit, the purple boxes in Figure 1a. The R-value for

these areas is then calculated and many other rigid window shifts are performed. The idea behind the

random rigid window shift is if a true match exists, there is reason to believe that many rigid shift

window comparisons will yield larger R-values. Conversely, if a true match does not exist there is no

reason to believe that rigid shift comparisons over the length of the mark will yield large R-values. It

should be noted the R-values in this step are assumed to be lower than the one defined by the

Optimization step, as that was the highest R-value for the two datasets. The next step in Validation

calculates R-values at random locations along the length of the datasets, Figure 2. The random

window shifts are expected to have lower R-values than the rigid window shifts. In the case of a non-

match, the rigid window shifts and the random window shifts in the Validation step will have similar
 15

low R-values. To conclude the algorithm’s process, a Mann-Whitney U-statistic is computed to join

all of the comparisons together. This statistic is referred to as a T1 value. T1 values close to zero

indicate the case of a non-match, while higher T1 values indicate the case of a match.

a b

Fi

Figure 1: a) Line profiles of known match comparisons. b) Line profiles of known non-match
comparisons.

Figure 2: Random shift windows, green boxes. Red boxes indicate region of best agreement
found during the Optimization step.

This algorithm is subject to current and future work involving the comparison of striated and

quasi-striated marks such as shear marks from slip joint pliers, which will be discussed later in this

thesis.
 16

Research Related to Firearms

In 1959 Biasotti published an extensive statistical evaluation of the individuality of bullets

fired from different firearms [32]. Thus far, this remains the most exhaustive statistical empirical

study ever published for firearms examination. Biasotti used twenty four .38 Special Smith and

Wesson revolvers: 16 used and eight new guns. He compared the land and groove impressions from

bullets fired from the study’s guns and in doing so created a new way to describe striated markings,

consecutiveness. This arguably was the beginning of the CMS method as a way to describe striae.

Numerous studies have been done relating to firearms in the last century of forensics work.

Nichols [16] provides excellent summaries of many of the more notable studies for bullets, gun

barrels, and cartridge cases. Included in his review are several studies examining the marks imparted

on bullets from rifled barrels, a study on the individuality of button rifled barrels, a study examining

sequentially manufactured firing pins, and several studies concerning breech face markings.

Traditionally, all of these factors-bullets, gun barrels, cartridge cases, and firing pins- are evaluated

by an examiner when presented with firearm evidence. However, the numerous studies concerning

the individuality of each gun and its respective working surfaces have not deterred attorneys from

attempting to throw out ballistics evidence [14].

In an effort to eliminate the arguments of subjectivity in ballistics examination, the idea of

microstamping was conceived by Todd Lizotte and Orest Ohar in 1994 [33]. Microstamping involves

placing unique, identifiable characters on the end of a firing pin or breech face of a gun. When fired,

the microstamp impresses the unique identifiers onto the fired cartridge case with the intention of

making identification of the firearm relatively straightforward; something an officer on a crime scene

could identify with a hand lens. The creators’ other intent with the introduction of this technology was

to make it possible to track patterns of gun crimes. If cartridge cases are left at the scene of several

crimes that trace back to the same gun, a pattern can emerge which could be helpful for law

enforcement [34].
 17

Current microstamped marks have six to eight alphanumeric characters surrounded by a gear

code placed on the end of a firing pin, Figure 3. The alphanumeric marks are intended to act similar to

the way a license plate acts to identify a car where each code can be traced back to a specific firearm.

The gear code surrounding the alphanumerics is intended as a backup, i.e., it provides a way to

identify the cartridge in the event the alphanumeric identifiers cannot be read. It acts somewhat like a

barcode, as it is read in six-bit binary in zeros and ones. The gear code, outlined in Figure 3, is

divided into eight equal sectors of 42 degrees with a starting wedge of 24 degrees at the top of the

mark. These eight sectors, read clockwise, correspond to each alphanumeric identifier as it is read left

to right. The first sector of the gear code corresponds to the first alphanumeric, S, the second sector to

2, etc. Within each sector there are 7 degree increments, which correspond to the aforementioned

ones and zeros. For example, the first sector reads 011001, which corresponds to the letter S. The

entire gear code in Figure 3 can be translated to read S23-SX7-SS.

Figure 3: a) Microstamp of Sig Sauer cartridge, b) microstamp of Sig Sauer cartridge with
overlay

Needless to say, the introduction of this idea and technology has largely polarized lawmakers

and the public. Those critical of guns and gun law have pushed for legislation requiring all guns to

have a microstamped firing pin. On the other side, pro-gun advocates vehemently reject any
 18

requirements to have such regulations applied to guns while manufacturers claim that implementing

such technology will raise costs and force many companies out of business [35].

Despite microstamping being a hot button issue with gun lovers and gun haters, a fair amount

of research has been performed to evaluate the reliability of the transfer and the durability of the

microstamp. Perhaps the most extensive study performed occurred at the University of California,

Davis [36]. This study was very extensive and encompassed many different aspects of microstamping

including the durability and longevity of characters, their legibility, obliteration, the costs associated

with implementation along with extensive appendices containing all the data. This study is extensive

and several key points should be considered. First, the quality of transfer is heavily dependent on the

firearm, and can also be affected by the ammunition used. Some flattening and degradation of the

alphanumerics were seen throughout the study. Additionally, the radial bar code structure showed

severe wear, though it should be noted that since this study the gear code structure has been revised to

produce a more discernible mark [34]. Finally, destroying the microstamp was very easily

accomplished when the firing pin was removed from the weapon.

It remains to be seen if microstamping will gain widespread legislation throughout the United

States. However, further research into the durability and transfer of microstamping is necessary to

come to definitive conclusions before legislation is passed requiring this technology. For this reason,

an examination of microstamping is also undertaken in this thesis and the results are discussed in later

chapters.

References

1. Hamby, J.E., and Thorpe, J.W., “The History of Firearm and Toolmark Identification,” AFTE

Journal, Vol. 31, No. 3, 1999, pp. 266-283.

2. Meyers, C.R., “Firearms and Toolmark Identification: An Introduction,” AFTE Journal, Vol.

25, No. 4, Oct. 1993, pp. 281-285.

 19

3. Springer, E.S., “Toolmark Examinations-A Review of Its Development in the Literature,”

Journal of Forensic Sciences, Vol. 40, No. 6, Nov. 1995, pp. 964-968.

4. “ATF’s NIBIN Program,” April 2008, <http://www.nibin.gov/nibin.pdf>. Accessed 2013 Jan.

17.

5. “National Intetrated Ballistic Information Network (NIBIN),” March 2012,

<http://www.atf.gov/publications/factsheets/factsheet-nibin.html>. Accessed 2013 Jan. 17.

6. Association of Firearm & Tool Mark Examiners Glossary, fifth edition, 2007.

7. Monturo, C., “The Effect of the Machining Process as it Relates to Toolmarks on Surfaces,”

AFTE Journal, Vol. 42, No. 3, 2010, pp. 264-266.

8. Miller, J., “An Introduction to the Forensic Examination of Toolmarks,” AFTE Journal, Vol.

33, No. 3, 2001, pp. 233-248.

9. Biasotti, A.A., Murdock, J., Firearms and Toolmark Identification. In D. L. Faigman, D. K.

Kay, M.J. Sacks & J. Sanders, (Eds.), Modern Scientific Evidence: The Law and Science of

Expert Testimony. Volume 2. St. Paul: West, 1997.

10. Regan, B., “Reasonable Doubt,” Popular Mechanics, Aug. 2009.

11. Goff, L., “Quick Study: Crime Scene Science”, Reader’s Digest, Feb. 2011.

12. Nichols, R.G, “Consecutive Matching Striations (CMS): Its Definition, Study and

Application in the Discipline of Firearms and Tool Mark Identification,” AFTE Journal, Vol.

35, No. 3, 2003, pp. 298-306.

13. Gryzbowski, R.A., and Murdock, J.E., “Firearm and Toolmark Identification-Meeting the

Daubert Challenge,” AFTE Journal, Vol. 30, No. 1, 1998, pp. 3-14.

14. United States v. Darnell Anderson, Joseph Jenkins, Edward Warren, James Bates, Obbie

English, Superior Court of the District of Columbia, (2010).

15. Burd, D.Q., and Kirk, P.L., “Toolmarks: Factors Involved in Their Comparison and Use as

Evidence,” Journal of Criminal Law and Criminology, Vol. 32, No. 6, 1942, pp. 679-686.
 20

16. Nichols, R.G, “Firearm and Toolmark Identification Criteria: A Review of the Literature,”

Journal of Forensic Sciences, Vol. 42, No. 3, 1997, pp. 466-474.

17. Burd, D., Kirk, P., “Individual and class characteristics of tools,” Journal of Forensic

Sciences, Vol. 13, No. 3, July 1968, pp.390-396.

18. Vandiver, J., “New screwdrivers production and identification,” AFTE Journal, Vol. 8, No. 1,

Mar 1976, pp. 29-52.

19. Watson, D., “The identification of toolmarks produced from consecutively manufactured

knife blades in soft plastics,” AFTE Journal Vol. 10, No. 3, July 1978, pp. 43-45.

20. Tuira, Sgt., “Tire stabbing with consecutively manufactured knives,” AFTE Journal, Vol. 14,

1982, pp. 50-52.

21. Butcher, S. and Pugh, D., “A study of marks made by bolt cutters,” Journal of the Forensic

Science Society, Vol. 15, No. 2, 1975, pp. 115-126.

22. Hornsby, B., “MCC bolt cutters,” AFTE Journal, Vol. 21, No. 3, 1989, pp. 508.

23. Hall, J., “Consecutive cuts by bolt cutters and their effect on identification,” AFTE Journal,

Vol. 24, No. 3, 1992, pp. 260-272.

24. Reitz, J., “An unusual toolmark identification case,” AFTE Journal, 1975, Vol. 7, No. 3,

1975, pp.40-43.

25. Warren, G., “Glass cutter impression identification,” AFTE Journal, Vol. 23, No. 4, 1991,

pp.925-927.

26. Haag, L., “Matching Cast Bullets to the Mould that Made Them and Comparisons of

Consecutively Manufactured Bullet Moulds,” AFTE Journal, Vol. 39, No. 4, 2007, pp. 313-

322.

27. Cassidy, F.H., “Examination of Toolmarks from Sequentially Manufactured Tongue-and-

Groove Pliers,” Journal of Forensic Sciences, Vol. 25, No. 4, Oct. 1980, pp. 796-809.
 21

28. Faden, D., Kidd, J., Craft, J., Chumbley, L.S., Morris, M., Genalo, L., Kreiser, J., and Davis,

S., “Statistical Confirmation of Empirical Observations Concerning Toolmark Striae,” AFTE

Journal, Vol. 39, No. 3, 2007, pp. 205-214.

29. Bachrach, B., Jain, A., Jung, S., and Koons, R.D, “A Statistical Validation of the

Individuality and Repeatablility of Striated Tool Marks: Screwdrivers and Tongue and

Groove Pliers,” Journal of Forensic Sciences, Vol. 55, No. 2, Mar. 2010, pp. 348-357.

30. Chumbley, L.S., Morris, M.D., Kreiser, J., Fisher, C., Craft, J. Genalo, L.J., Davis, S., Faden,

D., and Kidd, J., “Validation of Tool Mark Comparisons Obtained Using a Quantitative,

Comparative, Statistical Algorithm,” Journal of Forensic Sciences, Vol. 55, No. 4, Jul. 2010,

pp. 953-961.

31. Petraco, N. et al., “Application of Machine Learning to Toolmarks: Statistically Based

Methods for Impression Pattern Comparisons,” Document 239048, NCRJS, July 2012.

32. Biasotti, A.A., “A Statistical Study of the Individual Characteristics of Fired Bullets,” Journal

of Forensic Sciences, Vol. 4, No. 1, Jan. 1959, pp.34-50.

33. Lizotte, T., “Advanced Intentional Firearm Microstamping,” Presented at Anti-Weapon

Trafficking Training, Mexico City, Mexico, 17 Oct 2011.

34. Ohar, O.P., Lizotte, T. E., “Extracting Ballistic Forensic Intelligence: Microstamped Firearms

Deliver Data for Illegal Firearm Traffic Mapping – Technology,” Proc. Of SPIE Vol. 7434,

743416, (2009).

35. Shapiro, O., “Empire state: Kills CoBIS; threatens microstamping (again),” Shooting Sports

Retailer, Vol. 30, No. 3, May/Jun. 2012, pp. 18.

36. Howitt, D., Tulleners, F. A., Beddow, M. T., “What Micro Serialized Firing Pins Can Add To

Firearms Identification in Forensic Science: How Viable are Micro-Marked Firing Pin

Impressions as Evidence?,” UC Davis/ California Policy Research Center.

< http://forensicscience.ucdavis.edu/pdf/microserial.pdf>.
 22

CHAPTER 2. OBJECTIVE COMPARISON OF MARKS FROM SLIP-

JOINT PLIERS
A paper to be submitted to the Association of Firearm and Tool Mark Examiners Journal

T. Grieve1, L. S. Chumbley1, J. Kreiser2, M.Morris1, L. Ekstrand1, S. Zhang1

1
Ames Laboratory and Iowa State University, Ames IA 50011
2
Illinois State Police, Retired, 3112 Sequoia Dr., Springfield, IL 62712

Introduction

In the last twenty years, several different court cases, including perhaps the most well-known,

Daubert v. Merrell Dow Pharmaeuticals, Inc., have called into question the validity of scientific

testimony, especially as it relates to firearm and toolmark examination. As a result, recent research

has sought to justify a basic assumption made by forensic examiners: each tool makes its own unique

mark. Many different tools and their marks have been examined in the research setting including

screwdrivers [1-4], tongue and groove pliers [4, 5], and chisels [3].

Striated screwdriver marks have been well studied and characterized by stylus profilometry

and confocal microscopy. These characterizations have been used to analyze potential matches and

non-matches via statistical validation in several different studies [1-4]. In general the results have

shown that striated marks can be compared objectively using computer algorithms with a fairly high

success rate. Studies of somewhat irregular marks also exist, although to a lesser extent. Cassidy

first published a study on the examination of toolmarks from sequentially manufactured tongue and

groove pliers, as they are frequently used to twist off doorknobs to break into buildings [5]. This

study, while not based in statistical validation, did establish that the tongue and groove pliers only

produce individual characteristics due to the teeth being broached perpendicular to the direction of the

striated mark. Bachrach et al. more recently examined the marks produced by the application of

tongue and groove pliers to different materials (lead, brass and galvanized steel) and used statistical
 23

comparisons to objectively compare the marks [4]. Bachrach et al. found the tongue and groove pliers

marks could readily be compared when made on the same media. However, the empirical error rate

increased when comparing marks made on different media. Chisel marks have been evaluated by

Petraco et al. [3], but the patchy striated chisel marks used in this research proved too difficult for the

developed suite of software currently in use to provide useful information during comparison. Thus,

while a small body of work exists on less than perfectly striated marks, the results are somewhat

disappointing at this time.

In a previous study [2], fifty sequentially manufactured screwdriver tips and their marks

made at different angles were examined and compared though a statistical algorithm to determine the

strength of evidence of a positive match between a mark and the tool that made it. This algorithm has

been used extensively to evaluate the evenly striated marks of screwdrivers, however it has not yet

been used to evaluate less striated marks or impression marks. As a first step toward investigating the

applicability of the current algorithm, quasi-striated marks such as those made by slip joint pliers

when cutting wire were examined. Slip joint pliers were chosen since no studies currently exist on

this subject to the authors’ knowledge. Additionally, they were expected to produce a more difficult

mark for analysis, due to the manner in which cutting occurs. When cutting a wire with slip joint

pliers, the mark produced reflects both striations from the actual cutting and smearing, due to shearing

of the material during the process. This results in a mark that is not continuous from the beginning of

the cut to the end. Thus, the surface topography that exists at the initial cut edge of the mark could

vary substantially from what is seen at the final cut edge.

Experimental

For this experiment, 50 pairs of sequentially manufactured slip joint pliers were purchased

from Wilde Tool Co., Inc. so as to be as nearly identical as possible. It is well known the
 24

manufacturing process greatly affects the resulting toolmarks a tool mak

makes
es due to the surface features

imparted on the tool during manufacturing [6, 7]. For this reason, a detailed description of the way the

pliers used in this study were manufactured is in order.

All of the plier-half blanks examined in this study were hot forged from the same die,

followed by cold forging from the same forging die. Following forging, holes were punched to seat

the fastener, i.e. the bolt that will hold the two halves of the pliers together. At this point a difference

is introduced in the blanks. On slip joint pliers, one half of the pliers has a small hole, while the other

half has a larger, double hole allowing the user to gain a better grip when using the pliers (see Figure

1). Once the plier holes were punched the teeth and shear cutting surfaces were created using a

broaching process. It is this machining method that creates the scratch minutiae on the surface of the

plier halves responsible for producing the characteristic toolmark that is of interest in forensic

examinations.

Figure 1: Slip joint pliers in their unfinished and finished states. From left to right: plier halves
(single and double hole) before broaching; an example flat side of pliers that will be polished;
finished and labeled pliers (sides A and B).
 25

The plier halves for this study were cut on two separate broaching machines; halves with the

smaller hole were all broached on one machine, while the halves with the double hole were broached

on a second. At this point in the process the manufacturer stamped numbers 1-50 on each plier half as

they were finished being broached. Thus, the 50 pairs could be assembled with confidence that they

were actually made sequentially. After broaching, both halves were given the same heat treatment

and shot peened to surface harden the metal. The long, flat surface was then polished and the pliers

were assembled and gripped. As a final step the company branded the double hole side of each pair of

pliers. For the purposes of this study each half of the pliers was assigned as either A or B, with Side

B being the branded half of the pliers (see Figure 1).

To make the samples, copper wire of 0.1620” diameter and lead wire of 0.1875” diameter

were obtained and cut into two-inch lengths with bolt cutters to distinguish the ends from the cuts

made by the pliers. Next, the cut lengths of wire were placed centered in the plier jaws on the cutting

surface with pliers side B facing down. Alternating shear cuts of lead and copper were made with

each pair of pliers for a total of 21 cuts. All odd numbered cuts were lead samples; all even numbered

cuts were copper. The total number of copper samples thus obtained was 1000, with 500 cuts in

contact with Side A, 500 cuts with side B.

For the purpose of this study, only the copper samples were evaluated. Each cut mark surface

was scanned optically with an Alicona Infinite Focus G3 profilometer at 10x magnification to acquire

the surface geometry of the mark. An example of a typical scan is shown in Figure 2. The tool mark

is seen to be quasi-striated, i.e. parallel linear striae do exist but it clearly varies across the surface of

the cut mark.

 26

Figure 2: Areas examined during comparisons. Dashed line is referred to as the “short edge,”
the solid line is referred to as the “long edge.”

When the data are acquired, noise spikes occur around the edges of the mark where the cut

surface drops off because there is no surface here for the profilometer to scan. This noise is non-

informative for the matching process, and is not desirable in the data file. Therefore, the raw data are

processed using a computer routine to remove the extraneous noise spikes. This process is referred to

as a cleaning routine and does not affect the data that characterizes the cut surface. An example of a

clean and uncleaned data file can be seen in Figure 3.

a b

Figure 3: a) Raw data; b) cleaned data with noise spikes removed

 27

All raw data files contained trended data. Simply put, due to the manner in which the data

were collected the line profile of a mark data file had an increasing linear trend in the z direction

moving from one side of the mark to the other. Such a trend is common when using profilometers

since the surface analyzed is rarely exactly parallel with the direction of scanning. Because the files

were a rectangular collection of 3D data (shown in the uncleaned data of Figure 3a), trending was

corrected by subtracting a plane matching that of the trended data from the file. To accomplish this,

the detrending routine selects left and right diagonal points from the data (approximately 40 on each

side, 80 in total) and uses a linear least squares method to fit the appropriate plane for the data. It then

subtracts the fitted plane from the data to achieve an appropiately leveled data file for comparison. As

a reference, these final data files are roughly 2200 by 4500 pixels.

Comparisons between the marks were made using the previously described algorithm [2].

The comparisons were divided into two different groups, those made close to the end of the mark, as

designated by the solid line in Figure 2, and those made close to the start of the mark, shown by the

dashed line in Figure 2. From this point on, the dashed line data will be referred to as the short edge

and the solid line data as the long edge. These mark locations were chosen to examine differences

between the beginning of the cut, where the mark has short and variable length striae, and the end of

the mark, where the striae are longer and appear to be more regular.

Each side of the pliers was considered to be a separate data set, the assumption being, as

confirmed by forensic examiners, each side acts as a different surface. Given there are 50 pairs of

pliers, with two sides for each pair of pliers and ten replicate cuts for each side of each pair of pliers,

the total number of samples possible for examination came to 1,000 discrete data sets.
 28

Results

A sampling format was set up to compare three different groups of data: known matches,

known non-matches from the same pair of pliers (i.e. different sides), and known non-matches from

different pairs of pliers. The comparison setups are as follows:

Set 1: Compare known matches. These should be marks from the same side of pliers.

Comparisons were made between marks 2 and 4 and between marks 6 and 8 for each side of the

pliers, side A and side B.

Set 2: Compare known non-matches from the same pair of pliers. Comparisons were made

between side A and side B for marks 10, 12 and 14.

Set 3: Compare known non-matches from different pairs of pliers. The samples were divided

into 12 groups of four, each numbered consecutively, e.g. tools 1-4, 5-8, etc. Comparisons were made

for both side A and side B. Table I shows an example comparison setup for the first group of pliers.

Table I: Comparisons for Set 3, Group 1

Comparison Plier number Side Mark number Plier number Side Mark number
A 1 A 16 2 A 16
B 3 A 16 4 A 16
C 1 A 18 4 A 18
D 2 A 18 3 A 18
E 1 A 20 3 A 20
F 2 A 20 4 A 20

The same algorithm used in an earlier work for striated marks [2] was applied in this study to

examine the quasi-striated marks made by the slip joint pliers. The algorithm has two primary steps:

Optimization and Validation. During the Optimization step, the regions of best agreement between

the two marks are determined by the maximum correlation statistic, or “R-value.” The size of the

region is assigned by the user and is hereafter referred to as the “Search Window.” The second step of
 29

the algorithm, Validation, uses both rigid and random window shifts to verify the regions chosen in

the Optimization step indeed correspond to a true match. These windows are hereafter referred to as

the “Valid Windows” and their width is also user determined. The R-values in this step must clearly

be lower than the R-value in the Optimization step, as the highest R-value has already been

calculated. However, in the instance where a true match exists, the R-values associated with the rigid

shift valid windows should be larger than those associated with the random shift valid windows, the

assumption being, if an excellent match exists at one location then very good matches should exist at

any number of corresponding locations. If true, this is indicative a true match does exist. Conversely,

rigid window shifts do not produce systematically larger R-values than random shifts in the case of a

true non-match, since the high values found during the Optimization step exists due to random chance

rather than any physical relationship between the items being compared. Further discussion of this

algorithm can be found in the literature [2].

Originally, the size of the search and valid windows were set at the comparison software’s

default 200 and 100 pixels, respectively, and the comparisons were conducted with samples from the

first 20 pairs of pliers. This setup produced 400 different comparisons for the long and short edge

comparisons. When a comparison is made, indication of a true match is found when the T1 value of

the statistic returned is relatively high. Little or no relationship between the marks results in T1

values centered near 0.

Results of these early comparisons can be found in Figure 4. In these box plots, the bold line

in the middle of the box represents the median, the lower quartile by the bottom line of the box, and

the upper quartile by the top line of the box. The whiskers are one and a half times the difference

between the upper and lower quartiles. Any outliers outside the whiskers are denoted by dots. In these

plots, known matches are in the comparisons designated Set 1, while Sets 2 and 3 show comparisons

between known non-matches from different sides of a pair of pliers and non-matches between

different pairs of pliers, respectively. It is evident that with these window sizes, the success of
 30

identifying known matches was relatively low, there being little separation between the returned T1

values of known matches and non-matches.

a b

Figure 4: Original data comparisons for (a) short edge, (b) long edge.

From the minimal success of the first attempt at matching the plier marks, several changes

were decided upon for further comparisons. First, the data shown in Figure 4 compared trended data.

This was corrected in subsequent comparisons. Second, it was decided to vary the window size for

all plier mark samples. The initial values used were chosen simply because they had proven effective

for comparison of fully striated marks. A series of experiments was conducted within each plier

comparison set where the window sizes were varied to evaluate the effect window size has on the

resulting T1 value. In other words, the question asked was: does the size of the window play a large

role in the discrimination between known matches or known non-matches? In this series of

experiments Search and Valid windows were assigned four different values. The Valid window was

always half the size of the Search window. Search windows were set at values 100, 200, 500, and

1000 pixels, respectively, to examine the effects of one smaller Search window and two larger Search

windows. These new settings were extended to all 50 pairs of pliers and their corresponding

toolmarks in the copper wire, bringing the total number of comparisons to 3,952.
 31

The results of these comparisons can be found in Figures 5 and 6. Observation shows that the

T1 value increases dramatically with increasing window size. While known non-matches return

values centered around zero regardless of window size, the T1 value for known matches increases

from just slightly over zero to an average of 6.36 and 6.09 for the largest window size for the long

and short comparisons, respectively. However, the data range increases as well. At the larger window

sizes, numerous outliers exist and failure of the algorithm occurs in some cases, especially for the

short edge comparisons.

 32

a b

Figure 5: Long edge comparisons. a) Known matches from the same set of pliers. b) Known
non-matches from the same set of pliers. c) Known non-matches from different sets of pliers.
 33

a b

Figure 6: Short edge comparisons. a) Set 1: Known matches from the same set of pliers. b) Set
2: Known non-matches from the same pair of pliers. c) Set 3: Known non-matches from
different pairs of pliers.

The large number of observed failures directly results from the constraints placed on the way

the Search and Valid windows are chosen and compared. One of the standard conditions under which

the algorithm operates is the Search and Valid windows are never allowed to overlap. In some cases,
 34

especially with the short edge comparisons, the shorter length of line from which data can be selected

and compared results in far fewer data points for comparison. This problem is exacerbated as the

window sizes increases. For larger sizes, there simply is not enough data available to meet these

conditions in all instances. Thus, this stipulation can cause the algorithm to return no T value.

Table II summarizes the instances in which the algorithm failed to return values. It can be

clearly seen that the return rate decreases with the shorter line profiles as the window size increases.

As a reference, set 1 has a total of 200 comparisons, set 2 has 150 comparisons and set 3 has 144

comparisons.

Table II: Cases in which the algorithm returned no T values for each window size

Long edge comparisons

Set 100-50 200-100 500-250 1000-500
1 0 1 1 1
2 0 1 3 3
3 0 2 3 5
Short edge comparisons
Set 100-50 200-100 500-250 1000-500
1 0 0 1 9
2 1 0 3 19
3 1 0 3 24

As a first attempt at a solution, two additional window ratios were examined: 4 to 1 and 6 to

1. It was hoped that by limiting the size of the Valid windows less spread in the data would be seen.

For each new ratio, four different window sizes were chosen and the algorithm was run again

following sets 1, 2 and 3 at both the long and short edge locations on the mark. For these exploratory

tests the data were limited to pliers 1-25, the assumption being the abbreviated data set would be

representative of the full 1-50 pliers data. Results of this examination can be found in Figures 7 and

8. This set of parameters does indeed appear to have a significant effect in reducing the number of

outliers and spread of the known matches (i.e. Set 1) as compared to the 2:1 ratio data. A slight

degradation in the maximum values obtained was seen for the known matches. Less change is seen
 35

in the results for the known non-matches (Sets 2, 3). Average values still were centered around zero

and spread seemed to increase somewhat in some cases for the known non-matches.
 36

Figure 7: Results of varied ratio long edge comparisons.

37
 38

Figure 8: Results of varied ratio short edge comparisons.

Discussion

When using the developed algorithm, ideally the data should show a clear separation between

T1 values for known matches as opposed to known non-matches, with no overlap occurring, even

when considering outliers. While elimination of overlap in the outliers has not been achieved it is

clear that a high degree of separation is seen in the majority of cases when the search parameters are

adjusted from the defaults used for the striated screwdriver marks. This suggests that the current

algorithm is more robust than it initially appeared, and could be suitable for discrimination if

performance can be enhanced and the spread in the data can be decreased to produce complete

separation between known matches and non-matches. These tests also indicate the size of the Search

and Validation windows can have a critical role in determining when a match can be discriminated

from a non-match. Since the size and number of Valid windows is user defined, future work must

involve a series of experiments to determine what operation parameters are best suited for each
 39

individual class of marks. For example, the relatively small Search and Valid window sizes that

worked well for screwdriver marks were inadequate for the plier marks. However, increasing the

Search and Valid window size proved effective in producing a clear separation between known

matches and non-matches for slip joint pliers and changing the size ratio has an effect on the spread

of the data.

Outliers are seen in all the data sets, both known match and known non-match. Examination

of these data files points to a consistent problem with the current state of the algorithm, which the

authors refer to as the “opposite end” match problem. This seems to be an area where further

improvements can be made. In earlier work involving screwdriver comparisons [2], it was noted the

algorithm often returned false match values, incorrectly identifying the match areas on opposite ends

of the mark’s cross-sectional profile. “Opposite end” matches appear to occur most often in known

non-matches, however non-match values have been returned for known matches as well with similar

opposite end match problems. In detrending the data, many of these problems have been eliminated;

however a few opposite end match problems still exist. One such example can be seen in Figure 9 for

a plier comparison datafile, which consists of detrended data. One data set is shown at the top while

the second is shown at the bottom. Simple chance where the opposite ends of the mark have a very

similar profile over the small area of the search window, as denoted by the box, has resulted in the

computer declaring an excellent match. Obviously, such a match is physically impossible, no matter

how good the numbers.

 40

Figure 9: Incorrect opposite ends match for long edge comparison of known non-matches from
different pairs of pliers. The search and valid windows were 450 and 75. T1 value is 8.137.

In its current form, the algorithm has maximum flexibility, allowing marks to be compared

along a linear direction both forwards and backwards. Such a methodology requires no contextual

information to be known about the mark. A fully striated mark may leave few clues as to what is the

“left” side of the mark vs. the “right” side, as determined by how one holds the screwdriver, Figure

10. As shown by the bold arrows, pulling the screwdriver across the surface in opposite directions

leaves the same mark, but it is rotated 180 degrees. While this situation is usually easily recognized

by a trained examiner making a test mark, it is more of a problem for an automated system. To the

machine, both situations result in a series of parallel lines. If the scan is constrained to run

comparisons in only 1 direction (dotted line), this match may be missed since “left” could be viewed

as “right” and vice versa. For this reason currently the algorithm is written to be as flexible as

possible with comparisons run in both directions so it is not necessary to know which side of the mark

was on the left and which was on the right as it was being made.
 41

Determining the correct scanning direction is less of a problem for a cut wire, where

contextual information such as “left” and “right” can be easily assigned due to the macroscopic shape

of the object itself, Figure 10b. In this instance the situation is somewhat similar to distinguishing

between class characteristics in a firearm examination.

Right

Right
Left
Left Left
Left
Right

Right

a. b.

Figure 10: a) Fully striated marks hold few clues to “left” vs. “right for the automated scan as
denoted by the dashed line. b) Cut wire sample scan directions are easily distinguishable by the
macroscopic shape.

Currently each data file needs to be examined separately in order to determine whether an

“opposite end” match has occurred. A screening option is being considered that will automatically

determine whether an “opposite end” match has occurred and alert the user to this possibility. The

user can then examine only those files so flagged and decide whether an incorrect match has

occurred. Clearly, in this instance the examiner will have to use their contextual knowledge of the

marks being compared to make this determination.

 42

Summary and Conclusions

An objective analysis of 1000 cut copper wire samples produced using 50 sequentially

manufactured pliers was carried out using a previous algorithm to successfully compare striated

marks produced by screwdrivers. Early efforts using the algorithm produced inconclusive results

when using the same parameters used successfully for the screwdriver marks. Further experiments

showed changing the comparison parameters, specifically the sizes of the search and validation

windows, could produce successful identification of known match/non-match comparisons. Future

improvements to the algorithm are planned to screen the identified matched search windows to

eliminate the possibility of clearly incorrect “opposite end” matches.

Acknowledgments

The authors are extremely grateful to Adam Froeschl of Wilde Tool Co., Inc. for making our unusual

request for sequentially manufactured slip-joint pliers possible. This study was supported by the U.S.

Department of Justice, National Institute of Justice, through the Midwest Forensics Research Center

at Ames Laboratory, under Interagency Agreement number 2009-DNR-119. The Ames Laboratory is

operated under contract No. W-7405-Eng-82 by Iowa State University with the U.S. Department of

Energy.

References

1. Faden, D., Kidd, J., Craft, J., Chumbley, L.S., Morris, M., Genalo, L., Kreiser, J., and Davis,

S., "Statistical Confirmation of Empirical Observations Concerning Toolmark Striae," AFTE

Journal, Vol. 39, No. 3, 2007, pp. 205-214.

2. Chumbley, S., Morris, M., Kreiser, J., Fisher, C., Craft, J., Genalo, L., Davis, S., Faden, D.,

and Kidd, J., "Validation of Tool Mark Comparisons Obtained Using a Quantitative,
 43

Comparative, Statistical Algorithm," Journal of Forensic Sciences, Vol. 55, No. 4, 2010, pp.

953-961.

3. Petraco, N. et al. "Application of Machine Learning to Toolmarks: Statistically Based

Methods for Impression Pattern Comparisons," Document 239048, U.S. Department of

Justice, July, 2012.

4. Bachrach, B., Jain, A., Jung, S., and Koons, R., "A Statistical Validation of the Individuality

and Repeatability of Striated Tool Marks: Screwdrivers and Tongue and Groove Pliers,"

Journal of Forensic Sciences, Vol. 55, No. 2, 2010, pp. 348-357.

5. Cassidy, F. H., "Examination of Toolmarks from Sequentially Manufactured Tongue-and-

Groove Pliers," Journal of Forensic Sciences, Vol. 25, No. 4, 1980, pp. 796-809.

6. Miller, J., "An Introduction to the Forensic Examination of Toolmarks," AFTE Journal, Vol.

33, No. 3, 2001, pp. 233-248.

7. Monturo, C., "The Effect of the Machining Process as it Relates to Toolmarks on Surfaces,"

AFTE Journal, Vol. 42, No. 3, 2010, pp. 264-266.

 44

CHAPTER 3. CLARITY OF MICROSTAMPED IDENTIFIERS AS A

FUNCTION OF PRIMER HARDNESS
AND TYPE OF FIREARM ACTION
A paper published in The Association of Firearm and Toolmark Examiners Journal, Volume 44,
Number 2, pp. 145-155

L.S. Chumbley, J. Kreiser*, T. Lizotte†, O. Ohar†, T. Grieve, B. King, D. Eisenmann

Iowa State University, Ames Laboratory

Ames, Iowa

*Illinois State Police, Retired

Springfield, Illinois
†
Pivotal Development, LLC
Manchester, NH

Introduction

In recent years the area of comparative forensic examinations have come under increasing

attack, with various charges being made in popular literature that they are unscientific and highly

subjective in nature [1, 2]. These allegations have arisen due to a combination of controversial court

cases [3], mistakes in fingerprint identification [4], selective use of remarks made in a National

Research Council (NRC) study on the subject of ballistic imaging [5], and a later highly critical NRC

study on forensic science in general [6]. While the completeness of the latter study especially has

been called into question [7] the fact remains that forensic examiners often find themselves on the

defense when it comes to presenting their expert opinions.

The success of DNA evidence in providing numerical assessment of duplication made

possible by known population statistics has created a call for comparative examinations to reach a

similar level of confidence. Such a mandate is somewhat unreasonable given the nature of the

evidence and the factors associated with the various types of analyses involved. However, there is no
 45

question that some degree of objectivity can be (and in some instances has been) introduced into

comparative examinations [8]. However, a problem lies in determining by which method to apply

comparative standards. This is a difficult proposition given the wide range of examinations possible,

e.g. questioned documents, fingerprints, tool marks, tire impressions, shoeprints, etc. and of course,

firearms. For the purposes of this paper, past efforts and current suggested solutions aimed at

introducing additional objective analysis into the area of firearm and tool mark examinations will be

the only area discussed.

Forensic identification of firearms and tool marks make use of the fine series of markings that

are impressed or scratched on bullets, cartridges, and surfaces when they come in contact with the

tool under consideration, be it a common hand tool or components of a firearm. The markings often

exist in the form of a fine series of parallel scratches and one of the earliest efforts to introduce

statistical analysis was suggested in 1959 by Biasotti [9]. This approach is based on observation and

tabulation of groups of “consecutive matching striae” in firearm and tool mark examinations [10] and

is known as the CMS method. Considerable work has been done investigating this possible technique.

More recently, quantitative measurements of tool marked surfaces using surface and optical

profilometers have been evaluated using a statistical algorithm to identify possible match pairs in a

completely objective manner [8]. However, this study showed that trained examiners making

subjective judgments are still able to distinguish between true matches and nonmatches at a higher

level of success than these objective methods [8].

It is well known that using the fine markings present as a means of identification has certain

problems and limitations, especially in the case of firearms, and these have been documented quite

extensively [11, 12]. In recent years a method has been developed that seeks to augment traditional

firearms identification by purposefully placing unique identifiers on certain critical pieces of a

firearm, such as the firing pin, breech face, etc. that are stamped into a cartridge when fired [13].

Termed Microstamping, this technique has received a large amount of political and media attention.
 46

In some cases local and state officials have introduced bills aimed at implementing microstamping of

either firearms or ammunition, perhaps without a proper understanding of the process or a

consideration of best practices concerning the use of this technique [14].

Certainly, one of the difficulties in any shooting investigation is to locate possible “suspect”

firearms that can be test fired to generate marks that can be compared to recovered items of evidence.

In theory, recovered items of evidence with microstamping could yield information that could assist

investigators in locating the responsible firearm much more quickly. However, while microstamping

does have the potential to greatly aid in firearm identification it clearly is not a panacea for the

difficulties associated with traditional examinations. For example, the criminal can always remove

firing pins, alter scratch patters by the use of abrasive polishing media, etc. Steps can be taken to

minimize the effect of such alterations by use of microstamping in several places but such

possibilities cannot be prevented entirely and will always exist. These considerations are not the topic

of this discussion.

What is of importance and should be understood by those who suggest or are contemplating

implementing laws utilizing microstamping is the effort that must be undertaken in order to optimize

the microstamped mark and ensure maximum transfer of the pattern. In other words, microstamping

involves more than just “blasting a number onto a firing pin using a laser”, which to the layman may

seem how the technique works. For each model of firearm an optimization process must be run. The

optimization process considers many physical characteristics of the area of the firing pin that strikes

the primer and how the laser used for engraving interacts with this area. These characteristics would

include material hardness, as well as shape, size and curvature of the firing pin. The optimum number

of characters and their arrangement for maximum clarity must also be considered, along with laser

parameters such as power input necessary to achieve this clarity. Thus, optimization is a complex

process involving a series of experimental determinations that must be conducted for each model

firearm of each manufacturer. [13]. Once completed the determined set of parameters can be applied
 47

to other firearms of the same type and material specifications in a production process. The cost of

optimization becomes small once an appreciable number of parts have been produced. However,

when one considers the large number of different firearm brands and models produced by any one

manufacturer, the effort to optimize all possible firearms becomes a significant research project of

considerable cost that must initially be undertaken. Such a project is separate and apart from the

economic costs that might be incurred by a company required to adopt microstamping. The latter

includes industry fears related to the purchase and maintenance of equipment, training of operators,

the speed of the process and its effect on production, etc. For example, if laws requiring that unique

identifiers be placed on numerous separate parts are passed, industry will have to ensure that guns are

assembled as a unique set of parts, rather than in a batch process of interchangeable parts, as is

currently typical.

Another consideration is the nature of the unique identifier selected for placement on each

firearm. Possibly the most common perception is that microstamping would involve placing the

serial number of the firearm on the firing pin. While large numbers of characters can be placed on a

firing pin [15] the most viable suggestion involves placing a more limited number of identifiers on the

pin, analogous to present license plates. This would provide for larger characters that are more easily

produced on a firing pin, transferred during the firing process, and recognized by an examiner. By

using a combination of alphanumeric characters, a six-digit code would provide a database of 36 x

106 unique designations (i.e. almost 2.2 billion possibilities), ten times the approximate number of

firearms in the U.S. today. A rapid field identification then becomes a simple matter of tracing the

number, in the same manner that license plates are traced today. In cases where the characters are not

readily readable a subsequent examination by a trained examiner would be necessary.

However, the question then arises as to who would oversee the assignment of identifiers and

maintain database integrity. Ideally, an oversight board could perform this function in much the same

way as the American Society for Testing of Materials (ASTM) oversees material specifications or the
 48

Accreditation Board for Engineering and Technology (ABET) accredits the quality of university

engineering programs in this country. These organizations are voluntary societies whose stated goals

are to preserve the quality of the members, industries, and institutions that they represent. A similar

arrangement, possibly consisting of sportsman associations, industry representatives, and advocacy

groups, might be formed to maintain a database and assign codes to participating companies that

choose to implement microstamping. The goal of the group would be to ensure that database integrity

is safeguarded while at the same time offering material assistance to law enforcement agencies.

Given the above considerations it is apparent that legitimate questions exist related to both

the technical aspects, production costs, and database management associated with microstamping that

should be addressed before wide scale implementation is legislatively mandated. However, it should

be noted that none of the above objections are inherently insurmountable. While it is likely that

microstamping will never approach the discriminating power associated with DNA evidence, it is a

viable method for providing rapid identification of a firearm in many cases, possibly decreasing the

current high workload of forensic examiners.

The purpose of this exploratory study is to examine one aspect of microstamping, namely, the

performance of a microstamped identifier on a small test set as a function of ammunition brand,

hardness, and firearm action type. Three different firearms representing the two most common

operating principles for semiautomatic pistols were chosen as well as 10 different brands of

ammunition. The results of the study and discussions concerning the various effects of primer

hardness and firearm brand are presented below. It is hoped that studies of this type can guide future

decisions as to the nature of the microstamped identifier that should be used, the probability of

unambiguous transfer, and the parameters that most affect clear transfer of the identifier.
 49

Experimental

The test set for this study involves use of three different 9mm semiautomatic handguns,

namely, a Sig Sauer model P226 semiautomatic pistol (short recoil action), a Taurus model PT609

semiautomatic pistol (short recoil action) and a Hi-Point model C9 semiautomatic pistol (simple

blowback action) where the firing pin also acts as an ejector. These guns were selected to represent a

range of performance and ejection properties and the actions are typical of the types of that leave fired

cartridges at crime scenes. Additionally, the firearms represent three different market price points, the

Sig Sauer being a higher priced firearm, the Taurus a medium priced item, and the Hi-Point being a

lower priced firearm.

Microstamping of the firing pins was optimized for a 6 character alphanumeric code and a

circumferential gear code for each firearm, which is intended to confirm the alphanumeric code. The

gear code is deciphered by dividing the circular code into eight equal sectors, excluding the wedge at

the top of the gear code in Figure 1. Beginning at the wedge, the code is read clockwise. Within each

sector, the notches are read as a six-bit binary code. For example, the first sector is read as 011001,

which corresponds to the letter “S” and the first identifier in the alphanumeric code. Subsequent

sectors correspond to the alphanumeric identifiers being read left to right. Further details concerning

use and interpretation of the gear code are available in the literature [13].

The optimization process involved a cycle of fire analysis to ensure optimal mark transfer by

identifying the surfaces, locations and vectors that provide the highest capability of transfer and

repeatability [13]. Both codes are designed to act in different ways to the multivariate kinetic motion

and the various instability vectors acting upon the cartridge during the cycle of fire. Both codes are

designed to be spatially out of phase with each other, ensuring that degradations (such as pin drag and

smear) which might wipe out certain characters in one code provide a high probability of survivability

for that character on the other code surface. Reading both codes provides a means of extracting the
 50

final code. One example of a stamped impression is shown in Figure 1, imaged using a scanning

electron microscope (SEM).

Figure 1: SEM image of a microstamped mark on a cartridge fired by the Sig-Sauer. Note the
gear code surrounding the alpha-numeric identifier.

The ammunition chosen for the study represents a considerable range of possibilities.

Ammunition brands were selected with a consideration of primer hardness [15] and a desire to

include sealant coated and manufacturer imprinted primers. Ten different brands were selected and

are listed in Table I in the order in which they were fired from the handguns. Before firing all of the

cartridges were marked using an electric scribe with a letter to denote the firearm used and then

sequentially marked from 1 to 1000 to make the firing sequence identifiable, Figure 2a. Thus, the T

306 cartridge was the 306th cartridge fired by the Taurus pistol. The order of ammunition used was

randomly selected by drawing names out of a hat.

The cartridges were loaded ten at a time into a magazine and fired. The highest shot order

number being loaded first and the lowest shot order number loaded last. The lowest number would

then be fired before the higher numbers. In the event a cartridge did not fire on the first try, the

cartridge was not removed from the chamber and a second pull of the trigger was tried (in the Sig-
 51

Sauer and Taurus pistols that were both single action and double action). If the cartridge failed to fire

on the second try, no further attempts to fire it were made and the misfired cartridge was placed in

order with the fired cartridge cases. A second attempt at firing was not carried out using the Hi-Point

pistol, which is only single-action. The spent rounds were collected during firing using a lightweight

cage / net that could be affixed to the gun hand of the person conducting the firings, Figure 2b.

Table I: Ammunition brands studied.

Firing Ammunition Primer Cartridge Description
Order Brand Type Material
1 Brown Bear Berdan Lacquered Steel 115 gr., full metal jacket, brass
primer
2 DAG Boxer Brass 124 gr., full metal jacket, brass
primer
3 Federal - Boxer Brass 115 gr., full metal jacket, nickel
American Eagle primer
4 Remington - Boxer Brass 115 gr., Flat Nose Enclosed Base,
UMC nickel primer, letters “H F” stamped
into the primer
5 PMC Boxer Brass 115 gr., full metal jacket, brass
primer
6 Silver Bear Berdan Zinc-plated steel 115 gr., full metal jacket, brass
primer
7 CCI Blazer Boxer Aluminum 115 gr., full metal jacket, nickel
primer
8 Cor-Bon Boxer Brass 147 gr., full metal jacket, nickel
primer
9 Independence Boxer Brass 115 gr., full metal jacket, nickel
primer
10 Sellier & Bellot Boxer Brass 115 gr., full metal jacket, brass
primer, covered with red lacquer
sealant

The pistols were cleaned after each 100 rounds. Cleaning consisted of brushing out the bore

with a nylon brush soaked in “PRO-SHOT 1 Step Gun Cleaner & Lubricant”. The bore was then

wiped out with a clean cotton flannel cleaning patch. The breech was thoroughly brushed using a

tooth-brush like commercial nylon brush. The top of the magazine and magazine follower were

wiped with an oily cleaning patch.

 52

The fired cartridge cases were placed back into the original box/tray from which they came

and the box was labeled with the pistol letter designation and the corresponding shot order numbers.

Thus a box labeled S601—S650 would contain shots 601 through and including shot 650 fired by the

Sig Sauer pistol. Cartridges missing from a tray would reflect casings that could not be found at the

firing range.

Figure 2: a) Unfired cartridge with inscribed identifier. b) Firing in progress with catch-basket.

After firing the primers of the cartridges were examined and graded as to the quality of the

microstamped impression. In conducting an assessment of this nature it becomes a matter of concern

whether a character is truly visible or whether the examiner, knowing what the character is supposed

to be, unconsciously ascribes greater clarity than actually exists. For example, after seeing 95 clear

impressions of a code it would be difficult to not immediately interpret the 96th cartridge as being

clear, even though some smearing may be present. Ideally one would want a different person to view

each separate cartridge without knowing what the identifier was supposed to be. This was obviously

not possible in this study. In order to somewhat account for this possibility two examinations were

undertaken. Firstly, Mr. Kreiser examined the cartridges and was instructed to be conscientiously

conservative in assigning his assessment. The examination involved use of a stereomicroscope

equipped with a polarized light for illumination and a simple rubric where the number of characters
 53

clearly visible using a stereoscopic examination was tabulated. Thus, a “C6” assessment means all

six characters were clearly visible while a “C3” would mean only three characters could be read

easily immediately. For this examination only the alphanumeric identifier was evaluated and

observations concerning multiple stamped identifiers, misfires, etc. were also noted. Secondly, the

cartridges were viewed and evaluated by T. Grieve, who has no training in forensic examinations at

all. The examination again involved a stereomicroscope with a polarized light source. In addition to

the alphanumeric identifier she examined whether there was any observable transfer of the gear code.

This evaluation was qualitative and did not determine what percentage of the code was visible, only

whether any useable portion survived. Thus, a “Y” evaluation meant that at least part of the code

transferred while “N” meant none was visible.

Note that the evaluation rubric employed by Mr. Kreiser might represent a “worst case

scenario” for the alphanumeric identifier while that used by Ms. Grieve is a “best case scenario” for

the gear code. Neither evaluation rules out the possibility of identifying either more characters or

more of the gear code using a more advanced imaging technique, nor does it necessarily preclude

reconstructing the entire code [13]. As an example of what might be visible using a more advanced

technique, certain cartridges having low C and gear code ratings were examined using a JEOL SEM

capable of both secondary (SEI) and backscattered (BES) electron imaging. Both imaging techniques

were used and the best images were chosen for presentation.

Vickers hardness measurements of the primers from the 10 selected ammunition types were

made using a LECO LM 247 AT microhardness tester. Loading was set at 50g and dwell time was 13

seconds. The measurements were made on the already fired primers as far as possible from the firing

pin impression in order to minimize any work hardening effects.

 54

Results

Microstamp Evaluation:

The results of the stereo-observations are summarized below in Tables II-IV. The data is

summarized both by firearm used and by brand of ammunition. The totals displayed in Table II

confirm that the ratings by J. Kreiser are more conservative as anticipated and discussed above. It is

also apparent from examination of Table II that the results show a strong correlation between that the

transfer of the identifier and the price point of the firearm, i.e. the most advantageous transfer occurs

for the Sig-Sauer, the worst by the Hi-Point.

The lacquer present on the Sellier & Bellot ammunition initially prevented clear observation

of the numbers and gear codes for the Taurus and Hi-Point fires, so cartridges 901-1000 for these

firearms were not graded by J. Kreiser and therefore are not shown in Table II. This results in

somewhat lower totals for the Taurus and Hi-Point samples. The optical analysis carried out by T.

Grieve is delineated in Table II by the use of italics. Note that the lacquer was subsequently removed

from 95 of the cartridges after J. Kreiser had examined them and before T. Grieve conducted her

examination. (Note: Five cartridges with lacquer were reserved to conduct further imaging

experiments on at a later time) and the totals obtained are included in the comments section. In

either case, it is clear that the use of lacquer has significantly degraded the ability to achieve total

identifier transfer.

Table II: Quality of microstamp as a function of firearm. Note that the numbers are out of 1000
fires for the Sig Sauer, out of 900 for the Taurus and Hi-Point. T. Grieve numbers in italics.
Strike Grade Summary
Sig Sauer Comments
C6 C5 C4 C3 C2 C1 C0 Cartridge #808 was lost and not graded or included in
948 30 14 5 1 0 2 the totals. There were 36 C6 double impressions. There
968 19 7 2 1 1 2 were 3 C5 double impressions. Cartridges S901-S1000
were graded after the lacquer was removed by T.
Grieve.
 55

Table II (Continued)

Taurus
C6 C5 C4 C3 C2 C1 C0 There were 26 C6 double impressions, 1 C5 double
848 43 3 1 3 2 0 impression, 1 C4 double impression and 1 C1 double
854 35 5 3 2 1 0 impression. 3 C6 misfires appeared. Cartridges 901-
1000 ungraded by J. Kreiser. Cartridges T901-T1000
graded after the lacquer was removed by T. Grieve
produced C6:56, C5:26, C4:10, C3:1, C2:1, C1:0, C0:0
Hi-Point
C6 C5 C4 C3 C2 C1 C0 There were 52 C6 double impressions, 14 C5s, one C4,
663 139 47 26 15 5 4 one C3 and one C2. There was one C6 triple
684 113 65 25 7 4 1 impression. Of the 12 misfires, 6 were C6, 4 were C5, 1
was C4 and 1 was C0. Cartridges H901-H1000
ungraded by J. Kreiser. Cartridges H901-H1000 graded
after the lacquer was removed by T. Grieve produced
C6:49, C5:15, C4:12, C3:8, C2:4, C1:5, C0:2

It is interesting that it was often found that poorly marked cartridges would be grouped

together. This tendency was seen for all firearms but clearly occurred more often for the lower cost

Hi-Point. For example, for the Hi-Point 125 of the 237 non-C6 ratings found by Kreiser came in runs

of two to five consecutive cartridges. The tendency for multiple groups of poorly marked cartridges

seemed to be exacerbated by the presence of lacquer. For example, of the 52 non-C6 ratings found by

Kreiser for the Sig Sauer firings, eight groups of two and one run of nine non-C6 ratings occurred, i.e.

25 out of 52, all in the Sellier & Bellot cartridges. For the Taurus both Kreiser and Grieve found four

runs of two or more for the non-Sellier & Bellot ammunition; in the Taurus Sellier & Bellot cartridges

Grieve noted an additional six runs of two or more, the largest run being six consecutive non-C6

ratings.

Table III: Quality of microstamp as a function of ammunition, J. Kreiser results.

Summary of Cartridge Types
Brown Bear (#1-100) Comments
Gun C6 C5 C4 C3 C2 C1 C0
Sig 99 1 0 0 0 0 0
Taurus 93 7 0 0 0 0 0 2 misfires, C6
Hi-Point 92 7 1 0 0 0 0 1 triple impression
 56

Table III (Continued)

DAG (#101-200)
Gun C6 C5 C4 C3 C2 C1 C0
Sig 99 0 1 0 0 0 0
Taurus 89 9 1 0 1 0 0
Hi-Point 86 8 2 3 0 0 1 Ctg. 159 pierced
Federal American Eagle (#201-300)
Gun C6 C5 C4 C3 C2 C1 C0
Sig 97 3 0 0 0 0 0
Taurus 92 2 1 1 2 2 0
Hi-Point 62 23 8 3 2 1 0 Ctg. 251 lost
Remington UMC (#301-400)
Gun C6 C5 C4 C3 C2 C1 C0 Existing letters create interference with strike
pattern
Sig 99 1 0 0 0 0 0
Taurus 91 9 0 0 0 0 0
Hi-Point 92 6 2 0 0 0 0
PMC Bronze (#401-500)
Gun C6 C5 C4 C3 C2 C1 C0
Sig 100 0 0 0 0 0 0
Taurus 99 1 0 0 0 0 0
Hi-Point 64 25 9 1 1 0 0
Silver Bear (#501-600)
Gun C6 C5 C4 C3 C2 C1 C0
Sig 99 0 1 0 0 0 0
Taurus 89 10 1 0 0 0 0
Hi-Point 58 20 8 7 4 1 2 4 misfires, C6
CCI Blazer (#601-700)
Gun C6 C5 C4 C3 C2 C1 C0
Sig 99 1 0 0 0 0 0
Taurus 98 2 0 0 0 0 0
Hi-Point 73 15 5 5 0 2 0 1 misfire, C6
Cor-Bon (#701-800)
Gun C6 C5 C4 C3 C2 C1 C0
Sig 96 2 1 0 0 0 1
Taurus 97 3 0 0 0 0 0
Hi-Point 67 22 6 1 3 0 1 4 C5 misfires, 1 C4 misfire and 1 C0 misfire
Independence (#801-900)
Gun C6 C5 C4 C3 C2 C1 C0
Sig 99 0 1 0 0 0 0
Taurus 100 0 0 0 0 0 0
 57

Hi-Point 69 13 6 6 5 1 0 1misfire, C6
Sellier & Bellot (#901-1000)
Gun C6 C5 C4 C3 C2 C1 C0
Sig 61 22 10 5 1 0 1
Taurus - - - - - - - Lacquer prevented observation in Taurus and Hi-
Point
Hi-Point - - - - - - - Lacquer prevented observation in Taurus and Hi-
Point

Table IV: Quality of microstamp as a function of ammunition, T. Grieve.

Summary of Cartridge Types

Brown Bear (#1-100) Comments
Gun C6 C5 C4 C3 C2 C1 C0
Sig 95 2 0 1 1 1 0 Y=100 N=0
Taurus 89 11 0 0 0 0 0 Y=0 N=100
Hi-Point 86 13 1 0 0 0 0 Y=95 N=5
DAG (#101-200)
Gun C6 C5 C4 C3 C2 C1 C0
Sig 94 3 2 1 0 0 0 Y=81 N=19
Taurus 97 1 1 1 0 0 0 Y=0 N=100
Hi-Point 89 3 4 2 0 2 0 Y=95 N=5
Federal American Eagle (#201-300)
Gun C6 C5 C4 C3 C2 C1 C0
Sig 99 1 0 0 0 0 0 Y=100 N=0
Taurus 95 1 0 1 2 1 0 Y=55 N=45
Hi-Point 64 23 8 3 1 0 0 Y=95 N=4
Remington UMC (#301-400)
Gun C6 C5 C4 C3 C2 C1 C0
Sig 99 1 0 0 0 0 0 Y=100 N=0
Taurus 98 2 0 0 0 0 0 Y=0 N=100
Hi-Point 89 7 4 0 0 0 0 Y=98 N=2
PMC Bronze (#401-500)
Gun C6 C5 C4 C3 C2 C1 C0
Sig 100 0 0 0 0 0 0 Y=100 N=0
Taurus 100 0 0 0 0 0 0 Y=0 N=100
Hi-Point 63 16 13 7 1 0 0 Y=98 N=2
Silver Bear (#501-600)
Gun C6 C5 C4 C3 C2 C1 C0
Sig 99 1 0 0 0 0 0 Y=93 N=7
Taurus 82 13 4 1 0 0 0 Y=0 N=99
Hi-Point 63 14 12 5 3 2 1 Y=86 N=14
Blazer (#601-700)
Gun C6 C5 C4 C3 C2 C1 C0
Sig 100 0 0 0 0 0 0 Y=100 N=0
Taurus 99 1 0 0 0 0 0 Y=0 N=100
Hi-Point 83 12 3 2 0 0 0 Y=94 N=6
 58

Table IV (Continued)

Cor-Bon (#701-800)
Gun C6 C5 C4 C3 C2 C1 C0
Sig 98 0 1 0 0 0 1 Y=97 N=3
Taurus 95 5 0 0 0 0 0 Y=0 N=100
Hi-Point 74 13 9 3 1 0 0 Y=91 N=9
Independence (#801-900)
Gun C6 C5 C4 C3 C2 C1 C0
Sig 99 1 0 0 0 0 0 Y=100 N=0
Taurus 99 1 0 0 0 0 0 Y=0 N=100
Hi-Point 73 12 11 3 1 0 0 Y=97 N=3
Sellier & Bellot (#901-1000)
Gun C6 C5 C4 C3 C2 C1 C0 Lacquer removed from cartridges
Sig 85 10 4 0 0 0 1 Y=77 N=23
Taurus 56 26 10 1 1 0 0 Y=0 N=95
Hi-Point 49 15 12 8 4 5 2 Y=78 N=17

SEM Evaluation:

After the optical examination a few of the lower-scoring cartridges were selected for SEM

examination. One example from each of the firearms used is shown below. Figure 3 shows cartridge

#S198, rated as C3-Y by T. Grieve and C4 by J. Kreiser. For comparison see Figure 1, obtained from

a cartridge rated as a C6-Y.

Figure 3: SEM imaging of cartridge #S198, DAG ammunition, Sig Sauer handgun
 59

It is left to the reader as an unbiased observer to decide how many of the alphanumeric

characters are visible. To the authors (who, admittedly, know the code) it appears the code is

S23SX7, i.e. complete identification can be made using a higher quality image. The gear code,

though visible, is difficult to discern in small regions of this particular cartridge. Figure 4 shows an

example cartridge from the Taurus, #T944. Rated a C2-N optically by T. Grieve (not rated by J.

Kreiser due to the lacquer), this example shows the problems involved when using a lacquered

cartridge. The four alphanumerics at the corners, difficult to discern using optics, are clearly visible

using SEM, being T13A5L. The gear code is totally lacking, and in general the gear code did not

transfer for the Taurus handgun.

An example from the Hi-Point series is shown in Figure 5. The Hi-Point had the poorest

transfer of the alphanumeric, although a high percentage of the cartridges had some gear code

available, causing a much higher rating in this area than the Taurus. Figure 5 makes it clear, however,

that the gear code was present over a relatively small area, in this case the upper right quadrant.

Rated as a C2-Y optically by T. Grieve and C3 by J. Kreiser, SEM imaging in this case sheds little

light on the identifier, possibly allowing one additional character of the identifier H60PZE to be

visible.
 60

Figure 4: Cartridge #T944, Sellier & Bellot ammunition, Taurus handgun

Figure 5: Cartridge #H519, PMC ammunition, Hi-Point handgun

Hardness Evaluation:

The primer hardness values obtained from the 10 types of ammunition used are shown in

Table V. The presence of lacquer on the Sellier and Bellot cartridges presents a special problem

when measuring hardness. Just as it is clear that the lacquer prevents an immediately recognizable

mark transfer while it remains on the cartridge, evaluating the hardness with the lacquer present is
 61

meaningless since the soft nature of the lacquer disrupts the method used to measure hardness,

producing meaningless results. Thus, the lacquer was removed and the values reported in Table V

reflect the actual hardness of the uncoated primer.

Table V: Vicker’s Hardness of the ammunition studied.

Ammunition Type Average Hardness (HV) Primer type Comments
Brown Bear 157.88 Brass 284 total C6
DAG 177.71 Brass 274 total C6
Federal American Eagle 165.30 Nickel 251 total C6
Remington UMC 236.31 Nickel 282 total C6; Primer
contained manufacturer-
stamped letters
PMC Bronze 150.29 Brass 263 total C6
Silver Bear 162.80 Brass 246 total C6
CCI Blazer 176.62 Nickel 270 total C6
CorBon 164.38 Nickel 260 total C6
Independence 167.17 Nickel 267 total C6
Sellier &Bellot 160.68 Brass Lacquer coated Primer,
removed for hardness tests.

Discussion

It seems clear from the above results that both brand of ammunition and type of firearm play

a role in identifier transfer. When considering ammunition no primary parameter could be identified

as ensuring complete identifier transfer, i.e., no consistent trends were observed as a function of either

primer material, type or hardness, and/or cartridge case material. For example, if one simply uses the

total number of C6 ratings per ammunition type as a rough comparison system, the three highest rated

ammunitions are the Brown Bear (115 gr., brass primer, 157.88 Hv), the UMC (115 gr., nickel

primer, 236.31 Hv), and the DAG (224 gr., brass primer, 177.71 Hv). Given that the transfer quality

does vary substantially, further study is necessary before any definitive statements can be made

concerning the effect of ammunition type. However, it is clear that the presence of lacquer is of

paramount importance in identifier transfer. For example, for the Sig Sauer results examiner J.
 62

Kreiser scored 52 non-C6 marks, 39 of which were seen in the Sellier & Bellot before the lacquer was

removed, i.e. 75% of the poor markings came in the lacquered ammunition. The effect of the lacquer

was so great on the Taurus and Hi-Point marks that Mr. Kreiser did not even attempt to rate these

cartridges. Even after removal of the lacquer the effect was still apparent; Ms. Grieve found that 15 of

the 32 non-C6 marks she recorded for the Sig Sauer (47%) came from the Sellier & Bellot cartridges

and 38 of 90 for the Taurus (42%). For the Hi-Point 46 of the 95 Sellier & Bellot cartridges examined

(48%) were non-C6.; this compares to an average of 24% non-C6 ratings for the rest of the

ammunition types examined.

The type of firearm seems to play the largest role in the overall quality of identifier transfer.

Depending on whose evaluation you chose to use, success rate for a C6 transfer for the Sig-Sauer was

in the range 95-97%, for the Taurus 91-94%, and for the Hi-Point 68-74%. The firearms used were

specifically selected to cover a range of pistol operating systems and prices and it is clear that the

higher priced firearms, possessing a short recoil action, result in the transfer of a more easily

distinguishable identifier than the Hi-Point which has a simple blowback mechanism with a firing pin

ejector.

It should be noted that the firing pin is involved in the ejection of spent cartridges from the

Hi-Point, and is necessarily in contact with the primer during this time. This makes it difficult to say

whether the multiple strike marks seen on spent cartridge primers from the Hi-Point came solely from

a multiple strike scenario (as would be the case for the Sig Sauer and Taurus firearms) or whether the

ejection mechanism also contributed to the multiple markings. It is certainly true that the Hi-Point

suffered a much higher rate of multiple markings than did either the Sig Sauer or the Taurus.

The poor transfer of the gear code in the case of the Taurus was investigated by examining

additional firing pins that had also been microstamped using the same identifier for the purposes of

this study. SEM images of the pins, shown in Figure 6, reveal that while the alpha-numeric number is

clear the gear-code is somewhat sparse in detail compared to the Sig Sauer cartridge of Figure 1, and
 63

is not as clearly defined in some areas, particularly in the arc quadrant encompassing the “A” of the

identifier.

Figure 6: SEM backscattered images of three pins microstamped for the Taurus firearm.

Measurement of the radii of curvature of the firing pins for the three handguns examined

revealed that the curvature of the Taurus pins is much greater than either the Sig Sauer or Hi-Point,

the radii being 664 microns, 883 microns, and 1180 microns, respectively. Presumably this makes it

harder for the gear code on the Taurus to effectively mark a primer.

Although the complete identifier did not mark in every case, this is not to say that it could not

have been reconstructed using more advanced imaging techniques. SEM imaging in many cases could

reveal more of the identifier and gear code than was visible using simple optics. Previous studies [13]

have shown that a combination of better imaging, examination of multiple cartridges from the same

weapon and a careful analysis of the gear code can bring out additional information that is not

immediately obvious by a simple examination. Such detailed studies again would have to be

conducted by a forensic examiner trained in the use of both the necessary equipment and the

methodologies used. Whether a simple optical examination using a low-powered magnifying glass by

an untrained examiner is possible is a matter that needs to be investigated, and efforts are underway to

secure funding to conduct a blind study of this type.

 64

Summary and Conclusions

In this study 10 different ammunition brands were fired from three different brands of

firearms that were equipped with firing pins containing a unique microscopic identifier. Differences

in the clarity of the microstamped identifier were evaluated using simple observation employing a

stereomicroscope. While some differences in clarity were seen as regards brand of ammunition, the

observed results could not be related to most of the ammunition variables examined, which included

primer material (brass vs. nickel), hardness, type (Boxer vs. Berdan), or cartridge material (brass,

aluminum, or steel). The only obvious difference in quality occurring when using lacquered

ammunition, which degraded identifier transfer. Greater differences were seen when comparing the

type of firearm, where the Hi-Point transferred less well than the Sig Sauer or Taurus. However,

while the Taurus alphanumeric identifier transferred extremely well the gear code transferred either

very poorly or not at all.

While readable microstamping was achieved on most of the cartridge cases, it was also clear

that it is not a perfect technology, even on optimized weapons, as the poorer transfer of the Taurus

gear code would indicate. As discussed in previous papers the interaction of any particular brand of

ammunition with any given firearm is stochastic in nature [16]. Such a variable process prevents

perfect transfer in all cases and makes interpretation of the results of this study difficult as regards

primer hardness effects.

Despite shortcomings, microstamping does have the potential to place valuable information

into the hands of the officer or detective at the scene of a crime in a timely fashion. If coupled with an

independent, voluntary oversight board, established and maintained by firearm manufacturers and

sportsman associations to control issuance of the identifier and maintain privacy, microstamping

could enable tracking of fired cartridges in an efficient and timely manner.

 65

Acknowledgments

This research project was funded by the U.S. Department of Justice, National Institute of Justice,

through the Midwest Forensics Resource Center at Ames Laboratory, under Interagency Agreement

number 2008-DN-R-038. The Ames Laboratory is operated for the U.S. Department of Energy by

Iowa State University, under contract No. DE-AC02-07CH11358.

References

1. B. Regan, ‘Reasonable Doubt,” Popular Mechanics, August, 46, 2009.

2. L. Goff, “Quick Study: Crime Scene Science”, Reader’s Digest, Feb. 2011.

3. Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993).

4. S.T. Wax, C.J. Schatz; “A Multitude of Errors: The Brandon Mayfield Case”, Champion

Magazine, September/October, p 6, 2004.

5. Report by a committee for the National Research Council for the National Academy of

Sciences, “Ballistic Imaging”, National Academies Press, March, 2008.

6. Report by a committee for the National Research Council for the National Academy of

Sciences, “Strengthening Forensic Science: A Path Forward”, National Academies Press,

March, 2009.

7. Comments made during the session “The Scientific Foundations of Forensic Science”,

Impression and Pattern Evidence Symposium, Clearwater Beach, Florida, August, 2010.

8. A. Biasotti, “A Statistical Study of the Individual Characteristics of Fired Bullets,” Journal of

Forensic Science 4(1), 34-50, 1959.

9. A. Biasotti, J. Murdock, “Criteria for Identification or State of the Art of Firearm and Tool

mark Identification,” Association of Firearm and Tool mark Examiners, 4, 16-24, 1984.
 66

10. L.S. Chumbley, M. Morris , J. Kreiser, C. Fisher, J. Craft, L. Genalo, S. Davis, D. Faden, and

J. Kidd, “Validation of Tool mark Comparisons Obtained Using a Quantitative, Comparative,

Statistical Algorithm,”, Journal of Forensic Science (accepted).

11. R.G. Nichols, “Firearm and Toolmark Identification Criteria: A Review of the Literature,”

Journal of Forensic Science 42 (3), 466-474, 1997.

12. R.G. Nichols, “Firearm and Toolmark Identification Criteria: A Review of the Literature –

Part 2,” Journal of Forensic Science 48 (2), 318-327, 2003.

13. O. P. Ohar, T. E. Lizotte, “Extracting Ballistic Forensic Intelligence: Microstamped Firearms

Deliver Data for Illegal Firearm Traffic Mapping – Technology,” Proc. Of SPIE Vol. 7434,

743416, (2009).

14. http://www.sciencedaily.com/releases/2008/05/080514092333.htm

15. F. Tulleners, “Vickers Hardness Values of Selected 40 S&W Primers,” AFTE Journal, Vol.

35, no. 2, Spring, 2003

16. O. Ohar and T. Lizotte, “The Stochastic nature of Firearm Firing Pin Impressions Revealed

by the use of Intentional Firearm Microstamp (IFM)” presentation given at Northeastern

Association of Forensic Scientists, Criminalistics Session II, November 5, 2009.

 67

CHAPTER 4. GEAR CODE EXTRACTION FROM MICROSTAMPED

CARTRIDGES
A paper published by The Association of Firearm and Toolmark Examiners Journal, Volume
45, Number 1, pp. 64-74

T. Grieve, L.S. Chumbley, J. Kreiser*, T. Lizotte†, O. Ohar†

Iowa State University, Ames Laboratory

Ames, Iowa

*Illinois State Police, Retired

Springfield, Illinois
†
Pivotal Development, LLC
Manchester, NH

Introduction

Over the past few years, intentional firearm microstamping has received a large amount of

attention from technical discussions [1-4], lawmakers [5], and the media [6-8]. Microstamping

involves placing alpha-numeric identifiers onto the surface of various components associated with the

firing of a firearm, such as the firing pin or breech face. These unique identifiers are then

automatically transferred to the cartridge upon firing due to the forces involved in the action. While

microstamping can be used to transfer large numbers of characters [9] more effort has been devoted

toward and eight character alpha-numeric on the firing pin tip with a circular gear code around the

circumference of the pin [2]. It is proposed that these microstamped identifiers can be used as a

simple, objective, and rapid means of identification of a particular gun, similar to the way a license

plate identifies a particular car. An example of a microstamped mark showing both alpha-numeric

identifier and circular gear code is shown in Figure 1.

 68

While simple visual observation can determine what the identifiers are if the microstamping

is clear, distortion of the transfer makes their identification much more difficult. If the alpha-numeric

characters are deformed, or partially removed due to the firing and cartridge ejection process, the only

means of identification for the original microstamped identifier might be the gear code. Thus, the gear

code could provide important information that could either fill in any gaps in a distorted alpha-

numeric code, or be used to replicate the code if the alpha-numeric identifier is entirely illegible.

Figure 1: a) Microstamped mark from a Sig Sauer P226 semiautomatic handgun. b)

Microstamped mark with gear code overlay.

The gear code is deciphered by dividing the circular code into eight equal sectors, excluding

the wedge at the top of the gear code, which marks the start of the sequence (see gray line in Figure

1b). This translates to eight sectors of 42 degrees, delineated in Figure 1b by straight, white lines. The

code is then read clockwise, in six bit binary, where each bit is a 7 degree increment, as shown in

Figure 1b by black lines. The numbers “0” and “1” then correspond to whether the primer is left in

the unstamped or stamped condition, respectively. For example, the first section of gear code in

Figure 1b is then read as 011001, which corresponds to the letter “S” and is also the first character in

the alpha-numeric code. The subsequent sectors correspond to the identifiers being read left to right.

Thus, in Figure 1, the second sector represents the second character, 2, the third 3, etc. The gear codes
 69

contain the numbers 0-9 and all letters of the alphabet, excluding I, O and Q to eliminate any

confusion in evaluation. More information regarding gear codes, microstamping, and translation of

the digital code into the alpha-numerics can be found in the literature [2]. A table showing the digital

code and the corresponding alpha numeric is shown in Table I.

In this study the efficiency of transfer of gear codes from micro-etched firing pins to a variety

of ammunition types is reported. This paper constitutes a follow-up to an earlier study where the

alpha-numeric was examined [3]. Readers are encouraged to cons

consult
ult this earlier study for a full

understanding of the experimental design.

Table I: Variable pitch gear code table.

 70

Table I (Continued)

Experimental

Samples examined in this paper were described in a previous study [3]. Briefly, cartridges

were fired and examined using three different semiautomatic handguns: a Sig Sauer model P226

pistol, a Taurus model PT609 and a Hi-Point model C9. Six character microstamped firing pins were

optimized for these guns and ten different brands of ammunition repre
representing
senting a range of primer

hardness and types were selected. Each gun was used to fire 100 rounds of each brand of ammunition,

10 rounds per magazine, for a total of 1000 rounds per firearm. The brands of ammunition used can

be found in Table II.

Table II: Ammunition brands used in the study.

Firing Ammunition Brand Primer Primer Description
Order Type Material
1 Brown Bear Berdan Brass 115 gr., full metal jacket
2 DAG Boxer Brass 124 gr., full metal jacket
3 Federal - American Eagle Boxer Nickel 115 gr., full metal jacket

4 Remington - UMC Boxer Nickel 115 gr., Flat Nose Enclosed

Base, letters “H F” stamped
into the primer
5 PMC Boxer Brass 115 gr., full metal jacket
6 Silver Bear Berdan Brass 115 gr., full metal jacket
7 CCI Blazer Boxer Nickel 115 gr., full metal jacket
 71

Table II (Continued)

8 Cor-Bon Boxer Nickel 147 gr., full metal jacket

9 Independence Boxer Nickel 115 gr., full metal jacket
10 Sellier & Bellot Boxer Brass 115 gr., full metal jacket,
primer covered with red
lacquer sealant

Evaluation of the microstamped alpha-numeric identifiers has already been published [3].

Optical grades were given based upon the number of clearly legible alpha-numeric characters visible

using a stereomicroscope. If all six identifiers were clearly read, the cartridge received a grade of C6,

if only five identifiers were clear, the cartridge was graded C5, etc. For the current study, only fired

cartridges that received an optical grade of C2 or below were chosen for evaluation for the Hi-Point.

Since the Taurus and Sig Sauer generally received better optical grades, cartridges of less than C6

were evaluated. A total of 26 cartridges of poor grades were evaluated, seven from the Sig Sauer gun,

seven from the Taurus, and 12 from the Hi-Point.

The selected cartridges were cleaned and examined using a JEOL 6060LV scanning electron

microscope (SEM). Pictures were taken using either secondary electron imaging or backscattered

electron imaging, depending on which imaging technique made the gear code more legible. The SEM

images obtained were then examined using a free photo editing software (GIMP), the outline of clear

gear code was traced, and an overlay of the correct angles was placed upon the image to evaluate the

gear code.

Results

As with the previous microstamp study [3], the Sig Sauer had the best transfer of gear code

and legible identifiers, while the Hi-Point and the Taurus did not transfer identifiers and gear codes

quite as well. In this section examples of analyses from several selected cartridges will be presented,

followed by a summary of results for all of the cartridges examined.

 72

Sig Sauer

In Figure 2, Sig Sauer cartridge number 24 (Brown Bear) graded C2 optically is shown. More

detail is visible in the SEM image than when using a stereomicroscope and the identifier appears to be

S23-SX7 by simple SEM examination without resorting to the gear code. In this instance the gear

code is complete and can be clearly deciphered. All eight characters are visible and decode as S23-

SX7-SS, which confirms the assessment of the alpha-numeric based solely on SEM imaging.

Figure 2: SEM image of a) Sig cartridge #24, Brown Bear. b) Outlined gear code and overlay.

While generally the Sig Sauer had the best and most consistent transfer [3], this was not true

in all cases. Figure 3, shows an example of a poorly marked cartridge (Cor-Bon) that was graded C0

optically. The SEM image reveals more identifiers in addition to a partial gear code.

Figure 3: a) SEM image of Sig cartridge #707, Cor-Bon. b) Outlined gear code and overlay.
 73

Estimating exactly how many of the alpha-numerics can be deciphered using SEM is

somewhat artificial since the identifier is already known. While it is difficult to be totally objective, it

would appear that an unbiased observer might make a reasonable guess at 2-3 of the alpha-numerics,

possibly S*3 – S*7 at best, based solely on SEM imaging. While only part of the gear code can be

deciphered, it still yields enough information to confirm the first three identifiers and part of the

fourth. The first sector can be read as “S”, the second as “2”, the third as “3”. Complete transfer fails

at the fourth identifier.

Taurus

The Taurus firing pin did not mark gear codes nearly as well as that of the Sig. This was

partly due to the sharper radius of the pin [3] and partly due to the sparse gear code on the pin [3], i.e.

the code consisted of large continuous areas of stamped “1” or unstamped “0”. This absence of

surface relief was found to make it difficult to determine whether the cartridge was left unstamped to

denote a 0 or whether the cartridge simply was not marked at all. As a result, very little additional

knowledge as to the unique identifier was added by the presence of the gear code. An example is

shown in Figure 4, which is cartridge, number 233 (American Eagle). Optically, this cartridge was

graded C2, although the better imaging available using the SEM allows the first three alpha-numerics

to be read as T13 fairly easily, with suggestions of 2 additional identifiers, possibly a 5 or an S, and a

1. When examining the gear code the sectors for identifiers 3-8 are not visible at all; the first two

sectors of the code yield the correct identifiers T and 1.

 74

Figure 4: a) SEM image of Taurus cartridge #233, American Eagle. b) Outlined gear code and
overlay.

In general for the Taurus cartridges examined, only the first two identifiers could be extracted

from the gear code. Figure 5 shows an even poorer alpha-numeric and gear code transfer from

cartridge #296 (American Eagle) graded C1 optically. Again the SEM imaging allows 1 and 3 to be

ascertained from the alpha-numeric but only the number “1” is able to be deciphered using the gear

code, which falls in the second sector of the eight possible sections. All other sectors appear

distorted, precluding any interpretation with a high level of confidence.

Figure 5: a) SEM image of Taurus cartridge #296, American Eagle. b) Outlined gear code and
overlay.
 75

Hi-Point

Like the Taurus, the Hi-Point did not transfer its gear code as well as the Sig Sauer. However,

the Hi-Point pin did have a more robust gear code with considerable surface relief, which made it

somewhat easier to discern if the primer had indeed been marked. In Figure 6, cartridge #610 (CCI

Blazer) graded optically as C1 is shown. Again the SEM reveals more of the alpha
alpha--numeric than

could be seen optically as well as a fraction of the gear code. In Figure 6a the identifier appears to be

H60-PZ*, with the last alpha-numeric undistinguishable. When considering the gear code, “H” can

be read clearly, but the “6” is slightly muddled. As the outline shows in Figure 6b, the gear code for

the second identifier appears to read 000100, which would correspond to the number “4”. This is

obviously incorrect and forces aan

n examiner to decide between what appears to be a clear marking of

the alpha-numeric and the validity of the gear code.

Figure 6: a) SEM image of a Hi-Point cartridge #610. b) Outlined gear code and overlay.

In this particular cartridge, the primer seems to have been struck twice and smeared, which

distorts the alpha-numerics and obscures the correct gear code reading of (000110). Double strikes

were especially prevalent in the Hi-Point.

A second example is shown in Figure 7. This Silver Bear cart

cartridge,
ridge, #520, was graded C0

optically. However, when imaged with SEM reasonable guesses could be made as to the identity of
 76

most of the alpha-numerics. Although there is considerable uncertainty and judgment involved, the

identifier seems to be an H or an A, followed possibly by a 6, then 0. The second three-digit sequence

appears to be possibly a P, followed by Z, then maybe a 5. In this case the gear code lends valuable

assistance and permits unambiguous identification of the first two sectors, which translate as “H” and

“6”, confirming the tentative assessment of the image. The third sector almost reveals the third

identifier as “0”, but the last bit of the gear code didn’t transfer. However, since most of the “0” did

transfer on the identifier, an examiner might conclude that the first three digit sequence is H60.

Figure 7: a) SEM image of Hi-Point cartridge #520, Silver Bear. b) Outlined gear code and
overlay.

Like cartridge #610, cartridge #716 from the CorBon ammunition set also has an apparent

erroneous gear code for the second digit. Optically, this cartridge was graded as C2, but three

additional alpha-numerics are revealed through the SEM image. As seen in Figure 8, the first sector

of the gear code reads correctly as 010001 (H), but again the second sector reads as 000100 (4). From

the alpha-numerics that transferred, it’s clear that the second alpha-numeric is actually a “6” and not

“4” as the gear code suggests. The gear code corrects itself at the third sector and reads as 000000 (0).

The gear code also correctly gives us the missing alpha-numeric, H, changing the overall clarity

rating to C6. However, in a real-life setting the fact that the gear code does not match a corresponding

clear alpha-numeric indicator casts doubt on any identification based on the gear code alone. Thus,
 77

while the entire code can be reconstructed, in all probability this identification would be disregarded

as being unreliable. This instance points to a problem where an unclear marking of the gear code

leads to a false interpretation.

Figure 8: a) SEM image of Hi-Point cartridge #716, CorBon. b) Outlined gear code and overlay.

It is important to note at this point that the gear codes on the firing pins used for #610 and

#716 are correct and that the error is introduced during the marking. Examination of both #610 and

#716 using SEM show that both cartridges appear to have been double-struck. This presumably is the

reason for the apparently erroneous gear code markings.

Lacquered Cartridges

Lacquered cartridges, from the Sellier & Bellot ammunition, posed problems during the

optical and SEM evaluations, especially for the Hi-Point cartridges as it interfered with the transfer of

the identifiers and the gear code. As seen in Figure 9, Sig Sauer cartridge #909 (S&B) does not have

the clarity that the earlier cartridges did in either the alpha-numeric characters or the gear code. In

fact, the only parts of the gear code that can be readily deciphered are the first and last sections, both

of which read 011001 (S).

 78

Figure 9: a) SEM image of Sig Sauer cartridge #909, S&B. b) Outlined gear code and overlay.

The slightly smeared Sig Sauer transfer described above still appears fairly clear, however,

especially when compared to the poorest transfers from some of the Hi-Point cartridges. Figure 10 is

a good example of some of these transfers. Hi-Point cartridge #974 (S&B) in Figure 10a was graded

optically as C0 and its grade only improves to C1 with SEM and gear code analysis. By comparison,

the gear code on cartridge 937 did not fare as well as that of 974. The first half of the visible portion

is wiped out, making any analysis of the gear code futile. However, the SEM analysis does yield

another alpha-numeric character than the optical grade did, making the total clarity rating C2.

Figure 10: a) SEM image of HP cartridge #974. b) SEM image of HP cartridge #937.
 79

Like the unlacquered cartridges, the lacquered Taurus cartridges showed poor gear code

transfer, even to the extent of lacking the starting wedge marker (Figure 11). Though the lacquer

smeared the alpha-numerics of the Hi-Point extensively, the Taurus did not exhibit such extreme

distortion. As evidenced by Figure 11, the alpha-numerics are still legible. Cartridge #945 (S&B)

shown in Figure 11a was graded optically as C3 and with SEM evaluation the total clarity grade

conservatively becomes C4 and it could be argued a C6. Cartridge 944 was graded optically as C2,

but all 6 alpha-numerics are visible in the SEM image.

Figure 11: a) SEM image of Taurus cartridge #944. b) SEM image of Taurus cartridge #945.

Gear Code Analysis by Magazine

Often where a shooting has occurred several cartridge cases may be left behind. Assuming

that the gun used was equipped with a microstamped firing pin, one argument made in defense of

compiling, or adding, partially transferred markings is that given a large number of incompletely

marked cartridges from (presumably) the same firearm, could the entire identifier be reconstructed?

An analogy would be that part of an automobile license plate is better than no plate number at all. To

examine this hypothesis, cartridges from two magazines from each gun were examined optically with

a stereomicroscope, one from a non-lacquered ammunition set and the other from the lacquered S&B
 80

cartridges. Each magazine chosen had the highest number of non-C6 ratings to represent a possible

worst case scenario. Table III summarizes the grades of the chosen magazines. The bold, capital X’s

denote both the alphanumeric character and its corresponding section of gear code were legible, the

lower case, x’s denote only the alphanumeric character having a clear transfer and GC denotes only

the gear code being decipherable. If the table is blank it means for that cartridge neither the alpha-

numeric or gear code were decipherable.

Not surprisingly, the only complete alpha-numeric + gear code transfers occurred in the Sig

Sauer, both unlacquered and lacquered. It should be noted, however, that due to the presence of

lacquer in cartridges 901-1000, the transferred gear code was slightly smeared, but the code in many

cases could still be deciphered.

The Taurus cartridges again did not have all of the gear code on the unlacquered cartridges,

though they did assist in identifying the first one or two alphanumeric identifiers. The lacquered

Taurus cartridges were largely unhelpful in examining the gear code. The Taurus firing pin’s lack of

surface relief combined with the lacquer coated primers caused no gear code transfer in the Sellier &

Bellot cartridges. In some cases, even the start wedge of the gear code failed to transfer.

The Hi-Point gear codes were slightly more helpful than those of the Taurus. Still, the gear

code transfer did not extend beyond the “0,” and as evidenced by the table, in several cases did not

transfer or did not transfer legibly.

Despite the poor performance in some cases, it is still apparent that if one knows or could

safely assume that all ten cartridges found at a crime scene came from a single clip of ammunition,

the entire identifier could be reconstructed using the combined information for every magazine

examined in this study.

 81

Table III: Summary of gear code and alphanumeric character evaluation from low grade clips

Hi-Point
Unlacquered Lacquered (S&B)
Ctg. H 6 0 P Z E Ctg. H 6 0 P Z E
571 X X X x x x 981 X X X x
572 X X x x x 982 x x x
573 X x x 983 x x x x
574 x x x x 984 GC x x x x
575 x x x 985 GC x x x
576 x x x x 986 GC X X x
577 x x x x x x 987 x X x x x
578 X X x x 988 X X X x x x
579 x x x 989 GC X X x x
580 x x x x x x 990 x
Taurus
Unlacquered Lacquered (S&B)
Ctg. T 1 3 A 5 L Ctg. T 1 3 A 5 L
571 X x x x x x 911 x x x x x x
572 X X x x x x 912 x x x x
573 X X x x x 913 x x x x x x
574 X X x x x x 914 x x x x x x
575 x x x x x 915 x x x x x x
576 X x x x x 916 x x x x x x
577 X X x x x x 917 x x x x x
578 X X x x x x 918 x x x x x x
579 X X x x x x 919 x x x x x
580 X x x x x 920 x x x x
Sig Sauer
Unlacquered Lacquered (S&B)
Ctg. S 2 3 S X 7 Ctg. S 2 3 S X 7
191 X X X X X X 911 X x x x
192 X X X X X X 912 X X X X X X
193 X X X X X X 913 x X X X X X
194 X X X X X X 914 x x x x x x
195 X X X X X X 915 X X x x X X
 82

Table III (Continued)

Unlacquered Lacquered (S&B)

Ctg. S 2 3 S X 7 Ctg. S 2 3 S X 7

196 X X X X X X 916 X X X X X X
197 X X X X X x 917 X X x x x x
198 X GC X X GC X 918 x x x x x x
199 X X GC x GC GC 919 x x x x x
200 X X X X X X 920 x x x x x

Discussion

A summary of the results obtained in this study is shown in Table IV for the 26 cartridges

examined. As seen in the table, simply using the SEM as an evaluation tool measurably increased the

number of visible alpha-numerics, irrespective of the gear code. In fact, the gear code was only seen

to increase the number of identifiable alpha-numerics in a single instance, although it could be argued

perhaps that the gear code did confirm the guesses made based on SEM imaging. However, this help

must be balanced with those cases where the gear code seemed to be at odds with the visual data from

imaging (e.g. cartridges H610 and H716).

Table IV: Summary of grades from optical and SEM assessments.

Sig Sauer
Ctg. Number BrandOptical grade SEM grade Gear Code Total Identifiers
10 Brown Bear C1 C3 C5 C6
24 Brown Bear C2 C6 C6 C6
707 CorBon C0 C2 C2 C3
908 S&B C0 C2 C1 C2
909 S&B C5 C6 C2 C6
965 S&B C4 C4 C0 C4
985 S&B C4 C6 C0 C6
Taurus
Ctg. Number Brand Optical grade SEM grade Gear Code Total Identifiers
101 DAG C5 C6 C0 C6
233 Amer. Eagle C2 C3 C2 C3
275 Amer. Eagle C5 C5 C0 C5
 83

Table IV (Continued)

282 Amer. Eagle C2 C3 C0 C3

296 Amer. Eagle C1 C2 C1 C2
944 S&B C2 C6 C0 C6
945 S&B C3 C4 C0 C4
Hi-Point
Ctg. Number Brand Optical grade SEM grade Gear Code Total Identifiers
164 DAG C1 C6 C1 C6
218 Amer. Eagle C2 C3 C0 C3
420 PMC C2 C6 C3 C6
520 Silver Bear C0 C3 C2 C4
541 Silver Bear C1 C5 C0 C5
573 Silver Bear C2 C6 C3 C6
610 CCI Blazer C1 C5 C1 C5
716 CorBon C2 C5 C2 C6
880 Independence C2 C6 C3 C6
910 S&B C2 C4 C0 C4
937 S&B C1 C2 C0 C2
974 S&B C0 C1 C1 C1

As mentioned, most of the improvement in scores came not by use of the gear code but by the

improved imaging characteristics of the SEM. While this is encouraging, the SEM may not be

readily available to forensic examiners. It is interesting that the majority of the gear codes tended to

mark well in the initial sectors (e.g. up to the first three identifiers, especially with the Hi-Point and

Taurus) but less well in the remaining ones. Unfortunately, the code in the missing regions often

corresponded to the missing alpha-numerics, so the gear code rarely was able to clarify any

uncertainties in the last three alpha-numerics. For the evaluated cartridges, the clarity of the

transferred alpha-numerics as a function of position was examined to see if any trends existed that

might guide placement of the gear code in such a manner as to better allow reconstruction of the

missing alpha-numerics using the gear code, such as possibly a reversal of the gear code to run

counter-clockwise rather than clockwise. While no clear trends were discernible from the limited

amount of data obtained this remains an area worth investigating. If the firing / ejection mechanism

of a particular handgun consistently produces a smearing of the alpha-numeric in a certain area, it

might be possible to design the gear code such as to provide redundancy in an area that statistically
 84

provides good transfer clarity. Other possible areas to study include the effect of the shape of the

firing pin as pertaining to size and radius of curvature at the tip; the average force exerted on the

firing pin; and the effect of striker vs. hammer, etc. All these variables can be expected to play a role

in the quality of transfer. While the cycle of fire protocol used to place the unique identifier on the

firing pin ensured that the best possible transfer was achieved for the given set of conditions

associated with that particular firearm, it does not identify the particular variable and/or define the

optimum parameters in firearm design / manufacture that would ensure the best transfer of the

alphanumeric and gear code. Such a study might also be worthwhile.

It should also be noted that the quality of the gear code transfer was not examined in

cartridges that previously had received a rating of C6. Since the gear code is meant to be a backup for

those who might seek to remove the alpha-numeric code at the tip, this also is an area of further study.

A study of determining the identifier based solely on the gear code is planned for the future.

Conclusions

This study investigated the transfer of a digital circumferential gear code placed on the end of

the firing pin of three different firearms. As seen in a previous study that only evaluated the quality

of alpha-numeric transfer, this study showed that gear code transfer was not universal. However, with

partial information from both the identifiers and the gear code, some identification can be made,

especially when the information discernible from the gear code does not overlap that provided by the

readable alpha-numerics. That being said, a full gear code appears to be rare and dependent on the

weapon that made the impression. Also problematic was the gear code appeared to be at odds with

the alpha-numeric in certain instances. While the latter appears to be related to double strikes, which

can be recognized by an examiner, the former problem requires more study concerning exactly what
 85

combination of type of mechanism / pin / action / minimum pressure etc. is most likely to produce

good transfer.

While large pieces of the gear code did not transfer in many cases, SEM evaluation greatly

improved the clarity ratings for nearly all selected cartridges. This suggests that simply equipping

labs with small, relatively inexpensive SEMs (simple models can be had for ≈ 50K) may be more cost

effective than extensive research and development of improved gear code transfer.

References

1. Howitt, D., Tulleners, F. A., Beddow, M. T., “What Micro Serialized Firing Pins Can Add To

Firearms Identification,” UC Davis/ California Policy Research Center.

2. Ohar, O.P., Lizotte, T. E., “Extracting Ballistic Forensic Intelligence: Microstamped Firearms

Deliver Data for Illegal Firearm Traffic Mapping – Technology,” Proc. Of SPIE Vol. 7434,

743416, (2009).

3. Chumbley, L. S., Kreiser, J., Lizotte, T., Ohar, O., Grieve, T., King, B., Eisenmann D.,

“Clarity of microstamped identifiers as a function of primer hardness and type of firearm

action,” AFTE Journal, Vol. 44, No. 2, May 2012, pp. 145-155.

4. Krivosta, G., "NanoTag Markings from Another Perspective," AFTE Journal, Vol. 38, No. 1,

2006, pp. 41-47.

5. Cork, D. L., Rolph, J. E., Meieran, E. S., Petrie, C. V., “Microstamping: alternative

technology for tracing to point of sale,” Ballistic Imaging, National Academies Press, 2008,

pp. 255-271.

6. Shapiro, O., “Empire state: Kills CoBIS; threatens microstamping (again),” Shooting Sports

Retailer, Vol. 30, No. 3, May/Jun. 2012, pp. 18.

7. “Officials in R.I. advocate gun microstamps,” Rhode Island Lawyers Weekly, Dec. 3, 2007.
 86

8. Shapiro, O., “Microstamping! Coming to a crime lab near you?,” Shooting Sports Retailer,

Vol. 28, No. 4, Jul./Aug. 2010, pp. 20-24, 71.

 87

CHAPTER 5. SUMMARY AND CONCLUSIONS

The comparison of plier data with the statistical algorithm revealed the algorithm is more

robust than previously thought. Using parameters that were successful in evaluating evenly striated

screwdriver marks proved inconclusive when applied to the quasi striated marks of the pliers. Further

experiments varying the window size ratios proved more successful in separating the known match

and known non-match comparisons. Some incorrect match identifications were made by the

algorithm and were termed “opposite end” matches. Future work with this algorithm will include a

feature to detect when this has occurred and alert the user.

From the microstamp evaluation study, three different guns equipped with microstamped

firing pins were used to fire 1000 cartridges each. Ten different brands of ammunition were used to

examine the difference between primer hardness and the transfer quality of the microstamp. The fired

cartridges were evaluated for clarity of transfer by a trained examiner and a novice using a

stereomicroscope. Some differences in clarity were observed, which could not be attributed to the

examined variables: primer material, primer hardness, Boxer vs. Berdan primers, or cartridge

material. However, a noticeable drop in clarity ratings occurred with the lacquered cartridges, as the

lacquer interfered with the microstamp transfer. The most notable difference in quality of transfer

occurred when comparing the weapons. The Sig Sauer had the best transfer of both alphanumeric and

gear code, while the Hi-Point had the worst alphanumeric transfer but more gear code present than the

Taurus.

While the microstamp study primarily evaluated the alphanumeric identifier transfer, the gear

code study sought to assess the quality and legibility of the circumferential code surrounding the

alphanumerics on the microstamped cartridges. From this examination, the gear code transfer was not

universal. However, partial information from the gear code can be acquired. Complete gear code

transfer appears to be rare and entirely dependent on the weapon that made the microstamped
 88

impression. The Sig Sauer had the best and most complete gear code transfer, while the Taurus gear

code was sparse. Additionally, in nearly all of the examined cases, SEM evaluation improved the

clarity ratings of the alphanumeric transfer.

 89

ACKNOWLEDGMENTS

Special thanks to Professors Scott Chumbley, Max Morris, Larry Genalo, and Song Zhang for

their contributions and guidance. Thanks also go to fellow graduate student Laura Ekstrand, retired

toolmark examiner Jim Kreiser, and Adam Froeschl of Wilde Tool Co., Inc. A final thank you is

extended to my family and friends for their support.