the crime-solving promise of ballistic identification

The Crime-Solving Promise of Ballistic Identification

�

�

A Report in the Closing Illegal Gun Markets Series byThe Educational Fund to Stop Gun Violence

CRACKING THE CASECRACKING THE CASE

CRACKING THE CASEThe Crime-Solving Promise of Ballistic Identification

Our Mission The Educational Fund to Stop

Gun Violence was founded in

1978 as an educational nonprofit

dedicated to ending gun violence

by fostering effective community

and national action.

aw enforcement has

used ballistic identifi-

cation to solve the

toughest of crimes. In

2002, ballistic identi-

fication was instru-

mental in helping

Montgomery County Police Chief

Charles Moose and a team of

investigators track down the

Washington DC-area snipers. But

forward-thinking policymakers

who understand both the current

state of the art and the future of

ballistic identification technology

can make ballistics identification

an even more powerful tool for

law enforcement.

This report by The Educational

Fund to Stop Gun Violence is

designed to provide policymak-

ers with the information they

need to better arm law enforce-

ment in the war on crime. Based

on thoughtful analysis of exten-

sive information from the leading

ballistics technology experts, this

report will inform policymakers,

opinion leaders, advocates and

the media about the promise of

ballistic identification, now and

in the future.

Even as this report goes to press,

ballistic identification technology

continues to evolve. At the train-

ing seminar put on by the

Association of Firearm and Tool

Mark Examiners in late May

2004, exciting new breakthroughs

were announced. A powerful

three-dimensional imaging sys-

tem could be available later this

year and improved analytical

software will be released in 2005.

Finally, I want to acknowledge

the Educational Fund staffers

who worked on this study, espe-

cially Policy Director Eric

Gorovitz for taking the lead in

organizing, researching, and

writing the report.

This work is dedicated to

America’s firearm and tool mark

examiners, whose single-minded

dedication to the pursuit of truth

too often goes unrecognized.

A Message fromJoshua HorwitzExecutive Director ofThe Educational Fundto Stop Gun Violence

L

CRACKING THE CASE The Crime-Solving Promise of Ballistic Identification

A Report in the Closing Illegal Gun Markets Seriesby The Educational Fund to Stop Gun Violence

June 2004 �

�

CONTENTS

� Introduction 1

� Section I: Ballistic Identification Basics 3

Ballistic ‘Identification’ or Ballistic ‘Fingerprint’?

Firearm and Ammunition Basics

What Happens When a Gun Is Fired?

Ballistic Fingerprints: Marks on Cartridge Cases and Bullets

Forensic Use of Ballistic Fingerprints

� Section II: Today’s Ballistic Identification Technology 8

The Emergence of Computerized Databases

IBIS and NIBIN

IBIS and Reference Databases for New Handguns

Criticisms and Responses

� Section III: The Next Generation: Microstamping 16

Microstamping Technology

Implementation of Microstamping

� Section IV: Recommendations 18

� Ballistic Identification Glossary 21

� Endnotes 23

INTRODUCTION In October 2002, the world watched helplessly as

the Washington, D.C., area experienced a suddenrash of gun violence. Seven shootings, six of whichwere fatal, occurred in just two days in various D.C.-area suburbs. Investigators recovered bullets fromsome of the shootings and examined them under amicroscope. Unique, microscopic markings indicat-ed that the bullets had been fired from the same gun.A sniper was on the loose.

As police searched for clues to the sniper’s identi-ty, terrified residents altered their daily routines,avoiding gas stations, shopping mall parking lots andother public places generally regarded as safe.Charles Moose, then Montgomery County, Md.,police chief and leader of the investigation, madenumerous public appeals for information, some-times even speaking directly to the sniper.Responding to information from several eyewitness-es, police throughout the region stopped andsearched hundreds of white vans, tying up traffic onthe area’s major highways. Professional and amateurprofilers spun countless theories about the sniper’sidentity and motives. Yet the shootings continuedand the death toll mounted.

By the time police arrested two suspects threeweeks after the killing spree, 10 area citizens hadbeen murdered. Three more had been wounded,including a 13-year-old boy shot outside of hisschool. The sniper investigation occupied the fullattention of law enforcement agencies up and downthe East Coast and cost taxpayers millions of dollars.

In retrospect, the microscopic markings on thebullets that police recovered early in the investiga-tion provided strong evidence about the killers’ iden-tities. Those markings, and similar markings left oncartridge cases that were recovered later, constituteda unique “ballistic fingerprint” of the specific gun thesnipers used. If police could have identified themake, model and serial number of the snipers’ gunfrom its ballistic fingerprint, they could have usedthat information to access the existing crime guntrace system. That, in turn, would have led themdirectly to the Tacoma, Wash., gun store where thesnipers had acquired the Bushmaster XM15 assaultrifle they used to murder 10 people in the D.C. region

CRACKING THE CASE: The Crime-Solving Promise of Ballistic Identification

1

Montgomery County (MD) PoliceChief Charles Moose speaks to thepress during the October, 2002sniper shootings.

while terrorizing millions more.Unfortunately, because there is no existing,comprehensive system linking ballistic fin-gerprints to the guns that produced them,the evidence police recovered so quicklycould not help them identify the killers.

“Ballistic identification” is the science ofusing a ballistic fingerprint to identify thespecific firearm used in a shooting. A com-prehensive ballistic identification systemwould connect a bullet or cartridge caserecovered at a crime scene directly to themake, model and serial number of the gunfrom which that bullet or cartridge case wasfired. In effect, comprehensive ballistic iden-tification allows police to trace a gun evenbefore they recover that gun. It is impossibleto know whether ballistic identification ofthe snipers’ gun would have stopped theattacks any sooner, but police would haveconsidered it vital information during theinvestigation.

This Resource Guide explains the sciencebehind ballistic identification, answers themost common questions about the status ofthe technology, details how federal and statelaw enforcement deploy the technology, andrecommends how to best use current andemerging technologies to help law enforce-ment solve gun crimes.

The Resource Guide has four sections.Section I provides basic background infor-mation about the mechanics and terminolo-gy of ballistic identification. Section II

describes the early history of ballistic identi-fication and the technology in use today, andresponds to some of the most common ques-tions about existing systems. Section IIIintroduces an emerging technology that willgreatly enhance the power of ballistic identi-fication. Section IV presents seven recom-mendations for promoting the ability of lawenforcement to use ballistic identification tosolve gun crimes.

In addition to this Resource Guide, wehave developed a model BallisticIdentification Law suitable for use at anylevel of government. We have also producedsupporting materials that we hope will behelpful to advocates, policymakers, opinionleaders, the media and anyone else interest-ed in ballistic identification. All of thesematerials are available on our Web site,www.efsgv.org.

Finally, this Resource Guide was madepossible by generous support from theRichard & Rhoda Goldman Fund. We alsogratefully thank the following individuals andorganizations for helping us to understandthe history, technology and utility of ballisticidentification: Jennifer Budden and RahmMahanand at the NIBIN Program at the fed-eral Bureau of Alcohol, Tobacco, Firearmsand Explosives; Pete Gagliardi, Cybele Daleyand Michael McLean at Forensic Technology,Inc.; Todd Lizotte and Orest Ohar at IDDynamics (formerly NanoVia, Inc.); JosephKopera at the Maryland State Police CrimeLaboratory; Detective Mike Bonciemino atthe New York Police Department; DoreenHudson and Bill Moore at the Los AngelesPolice Department Criminalistics Laboratory;John Rush at the California CriminalisticsInstitute of the California Department ofJustice; and Lucien Haag of Forensic ScienceServices, Inc.


omprehensive ballisticidentification allows policeto trace a gun even beforethey recover that gun.

C

2 � Introduction

Ballistic ’Identification’ or Ballistic ‘Fingerprint’?Every firearm leaves unique, reproducible markings on each

bullet and cartridge case it fires. These markings function as a“ballistic fingerprint.” Today, ballistic fingerprints are inadver-tent but unavoidable byproducts of the way most guns aremade. In the near future, however, identifying marks will beintentionally designed into each firearm. “Ballistic identifica-tion” is the use of a ballistic fingerprint, whether accidental orintentional, to identify the specific gun that fired a recoveredbullet or cartridge case.

Understanding how ballistic identification works requiresbasic knowledge of what happens inside a gun when a cartridgeis fired. That, in turn, requires knowledge of some basic termi-nology. This section defines essential terms and explains theprocesses that enable ballistic identification.

Firearm and Ammunition BasicsBallistic fingerprints result from

the interaction between firedammunition, which has severalcomponents, and the interior of thegun, which has many parts. Webegin by defining the componentsof ammunition and the relevantparts of a gun.

A cartridge or round of ammu-nition includes four components.The primer is a percussion-sensi-tive chemical mixture embedded inthe base of the cartridge. The pow-der is a highly combustible com-pound that releases a tremendousamount of hot gas when burned.The powder sits inside the cartridgecase, separated from the primer by a perforated barrier. Thebullet, which sits directly over the powder, is the projectile thatleaves the gun when the cartridge is fired. The cartridge case,usually made of brass, contains the bullet, powder and primer.(Figure 1-1)

A cartridge of ammunition fits into the firing chamber of agun. This chamber is typically made of steel strong enough towithstand the extreme forces produced when the cartridge isdischarged. The barrel, which is slightly smaller in diameterthan the bullet and also made of very strong steel, begins at the

3

Section I

Ballistic Identification Basics

Figure 1-1: Cutaway of completecartridge showing the fourcomponents of ammunition.

front of the firing chamber. The barrel of ahandgun or rifle contains grooves, calledrifling, which spiral along the barrel’s interi-or surface. The spaces between the groovesare called lands. Shotguns typically do nothave rifled barrels. (Figure 1-2)

The rear of the firing chamber is a flat sur-face called the breech face. A firing pin sitsjust behind the breech face. When the triggeris pulled, the firing pin moves forwardthrough a hole in the breech face, striking theprimer cap and causing the cartridge to dis-charge. After the cartridge discharges from a

semi-automatic firearm, the extractor pullsthe just-fired cartridge case out of the firingchamber, and the ejector pushes the emptycase out of the gun. In a revolver, the car-tridge case remains in the firing chamber.(Figure 1-3)

What Happens When a Gun Is Fired?The discharge of ammunition from a

firearm involves a rapid series of mechanicalinteractions between the gun and the ammu-nition. These interactions occur under highpressure and with great force.

The series of interactions begins when thefiring pin strikes the primer cap. The impactof the firing pin on the primer cap compress-es the chemicals inside the cap, igniting theprimer. The burning primer then ignites thepowder inside the cartridge case. As the pow-der burns, it releases rapidly expanding gasesin all directions.

These expanding gases propel the bulletout of the cartridge case and down the barrel.At the same time, the gases press the walls ofthe cartridge case against the interior of thefiring chamber, and force the base of the car-tridge case back against the breech face.

In a semi-automatic firearm, the rearwardpressure of the cartridge case against thebreech face causes the breech face, which ispart of a moveable component called a slide,to begin moving backward. The empty car-tridge case, held to the slide by the extractor,moves back with the slide until the cartridgecase collides with the ejector. The impact ofthe moving cartridge case against the rigidejector pushes the cartridge case out of thefirearm through a hole in the slide (called aport).

Ballistic Fingerprints: Marks onCartridge Cases and Bullets

� Cartridge Case Fingerprints

A fired cartridge case bears several dis-tinctive marks, caused by different compo-nents of the firearm. (Figure 1-4)

The most prominent mark, the firing pinimpression, is created when the firing pinstrikes the primer. To the naked eye, the fir-ing pin impression looks like a small dimplein the primer cap (on center-fire ammuni-tion) or the rim (on rim-fire ammunition).Under a microscope, however, the firing pinimpression contains a unique imprint of thesurface imperfections present on the tip of


4 � Section I: Ballistic Identification Basics

Figure 1-2: Side view of a hand-gun with key internalcomponents identified.

Figure 1-3: Top view of a handgun with key components identified.

every firing pin. Other features, like the sizeand shape of the impression, provide addi-tional information.

Breech marks, which typically cannot beseen with the naked eye, result from theimpact of the cartridge case against thebreech face. The base of the cartridge casesurrounding the primer cap is harder thanthe primer cap and often bears stampedcharacters identifying the manufacture andcaliber of the ammunition. These characterscan interfere with any breech marks thatmight be present. The clearest marks, there-fore, appear on the surface of the primer cap,which is flat, relatively soft and typically freeof other impressions (except for the readilyidentifiable firing pin impression).

Ejector marks result from the collisionbetween the ejector and the cartridge case.Like firing pins, ejectors vary in position andshape, and have surface imperfections thatleave unique features on a cartridge case.

Additional marks can sometimes befound on the walls of a cartridge case. Thesemarks result from friction or impact betweenthe cartridge case and the interior of the fir-ing chamber or between the cartridge caseand the extractor.

� Bullet Fingerprints

The diameter of a bullet is slightly largerthan the interior diameter of a barreldesigned for a bullet of that size. As a result,when a cartridge is fired, the bullet scrapesalong the interior of the barrel. The lands,which occupy the space between the spiralrifling grooves, create gouges called impres-sions in the bullet, forcing the bullet to spin.

These impressions reflect the number oflands and the direction in which they spiral.In addition, the land impressions containmicroscopic scratches, called striations,caused by imperfections on the surface ofthe lands. (Figures 1-5 and 1-6)

Forensic examinersuse the number anddirection, and some-times the shape, ofland impressions tomake initial determi-nations about the typeof gun used to fire thebullet. One gun design,for example, may havesix lands spiraling tothe left, while anotherhas eight lands spiral-ing to the right.However, these charac-teristics, which are vis-ible to the naked eye,cannot alone identify aspecific gun becauseevery gun with thesame basic design willproduce the same pattern.

The microscopicstriations, on the otherhand, are unique. Thesurface imperfectionsthat create the stria-

tions result from machining the barrel. Thetools used to machine the barrel wear slightlywith each use and may even be sharpened,cleaned or replaced during the manufactureof a series of barrels. Consequently, each bar-rel bears distinct imperfections, some ofwhich leave marks on every bullet that trav-els through the barrel. Striations remainidentifiable even after many rounds are firedthrough the barrel.

To most of us, the ballistic fingerprint ona bullet or cartridge case looks like a set ofindecipherable nicks and dings. To a trainedforensic examiner, however, the ballistic fin-gerprint is an important piece of evidencethat can often help solve a crime.


Section I: Ballistic Identification Basics � 5

Figure 1-4: Fired cartridge case with firing pin impression and ejector mark.

Figure 1-5: Fired bullet showing landimpressions.

Figure 1-6: Magnifiedstriations in landimpressions of a firedbullet.

Forensic Use of Ballistic Fingerprints

� Early History of Ballistic Identification

The forensic potential of marks on bulletsand cartridge cases became recognized dur-ing the latter half of the 19th century, and bythe 1920s, such evidence was appearing incriminal trials around the country.1 In theseearly cases, available technology was limitedto magnifying glasses, rudimentary camerasand monocular microscopes. Even with theseprimitive tools, however, enterprising investi-gators were able to confirm or eliminatematches in several significant cases. Forensicinnovator Calvin Goddard offered ballisticidentification evidence in 1921 to help secureconvictions of accused murderers NicolaSacco and Bartolomeo Vanzetti. In 1929,using a comparison microscope adapted forthe purpose by his partner, Phillip Gravelle,

Goddard used similar techniques to absolvethe Chicago Police Department of participa-tion in the St. Valentine’s Day Massacre. (Figure 1-7)

Since then, ballistic identification hasbenefited from a long series of structural, sci-entific and technological advances. Lawenforcement agencies around the country(and the world) have established science-based crime labs, researchers have learnedmuch more about how to match bullets andcartridge cases to the guns used to fire them,and comparison microscopes have becomemore sophisticated. By the end of the 1980s,ballistic identification was an established

subspecialty of forensic science. Despite thisevolution, however, the basic tools and tech-niques have remained unchanged in theiressential details.

� Ballistic Identification Fundamentals

When both a firearm and a bullet or car-tridge case are recovered during an investiga-tion, a specially trained forensic expert calleda tool mark examiner can compare the ballis-tic fingerprint of the recovered bullet or car-tridge case with the ballistic fingerprint of asecond bullet or cartridge case test-fired fromthe recovered firearm. If the ballistic finger-print on the test-fired bullet or cartridge casematches the ballistic fingerprint on the recov-ered bullet or cartridge case, investigatorsknow that the recovered bullet or cartridgecase was also fired from the recovered gun. Aconfirmed link between a specific firearmand a bullet or cartridge case recovered froma crime scene constitutes a valuable lead,because investigators may be able to connectthe firearm to a person, who may thenbecome either a suspect or a source of infor-mation helpful to the investigation.

If investigators do not know the identity ofthe owner or possessor of the recoveredfirearm, they may be able to learn it by trac-ing the weapon. To trace a firearm, investiga-tors submit the make, model and serial num-ber of the recovered firearm to the Bureau ofAlcohol, Tobacco, Firearms and Explosives(BATFE), the federal agency with primaryresponsibility for regulating the gun industryand enforcing our national gun laws. Allmanufacturers, distributors, and licenseddealers are required by federal law to main-tain records of the acquisition and disposi-tion of each firearm they make, purchase orsell. Because federal law prohibits BATFEfrom maintaining any centralized database ofthese records, BATFE must contact the man-ufacturer of the recovered firearm, and themanufacturer must check its records todetermine who purchased the firearm.BATFE then continues down the chain of dis-tribution, requesting information from eachdistributor or dealer who bought the firearm.Eventually, BATFE identifies the retail dealer


6 � Section I: Ballistic Identification Basics

Figure 1-7: Re-enactment of the 1929 St. Valentine’s DayMassacre in Chicago, IL.

who first sold the firearm to a consumer, andwhose records should identify that buyer.Although the first retail purchaser may notnecessarily be involved in the shooting, thisinformation allows investigators to contact aperson who at some time possessed a firearmthat was involved in a crime. Firearm tracingis an extremely valuable tool for law enforce-ment agencies investigating gun crime.2

Another well-established use of ballisticfingerprints arises when investigators recoverone or more bullets or cartridge cases, but nofirearm. In these cases, the ballistic finger-print allows forensic examiners to connectshootings that might otherwise appear unre-lated. For example, examiners can comparecartridge cases recovered from two differentcrime scenes to determine whether the samefirearm was involved in both shootings.Examiners achieve these results by compar-ing ballistic fingerprints under a comparisonmicroscope, which enables side-by-sideexamination of two specimens. Investigatorsworking on the D.C.-area sniper shootings in2002, and the more recent highway snipershootings in Ohio, used this technique to linkthe shootings in each case.

Despite the long-standing effectiveness ofthese techniques, two significant limitationshave prevented law enforcement agenciesfrom making the best use of the informationballistic fingerprints provide. First, many guncrimes that could be linked through ballisticfingerprints remain unidentified because ofthe high volume of gun-related cases mosturban law enforcement agencies face. Withhundreds or thousands of shootings to inves-tigate each year, forensic examiners cannotpossibly compare the ballistic fingerprintsfrom each potentially connected pair of opencases.

One major obstacle to the routine use ofballistic fingerprint comparisons is the cum-bersome nature of manual recordkeepingsystems. The Miami-Dade Crime Laboratoryin Florida, for example, developed in the1950s an elaborate system of Rolodex cards,with the help of a specialist who had recentlyretired from Goddard’s crime lab in Chicago.The system required the examiner to fill out a

file card for each cartridge case, describingthe cartridge case and manually noting thelocation and characteristics of each mark.(Figure 1-8) To use the file in a subsequentinvestigation, the examiner had to flipthrough thousands of cards on huge Rolodexdrums looking for a possible match.3

Although this system could and did helpinvestigators solve crimes, it was difficult andtime-consuming to use.

Additionally, investigators cannot deter-mine, from the ballistic fingerprint alone,precisely which firearm was used to dis-charge the bullet or cartridge case.Investigators often have no leads other thana recovered bullet or cartridge case, andwithout a way to link ballistic fingerprints tothe guns that produce them, investigatorscannot initiate a trace and cannot identify apotential suspect. This problem was espe-cially visible early in the investigation of theD.C.-area sniper shootings when, despitelinking the shootings through ballistic finger-prints, authorities could not identify the pre-cise gun the shooter was using. Had theybeen able to do so, a routine trace wouldhave led them directly to the Washingtonstate gun dealer from whom JohnMuhammad acquired the Bushmaster XM15assault rifle that he and Lee Boyd Malvo usedin the shootings.

Fortunately, modern technology has pro-vided powerful solutions to both of theseproblems. The next two sections discussthese solutions and their potential to helplaw enforcement solve gun crimes faster andmore efficiently.


Figure 1-8: Manual ballistic fingerprint file card. (Adaptedfrom Kennington, R., The Matrix: 9mm Parabellum: AnEmpirical Study of Type Determination, 1992).

Section I: Ballistic Identification Basics � 7

LOADING & EJECTION SYSTEMMAKE & MODEL ________________________________________________

SYSTEM ______________________________________________________

SPECIAL CHARACTERISTICS ________________________

______________________________________________

______________________________________________

______________________________________________

BFM Breechface markCHM Chambering markCO CutoutDM DragmarkEJM Ejector markEXM Extractor markFPI Firing pin imprMC Magazine cutOSO One specimen onlyUA Unsupported areaUNR Unremarkable

Through most of the 20th century, ballistic identification wasan established investigative tool relying on relatively primitivetechnology: microscopes, cameras, file cards and a great deal ofpatience. Only in the last decade has ballistic identificationbecome a high-tech discipline. This section describes the cur-rent status of ballistic identification technology.

The Emergence of Computerized DatabasesIn the 1990s, computers joined comparison microscopes as

essential tools of forensic examination. With advances in digitalimaging technology and data storage capacity, forensic examin-ers envisioned a centralized database of images of bullets andcartridge cases that could be compared against a bullet or car-tridge case recovered from a crime scene. By the mid-1990s, twosuch systems emerged.4

The first system, developed by the Federal Bureau of Investi-gation (FBI), was called “Drugfire.” Drugfire used imaging soft-ware to capture, catalog and compare digital images of cartridgecases (bullets were added later). An examiner would capture animage of a recovered bullet or cartridge and compare it with

similar images from the database. Drugfire enabledthe examiner to see many images of potentialmatches on one screen, greatly speeding up theprocess. However, Drugfire did not rank the imagesby how close a match they were, leaving that deter-mination entirely to the examiner. More than 170 lawenforcement agencies nationwide participated in theDrugfire program.5

The second system, developed by the Bureau ofAlcohol, Tobacco and Firearms (ATF), was originallycalled “Ceasefire.” Like Drugfire, Ceasefire usedimaging software to capture images of the markingson bullets and included a sophisticated comparisonalgorithm that automatically identified likely match-

es. Rather than requiring the examiner to sift through dozens orhundreds of images, the computer presented the examiner witha ranked list of the most likely matches. When the ATF expandedCeasefire to include cartridge cases, it renamed the program theIntegrated Ballistics Identification System (IBIS). (Figure 2-1)

In 1997, the ATF and the FBI agreed to try to combineDrugfire and IBIS to reduce the cost and inefficiency of main-taining both systems. However, technical obstacles preventedintegrating the systems, sparking a pitched battle for suprema-cy. Adherents of the FBI’s Drugfire system preferred its imaging

8

Figure 2-1: Images of two firing pinimpressions presented for comparisonby IBIS. (Forensic Technology, Inc.)

Section II

Today’s Ballistic Identification Technology

technology and the degree of control theyretained, while supporters of ATF’s IBIS sys-tem praised its automation and speed.

After several years of wrangling, a com-promise emerged. The new system wouldadopt IBIS’s imaging technology and com-parison algorithms while relying on the FBI’stelecommunications network. Although afew devotees continued to use Drugfire ontheir own, IBIS became the standard central-ized system. The National Integrated BallisticInformation Network (NIBIN) was born.

IBIS and NIBINNIBIN’s central component is a large

database of digital images of cartridge casesand bullets recovered in criminal investiga-tions. Currently, 194 state and local lawenforcement agencies are connected toNIBIN through 228 terminals.6

� The IBIS Technology

Using a microscope attached to a com-puter, a trained IBIS technician selectsappropriate areas of a recovered bullet orcartridge case for the system to capture.7

(Figure 2-2) The IBIS software then capturesimages of those areas and creates a file con-taining the images and other demographic

information about the evidence (such as cal-iber, manufacturer, recovery location, etc.).IBIS uses a fixed, ring-shaped light to illumi-nate the evidence from all directions, ratherthan an “oblique” light, which illuminatesfrom one side. The ring light improves thecomparability of images by eliminating varia-tions in light placement, but may, in somecases, reduce the clarity of some of themarks. Recent versions of the IBIS equip-ment allow technicians to capture side-litimages as well, but the computer does notconduct comparisons on these images. Exa-miners can use the side-lit image, however,to help confirm or reject a potential match.

For a recovered bullet, IBIS scans andevaluates the land impressions at the verybase of the bullet, where the impressions aremost likely to remain useable after impact.(Figure 2-3) For a cartridge case, IBIS canscan and analyze, at the option of the techni-cian, the firing pin impression, the breechface striations on the primer cap and/or anyejector marks. An experienced techniciancan enter a cartridge case into IBIS in just acouple of minutes; bullets can take a fewminutes longer.

IBIS then compares the new image withthose already in the database. To improveefficiency, the operator can limit the searchto images with appropriate demographiccharacteristics. IBIS applies a complex algo-rithm to each image, which takes just a fewminutes, and returns a ranked list of imagesthat might be a match. Ideally, the highest-ranked potential match will be a true match,but examiners typically expect to review atleast the top 10 potential matches.

Finally, a trained forensic examiner com-pares the new image with those returned by

Section II: Today’s Ballistic Identification Technology � 9


Figure 2-2: An IBIS data acquisition station. (ForensicTechnology, Inc.)

Figure 2-3: Fired bullets mushroomed by impact; notethe visible land impressions at the base of each bullet.

IBIS to see if any are sufficiently similar tojustify direct comparison of the bullets orcartridge cases. If the examiner confirms alikely match, investigators must then acquirethe original bullet or cartridge case fromwhich the matching image was made. Thisway, the examiner can compare the actualphysical evidence under a microscope. Inother words, IBIS does not establish a “hit”on its own; only a forensic examiner candetermine the existence of an actual matchusing the physical evidence itself, rather thanthe digital images (Figure 2-4).

The primary benefit of IBIS is that the sys-tem quickly narrows the field of potentialmatches and identifies a manageable num-ber of candidates for an examiner to review.Compared to the manual systems uponwhich ballistic identification relied for thelast 70 years, IBIS works miracles.� NIBIN: A Valuable Law Enforcement Tool

NIBIN applies IBIS to a nationwide net-work of computers, linked to several regionalservers. BATFE provides the participatingagency with an IBIS terminal, which costshundreds of thousands of dollars, and con-nects the equipment to the servers on whichthe database resides. BATFE, through a con-tract with the company that makes the sys-tem, Forensic Technology, Inc. (FTI), alsomaintains the servers and the network, pro-vides technical support to participating

agencies, and pays for training of localagency personnel in the use of IBIS. Trainingoccurs at FTI’s facilities in Largo, Fla., andtakes about a week.

Each local agency provides its own staff toenter images of recovered bullets or cartridgecases from crime scenes and to evaluatepotential matches. The technician scans theevidence to create images, and submits themto the NIBIN database over the network. Thesystem adds the images to the database andcompares them to previously enteredimages.

NIBIN has been extremely valuable forlaw enforcement agencies that use it regular-ly. The database contains hundreds of thou-sands of images, and has received tens ofthousands of queries.8 In response to thesequeries, IBIS has helped investigators linkthousands of sets of cases that otherwisewould have remained unconnected, and pos-sibly unsolved. In one instance, IBIS linked13 shootings to a single 9-mm handgun.9

Consider these four examples:


10 � Section II: Today’s Ballistic Identification Technology

� An early adopter of the system, the

New York Police Department, has sub-

mitted nearly 70,000 cartridge cases

and more than 30,000 bullets to

NIBIN, and has had more than 1,100

confirmed “hits.”The detective in

charge of the program adds his own

training to FTI’s and follows a rigid

entry protocol that, he contends,

increases the chances of finding a

match.10

� One recent sutdy found that the use

of NIBIN greatly increased the pro-

ductivity of the Boston Police

Department’s Ballistics Unit, raising

six-fold the Unit’s monthly number of

ballistic matches and greatly increas-

ing the cost effectiveness of the Unit’s

gun crime investigations.11

Figure 2-4: IBIS presents possible matches for review byan examiner. (Forensic Technology, Inc.)

Each example demonstrates the lawenforcement value of NIBIN when localagencies use the system routinely. NIBIN’sability to link cases is unparalleled, but manylaw enforcement agencies—including somethat are part of the network—cannot realizethe system’s full potential because of inade-quate funding or cultural resistance toembracing the technology.

In addition, some law enforcement agen-cies have failed to recognize that everyoneloses potential leads when recovered evi-dence is not entered. If an investigator doesnot enter a recovered cartridge case into thesystem, a future investigator of anothercrime committed with the same gun will notdiscover a link between the two crimes. Inaddition, investigators of the first crime will

not learn that the weapon has been recov-ered in a later crime, and may miss anopportunity to close their case.

NIBIN has repeatedly proven its value as alaw enforcement tool. That value would beeven greater, however, if every law enforce-ment agency had the resources and inclina-tion to fully use the system.

Even at its fullest potential, however,NIBIN has one fundamental limitation as atool for ballistic identification: it only con-tains images of ballistic fingerprints fromcrime scenes. As a result, NIBIN cannot leadinvestigators directly to the specific firearmthat produced a recovered ballistic finger-print unless the weapon is eventually recov-ered. Two states have attempted to overcomethis limitation by implementing a new appli-cation of the IBIS technology: a referencedatabase of the ballistic fingerprints of newhandguns.

IBIS and Reference Databases for NewHandguns

Maryland and New York have recentlyestablished their own databases, also usingIBIS equipment and software, to captureimages of cartridge cases fired from newhandguns before their first retail sale. In the-ory, such a database would contain a com-plete file of images from every new handgunsold in the state. If one of those handguns islater used in a crime, IBIS could match theballistic fingerprint recovered from a crimescene with the previously entered referenceimage that was produced by the same gun.Unlike NIBIN, which can only make a matchafter a handgun has been used in at least twocrimes, these “reference” databases can gen-erate a lead the first time a handgun is usedin a crime. Investigators can trace the hand-gun as soon as they identify a match in thedatabase, even if they have not recovered thehandgun.

The systems in the two states are similar.In both states, the reference databasesinclude images of cartridge cases fired fromhandguns. Neither of the state databasescontains images of bullets, which are recov-



� The Los Angeles Police Department

implemented its “Walk-In Wednesday”

program in 2003 to encourage the sub-

mission of firearms evidence into NIBIN.

The new program has generated a dra-

matic shift in how detectives use NIBIN in

their investigations, and the LAPD has

found that frequent and early use of

NIBIN generates more leads faster, and

helps detectives direct their investiga-

tions more productively.12

� The Oakland, Calif., Police Department

responded to a dramatic rise in gun vio-

lence during the late 1990s in part by

entering into a unique arrangement with

the local BATFE office. Under this arrange-

ment, BATFE provides salary support for a

retired BATFE technician to enter

Oakland’s backlogged evidence into

NIBIN and to follow up on the results. This

arrangement has enabled officials in

Oakland, which lacked the resources to

support a trained technician, to tackle its

large backlog and to make numerous con-

nections among cases that otherwise

would have gone undetected.13

ered less often than cartridge cases and takelonger to enter. Nor do the state databasescontain images of cartridge cases fired fromrifles or shotguns, because the vast majorityof gun crimes are committed with handguns.These limitations reduce data entry andmaintenance costs. However, these restric-tions also prevent the systems from achiev-ing their full crime-solving potential.Recognizing the crime-solving value of IBIS,policy makers in both states have proposedexpanding their systems to include imagescaptured from long guns.

Each state requires gun manufacturers toinclude at least one test-fired cartridge case(Maryland requires two) with each handgunsold in that state. Each state also requiresmanufacturers to package and label the casesaccording to strict guidelines to ensure theproper identification of the handgun fromwhich the cases were fired. However,because manufacturers often sell their prod-ucts through distributors and cannot alwaystell which handguns will be sold in Marylandor New York, most gun manufacturersinclude test-fired cartridge cases with all ofthe handguns they ship, regardless of thedestination. Some manufacturers, however,

have not exercised sufficient care in follow-ing the law, resulting in cartridges casesbeing shipped with guns that did not firethem.14

When a dealer in Maryland or New Yorksells a new handgun, the dealer forwards thecartridge case(s) to the state police. The statepolice employ trained technicians who enterthe cartridge case(s) into the reference data-base using dedicated IBIS terminals. Eachstate also provides a procedure for capturingcartridge cases prior to delivery to the buyerif, for any reason, a dealer receives a newhandgun without the required cartridges.15

Maryland’s reference database, calledMaryland IBIS, was created in 2000 by theResponsible Gun Safety Act. All entries aresubmitted to MD-IBIS at the Maryland StatePolice Crime Lab in Pikesville, a suburb ofBaltimore. According to a report issued bythe Maryland State Police in September 2003,the system “provides a powerful weapon inlaw enforcement’s arsenal against crime.” Todate, the State Police have entered ballisticimages of more than 40,000 cartridge cases.Unfortunately, local agencies have queriedthe system fewer than 200 times, producingjust six hits.

New York calls its reference database theCombined Ballistic Identification System, orCoBIS. Created in 2001, CoBIS is similar tothe Maryland system. However, the NewYork State Police have created six regionalcenters around the state to enter images intothe database. Charles Simon of the New YorkState Division of Criminal Justice Services,which implements the program, repeatedlydeclined our requests for information aboutthe current status of CoBIS, responding that,“while we would be happy to discuss CoBISwith you in the future, we are unable to do soat this time due to ongoing projects whichhave not yet been formally announced.”According to the California Department ofJustice, however, CoBIS entered roughly21,000 images during 2002, but by March2003 had not yet produced a hit.16



In both Maryland and New York, critics have charged that the new databases are ineffective and costly. A close examination of how the programs were designed and implemented, however, suggests that their effectiveness and efficiency could greatly increase with a few relatively minor changes.

CRITICISM: Reference databases invade gun owner privacy.The National Rifle Association has falsely claimed, in opposing reference databases, that “forlawful gun owners, this scheme is national gun registration.” 17

RESPONSE: Reference databases contain no information about gun owners.Reference databases contain digital images of ballistic fingerprints, along with information identifyingthe gun that produced each image. That data includes the make, model and serial number of the gun,but no information about the purchaser. Police investigating a gun crime can use the identifyinginformation from the database to initiate a gun trace, as if they had recovered the gun, using existingrecords already maintained under current law. Reference databases contain no information about gunowners or gun buyers, and cannot function as a registration system.

CRITICISM: Reference databases have not produced results.The charge that the reference databases have been ineffective stems in part from the fact thatneither has produced many “hits.” The Maryland system has made just six matches, and theNew York system has yet to produce any.

RESPONSE 1: Few queries lead to few hits.Law enforcement agencies in both states have made few queries of the reference databases. As of theend of March 2004, all of Maryland’s law enforcement agencies had submitted a mere 177 queries toMD-IBIS, yielding six hits. Considering that the Baltimore Police Department alone investigates hun-dreds of shootings involving thousands of pieces of evidence every year, the small number of queriessuggests that Maryland’s law enforcement agencies are not using the system to its full potential.Similarly, New York City, which uses NIBIN so effectively, does not routinely query CoBIS. Experiencewith NIBIN has repeatedly demonstrated that increased use yields better results.

RESPONSE 2: Reference databases are too young to produce many hits.Reference databases, which have only existed for a few years, contain images taken from brand newhandguns—and few of these have had time to show up in crimes. Most crime guns are at least threeto five years old, so the guns that have been entered into the state databases should just now beginto show up in crimes. As time goes on and more guns that have been entered are used in crimes, thehit rate should improve.

RESPONSE 3: Interstate trafficking undermines enforcement.Each state database only contains ballistic fingerprints of handguns sold in that state, so crime gunsthat come from out of state will not show up at all.18 Maryland State Police worry some manufacturersmay even be evading the imaging requirement by funneling new handguns into Maryland throughretail dealers in other states, so that the guns are technically “used” by the time they enter Maryland’sstream of commerce. Because Maryland exempts used guns from the imaging requirement, guns dis-tributed in this fashion are excluded from the database. Expanding the system to other states andcentralizing the databases so that an agency can query more than one state at a time would greatlyimprove the likelihood of finding a match. BATFE has all of the necessary equipment and expertise,and could efficiently host a central new-gun database accessible in every state.



CRITICISMS AND RESPONSES

CRITICISM: The IBIS technology cannot handle a reference database.A feasibility study conducted by the California Department of Justice raised concerns aboutwhether an IBIS database of new handgun images could ever work.19 The California reportfocused on the fact that cartridge cases fired from different guns of the same model are morelikely to have common features, making them harder to distinguish. These shared features,called class characteristics, can narrow the focus of an investigation to a particular manufac-turer, but they could also overwhelm the ability of the IBIS algorithm to distinguish among cas-ings fired from different guns made by that manufacturer.

Because all of the cartridge cases in NIBIN were collected from crime scenes, the completedatabase contains ballistic fingerprints produced by a huge variety of guns, with relatively fewsamples of any specific model. The new databases, on the other hand, could potentially containhundreds or thousands of entries of a single model, because they capture all new guns prior tosale. The California study raised the question of whether the IBIS algorithms are sufficientlydiscriminating to distinguish among a large number of entries that share class characteristics.

RESPONSE: Effectiveness does not require perfection.The California study found that by including both firing pin and breech face impressions in the corre-lation, IBIS correctly identified the match 48 percent of the time. In another 12 percent of the trials,IBIS included the known match among the 10 most likely candidates, meaning that the examinercould find the match after reviewing 10 or fewer images. Of course, examiners could expand theirreview to include more than the top 10 candidates if a given investigation warranted the additionaltime. Although this performance is not perfect, it greatly improves on the existing manual system andquickly focuses examiners’ attention on likely matches. When other leads are not available, the abilityto focus an investigation on just a few known guns is a tremendous asset.

CRITICISM: Correlations are slow in reference databases.The California study concluded that a reference database in that state would grow to 670,000images within five years. At this size, the study stated, it could take about an hour to correlate asingle cartridge case.

RESPONSE: Better late than never.The California study included a test of the impact of increasing database size on correlation times,using databases containing hundreds of images. The study found, as a result of this test, that “correla-tion times are not a significant issue for a large database.” 20 In other words, IBIS performed the testcorrelations quickly even as the size of the database grew.

Even if it does take an hour to perform a correlation when the database gets very large, in theabsence of other leads, that is not a long wait. When the D.C.-area sniper was on the loose, hundredsof law enforcement officers spent thousands of hours searching for any clue, at tremendous publicexpense. A database capable of returning a potential lead in just an hour would have been a welcomeaddition to that investigation.

Finally, the main limiting factor for correlation times is the processing speed of the computer con-ducting the correlation. IBIS’s ability to compare images using large databases has increased dramati-cally since the system was introduced, and will continue to improve as computer processing speedsgrow even faster. When the California study tests were performed in 2001, top-end processors had aclock speed just a small fraction of that available today.

CRITICISM: Reference databases are expensive.Maryland spent $1.8 million acquiring the hardware necessary for MD-IBIS, while New Yorkspent even more. Estimates for California, which has a much bigger firearms market thanMaryland or New York, were astronomical. These costs make it difficult for states to embracenew gun databases, even if the law enforcement benefits are clear.



RESPONSE: Most of these costs are avoidable.By far, the largest contributor to the high costs of the Maryland and New York databases was the priceof new IBIS terminals. Of course, both states already had numerous IBIS terminals in place, as part ofNIBIN.21 However, according to BATFE, federal law restricts the use of NIBIN’s terminals to the captureof images of bullets and cartridge cases recovered in connection with a crime. As BATFE interprets thelaw, agencies cannot use their NIBIN terminals to enter data into a reference database because thereference databases contain the serial numbers of handguns.22 Even more significantly, agenciesinvestigating crimes cannot even use their NIBIN terminals to query the reference databases.

The Maryland State Police Crime Lab, where MD-IBIS resides, provides the clearest and mostabsurd example of the problems created by BATFE’s rule. A technician uses the lab’s NIBIN terminal tocreate a ballistic image and to query the central BATFE database. To query MD-IBIS, the technicianmust walk a few feet into an adjacent room to another, virtually identical terminal purchased sepa-rately by the state for hundreds of thousands of dollars, and spend several more minutes scanningthe same evidence again. Although the two scanned images should be virtually identical (assumingthe technician conducts the scans properly), the technician may not use the first image to access thestate reference database because the image was created on a NIBIN terminal a few feet away.

The high cost of the Maryland and New York systems can be virtually eliminated simply by author-izing the use of NIBIN terminals for both entering images into the reference databases and submit-ting images from recovered bullets and cartridge cases.

The New York Police Department, which is in the unique position of owning its own NIBIN termi-nal, has entered into a memorandum of understanding with BATFE allowing NYPD to query the statesystem, so long as the New York State Police, not the NYPD, review the results. That part of the CoBISprogram, however, has not yet begun.23

CRITICISM: Tampering and wear-and-tear defeat IBIS.Critics argue that criminals can easily defeat the IBIS technology by altering the firing pin orthe breech face. Indeed, for a time the National Rifle Association posted on its Web site a videoshowing criminals how to use common household tools to change a firearm’s ballistic finger-print. Further, critics allege that repeated use alone can cause sufficient alteration to make IBISineffective.

RESPONSE 1: Tampering is uncommon and unreliable.Criminals can alter the firing pin or breech face of a firearm, but that does not mean that IBIS will failto make a match. BATFE, commenting on the California study mentioned above, pointed out that onlytwo known instances of attempted alteration exist among the many thousands of images in theNIBIN database, and that in one of those, IBIS successfully made a match despite the alteration.24 IBIS’stechnology examines many, many data points, so a match is possible if even a portion of the originalballistic fingerprint remains intact.

Equally important, criminals are not known for their advanced planning skills. Fingerprints are easyto defeat, simply by wearing gloves, but that does not mean that police no longer dust crime sceneslooking for prints. Most criminals do not wear gloves; similarly most will not alter their firearms.

RESPONSE 2: Ballistic fingerprints are extremely durable.Forensic examiners know that ballistic fingerprints are extremely durable. Although small changesoccur each time a gun is fired, the ballistic fingerprint as a whole remains largely intact after hun-dreds, or even thousands, of discharges.

After years of incremental change, ballistic identification technology has advanced by leaps andbounds in the last decade. The IBIS technology has proven its value as a law enforcement tool, andsimilar technology can be adapted rapidly to establish a comprehensive ballistic identification systemcapable of identifying the specific gun used in a crime.

The next section discusses the future of ballistic identification.



16

Section III

The Next Generation: Microstamping

NIBIN and the state databases depend upon the transfer ofaccidental markings from the interior of a firearm to the bulletsand cartridge cases discharged from that firearm. The next gen-eration of ballistic identification technology replaces those acci-dental markings with microscopic codes intentionally insertedinto the firing chamber. This shift to intentional design meansthat the manufacturer can control what the identifying markslook like and where they are located. In addition, by duplicatingthese marks in various surfaces, manufacturers can assist lawenforcement by making it even harder for criminals to defeatballistic identification by tampering with the firearm.

Microstamping TechnologyThe basic technology of microstamping involves the use of

powerful lasers to make extremely precise, microscopic engrav-ings.25 By channeling a laser through a series of mirrors and lens-es, engineers can create a powerful beam that is a fraction of thewidth of a human hair. If the beam also passes through a stencil-like template before being reduced by the lenses, the beam cancut the image of the template into the target surface. This tech-nology has been used to produce the tiny nozzles in ink-jetprinters and to mark microchips to prevent counterfeiting.

A small company in New Hampshire called NanoVia, whichdeveloped this technology for the computer industry, beganexperimenting with using similar technology to etch a mark onthe tip of a firing pin in the early 1990s. (Figure 3-1) WhenNanoVia’s engineers put that firing pin into a pistol, fired a car-tridge and examined the cartridge case under a microscope,they found that the mark was readily visible inside the firing pinimpression on the cartridge case. Later tests determined thatthe mark remained clearly visible after many rounds were fired.(Figure 3-2)

NanoVia’s engineers found that they could put more than 20alphanumeric characters onto the tip of a firing pin. This devel-opment raised the intriguing possibility that a marked firing pincould label every cartridge with the make, model and serialnumber of the gun that fired it.

On the other hand, firing pins can be filed down or replaced,potentially undermining the effectiveness of such a system.Although most criminals are unlikely to take such measures,NanoVia went back to the drawing board, and two additionalinnovations emerged.

Figure 3-1 (Top): Tip of a firing pin marked with amicrostamp.

Figure 3-2 (Bottom): Firingpin impression left bymicrostamped firing pinon a fired cartridge case.

(ID Dynamics, LLC[NanoVia BallisticProgram Acquisition])

First, the engineers designed a redundantseries of tamper-resistant markings. Theydeveloped a coding system, analogous to abar code, comprising a series of dots thatencircle the area of the firing pin containingthe alphanumeric code. This duplicationallows the recovery of the code even if thealphanumeric figures are filed off. A thirdlevel of repetition, even farther back on thepin, could be added as well, and testing indi-cates that all three versions of the code trans-fer reliably to the cartridge case. If one triedto remove all three codes, the firing pinwould become too short to function.

Second, the engineers confronted theproblem posed by the rare, enterprisingcriminal who replaces the firing pin ratherthan altering it. Firing pins are small andinexpensive, and can be replaced withoutmuch difficulty. The breech face, however,cannot be replaced without replacing theentire slide. NanoVia’s engineers etched the

code into thebreech face,and found thatit transfersclearly to a dis-charged car-tridge. Theycould also pro-tect againstattemptedalterations byplacing thecode in multi-ple locations

on the breech face. (Figure 3-3) In addition,NanoVia tried putting the code on the wallsof the firing chamber to see if it would trans-fer to the side of the cartridge case. At thetime of this writing, that approach showspromise but has not yet produced a clear,consistent mark on the cartridge case.

Early critics of microstamping expressedconcerns that the markings would not besufficiently durable under the high stressesthat occur inside a firearm. One widelyrespected forensic scientist, Lucien Haag,was so skeptical about the durability of

microstamps that he decided to conduct hisown tests. Haag acquired marked firing pinsfrom NanoVia and tested them using gunsthat he thought were most likely to challengethe technology. His results indicate thatmarked firing pins continue to leave clearimpressions on cartridges even after hun-dreds of rounds, and even in guns that oper-ate under extremely high pressure.26 In short,microstamping appears to have a promisingfuture as a law enforcement tool.

Implementation of MicrostampingThe implementation of microstamping

requires the direct participation of firearmsmanufacturers, because they must build themicrostamp, ideally on both firing pins andfiring chambers, into firearm production.The technology for achieving both is avail-able today, but it must be adapted for use infirearm manufacturing.

Additional research and testing will identi-fy the optimal characteristics (such as the sizeand shape of the code characters) and place-ment of microstamps in the firing chamberand on the firing pin. Some of this develop-ment is under way at a company called IDDynamics, which acquired the technologyfrom NanoVia in 2003. The fundamentaltechnology is sound, however, and should beready for the market within two years.

The incorporation of microstamping alsoenables gun manufacturers to help lawenforcement in other ways. For example,microstamping invites the adoption of astandardized Firearm Identification Number(FIN) system that could ultimately replacethe current, inadequate serial number sys-tem.27 Like the Vehicle Identification Number(VIN) used for motor vehicles, the FIN couldcommunicate valuable information aboutthe firearm, such as the make, model, date ofmanufacture, etc. With a standardized codingsystem, all of this information could beincluded in the microstamp, allowing lawenforcement to extract detailed informationabout the gun directly from a recovered car-tridge case.


Section III: The Next Generation—Microstamping � 17

Figure 3-3: Breech face markedwith redundant microstamps.(ID Dynamics, LLC [NanoViaBallistic Program Acquisition])

Ballistic identification is a valuable law enforcement tool thathas not yet received the support it deserves. The leading technologyavailable today, the IBIS system, works well when it is properlyused. The emerging technology, microstamping, requires someadditional investment in development, but will, within a few years,greatly expand the power of ballistic identification. While the newtechnology continues to be developed, law enforcement should use the existing technology to the fullest extent possible. Even after microstamping becomes standard, reference databases builtaround IBIS will continue to be valuable because of the large number of guns that will be manufactured and sold between nowand then.

The following recommendations are designed to guide policy makers in

establishing ballistic identification systems that will efficiently aid law

enforcement now and in the future.

Law enforcement agencies that use NIBIN regularly derive tremen-dous benefits and efficiencies from the system. However, someagencies do not participate in the system at all, while others fail torealize its full crime solving potential. Additional resources wouldprovide local law enforcement with the support necessary fortrained staff to enter and review evidence on the most up-to-datesystems.

America’s robust illegal gun market ignores state borders. Evenworse, Maryland’s experience suggests that some manufacturersmay be intentionally funneling guns through dealers inPennsylvania in order to evade Maryland’s reference database. Suchactivity would not be possible if Pennsylvania had its own database.The more states that establish such systems, the harder it will be fortraffickers to evade detection. Even better, a single federal systemwould ensure that every firearm that enters the market would beincluded in a database, and would encourage widespread use bylaw enforcement agencies nationwide.

18

Every state, or the federal government,should establish a reference database ofballistic fingerprints for new guns.

Recommendation

2

Every law enforcement agency should haveaccess to NIBIN and the resources necessary forfull implementation of ballisitic identification.

Recommendation

1

Section IV

RECOMMENDATIONS

The greatest obstacle to efficiently using reference databases is BATFE’s refusal to allowNIBIN terminals to be used to capture and enter images or to submit queries. This ruleforces law enforcement agencies to spend scarce resources to buy duplicative equip-ment, and requires repetitive data entry that inhibits regular use. In addition, becausethe state reference databases are entirely separate, law enforcement agencies in onestate cannot access the databases in another state. If all of the state databases weremaintained on the NIBIN servers, but kept separate from NIBIN’s crime-gun database,law enforcement agencies could query multiple states if needed. Finally, centralizedhosting on the NIBIN system efficiently uses a compatible network that is already inplace, greatly reducing the cost of establishing reference databases for new guns.

Although the IBIS technology works well, microstamping provides an opportunity forgreatly easing law enforcement’s task in solving gun crimes. The gun industry, which haslong professed a desire for improved enforcement, has a central role to play in giving lawenforcement necessary tools. Microstamping is the most direct way that the gun industrycan help, because the technology allows the manufacturer to insert unique, identifyinginformation directly into the firing chamber of each firearm. However, even if develop-ment advances as quickly as it should, the first microstamped firearms will probably notshow up at crime scenes for another eight to10 years. In the interim, we should assist lawenforcement by supporting IBIS-based databases.

The federal government, through the National Institute of Justice or another appropri-ate agency, should provide funding to ensure completion of the last stages of develop-ment of microstamping technology. A federal/private partnership is the best way toensure that this technology promptly becomes available to assist law enforcement insolving gun crimes. If funding is left entirely to the market or to the gun industry, thetechnology will likely take much longer than it should to reach implementation, result-ing in a longer-than-necessary delay in the availability of microstamping as a lawenforcement tool.

Section IV: Recommendations � 19

New ballistic identification policies should encouragethe rapid development of microstamping, while con-tinuing to support IBIS-based databases.

Recommendation

4


BATFE should host the reference databases andprovide access to them through NIBIN terminals.

Recommendation

3

Federal funding should be made available tosupport the development of microstamping.

Recommendation

5

Firearm serial numbers today provide very little information about the gun.Guns made by different manufacturers may even have the same serial num-bers. Car makers, however, have used an informative, standardized numberingsystem for decades. Since 1980, cars and car parts have been labeled with aVehicle Identification Number (VIN). Much more than a simple serial number,the VIN contains detailed information about the car, identifying the countryand year of manufacture, the manufacturer, model and serial number, andsome information about equipment included on the vehicle. The VIN on amangled truck axle led investigators in the 1995 Oklahoma City bombing tobomber Timothy McVeigh. Microstamping enables the establishment of astandardized Firearm Identification Number system that will provide a greatdeal more information to law enforcement than a serial number alone.

The gun industry is in the best position to efficiently incorporate ballistic iden-tification. By building ballistic identification into the manufacturing process,whether through digital imaging or microstamping, gun companies can spreadthese costs over their entire product lines. On a per-unit basis, these costswould be minimal, especially as compared with the large public expenseincurred in the establishment of the Maryland and New York systems. Moreimportantly, shifting these costs to the industry affords gun manufacturers theopportunity to deliver on their claimed commitment to crime prevention andenforcement.

Recommendation

7The costs of creating ballistic identification systems should be borne by the gun industry and their customers, not by the general public.

The implementation of microstamping shouldinclude the establishment of a standardizedserial number system for firearms.

Recommendation

6


20 � Section IV: Recommendations

Ballistic fingerprint: A set of unique,reproducible markings left oneach fired bullet and cartridgecase by the firearm from whichthe bullet or cartridge case wasfired.

Ballistic identification: The use of aballistic fingerprint to identifythe specific, individual firearmused to fire a given bullet or car-tridge case.

Barrel: The tube on a firearmthrough which a bullet is pro-pelled when a cartridge is fired.

Breech face: The flat, vertical surfacethat forms the rear of the firingchamber of a firearm.

Breech mark: A microscopic markleft on the base of a fired car-tridge case by the surface of thebreech face. Breech marks aremost readily visible on the sur-face of the primer.

Bullet: The component of a car-tridge, usually made of lead, thatexits the firearm through thebarrel when the cartridge is fired.Some lead bullets are “jacketed”with a layer of copper alloy orother metal.

Cartridge: A unit of firearm ammu-nition containing four compo-nents: primer, powder, bullet andcartridge case.

Cartridge case: The component offirearm ammunition, usuallymade of brass, that holds theprimer, powder and bullet.

Crime gun database: A database,such as that established by theNIBIN program, containing bal-listic fingerprints of firearms

recovered in criminal investiga-tions. (See reference database)

Drugfire: A ballistic identificationsystem developed by the FederalBureau of Investigation in the1990s. Drugfire was replaced byIBIS when NIBIN was created.

Ejector: On a semi-automaticfirearm, a stationary metal bar orblock that forces a fired cartridgecase to eject from the firearm.

Ejector mark: An impression, usuallyvisible to the naked eye, left onthe base of a fired cartridge caseby the collision between the car-tridge case and the ejector.Microscopic details of an ejectormark are part of a firearm’s bal-listic fingerprint.

Extractor: On a semi-automaticfirearm, a small hook embeddedin the slide that hooks under therim of a cartridge in the firingchamber and pulls the dis-charged cartridge case out of thefiring chamber.

Firing chamber: The portion of afirearm, located at the back endof the barrel, that contains a car-tridge for firing.

Firing pin: A narrow rod which,when released by pulling thetrigger, springs forward andstrikes the primer of a cham-bered cartridge, causing the car-tridge to discharge.

Firing pin impression: An impres-sion, visible to the naked eye, lefton the primer of a fired cartridgeby the firing pin. Microscopicdetails of a firing pin impressionare part of a firearm’s ballisticfingerprint.

BallisticIdentificationGlossary

21

FTI: Forensic Technology, Inc., the companythat manufactures the IBIS system.

Groove: A spiraling channel cut into theinterior surface of the barrel of a firearmand designed to cause a bullet to spin asit leaves the barrel.

Hit: A match, confirmed by a trained foren-sic examiner, between the ballistic fin-gerprint on a recovered cartridge case orbullet and a previously-captured ballisticfingerprint.

IBIS: A computerized digital imaging systemthat captures and compares digital pho-tographs of fired bullets and cartridgecases. IBIS stands for “integrated ballisticidentification system.”

ID Dynamics: The company that acquiredNanoVia’s microstamping technology in2003.

Land: The raised surface between twogrooves on the interior of a rifled barrel.

Land impression: An impression in the sideof a bullet made by a land on the interiorof the rifled barrel through which thebullet was discharged.

Magazine: A spring-loaded ammunitionstorage and feeding device that attachesto a firearm. A magazine can be detach-able or fixed (i.e., non-detachable).

Microstamp: A microscopic array of charac-ters etched into the interior surfaces of afirearm during manufacturing, whichtransfers the characters to a cartridgecase when the cartridge is discharged.

NanoVia: The company that developedmicrostamping. (See ID Dynamics)

NIBIN: National Integrated BallisticInformation Network, operated by theBureau of Alcohol, Tobacco, Firearmsand Explosives (BATFE) and the FederalBureau of Investigation (FBI). NIBINuses the IBIS system to capture andcompare ballistic fingerprints from car-

tridge cases and bullets recovered atcrime scenes.

Port: An opening in the slide of a semi-auto-matic firearm through which a dis-charged cartridge case is ejected.

Powder: The component of firearm ammu-nition that ignites and burns when a car-tridge is fired, releasing a tremendousamount of rapidly expanding gas thatpropels the bullet along the barrel.

Primer: A percussion-sensitive chemicalmixture contained in the base of a car-tridge. The primer explodes when struckby the firing pin, igniting the powder.

Reference database: A database containingballistic fingerprints of new firearms,along with the make, model and serialnumbers of those firearms. (See crimegun database)

Rifling: A spiraling pattern of grooves on theinterior surface of the barrel of mostfirearms, designed to cause the bullet tospin as it moves down the barrel.

Slide: On a semi-automatic firearm, a mov-ing component that encases the barreland the firing chamber. When a cartridgeis discharged, the slide moves toward therear of the firearm, pulling the fired car-tridge case out of the firing chamber (seeextractor) and driving it against the ejec-tor. The slide then moves forward, forc-ing a new cartridge from the magazineinto the firing chamber.

Striation: A microscopic scratch left on thesurface of a bullet by the lands of therifled barrel of a firearm.

Tracing: An investigative technique usingexisting records to identify the first retail purchaser of a firearm that wasrecovered in connection with a criminalinvestigation.

\CRACKING THE CASE: The Crime-Solving Promise of Ballistic Identification

22 � Ballisitic Identification Glossary

1 For a more detailed summary ofthe history of ballistic identifica-tion, see Hamby, JH & Thorpe, JW,“The History of Firearm andToolmark Identification,” 31 AFTEJ. (1999).

2 In fiscal 2003 alone, BATFE’sNational Tracing Center, located inFalling Water, W.Va., processed over280,000 crime gun trace requests.(ATF Snapshot 2004, accessed onthe World Wide Web at http://www.atf.gov/about/2004snap.pdfon May 5, 2004). A new law, adopt-ed by the Unites States Congresswithout a hearing in February,2004, prohibits BATFE from releas-ing to “any person or entity” anyinformation regarding tracerequests. Accordingly, detailedinformation about the results ofthese hundreds of thousands ofinvestigations, even in aggregrate,is extremely difficult to obtain.

3 Kennington, RH, “The Matrix: 9mmParabellum—An Empirical Study ofType Determination” (1992).

4 “National Integrated BallisticsInformation Network (NIBIN) forLaw Enforcement,” CongressionalResearch Service Report forCongress, July 3, 2001 (Order codeRL31040).

5 Ibid. 6 Personal communication, Cybele

Daley, Executive Director,Government Affairs, ForensicTechnology, Inc., April 17, 2004.

7 The following discussion drawsheavily on information we receivedduring a visit to the Largo, FL,training center of ForensicTechnology, Inc., on December 2,2003, and in subsequent conversa-tions. We thank ForensicTechnology, Inc., for allowing us tovisit their facility and for educatingus about how IBIS works.

8 BATFE declined to provide moreprecise data, referring to new legal

prohibitions resulting from theJanuary, 2004 enactment of PublicLaw 108-199. Provisions of that law,collectively referred to as the“Tiahrt amendments” in honor ofthe Kansas Republican who pro-posed them, prohibit BATFE fromdisclosing to the public “the con-tents or any portion thereof of anyinformation required to be kept by[federal firearms] licensees.” P.L.108-199. That informationincludes, among other things, theidentity of gun purchasers and theserial numbers of the guns theypurchase. Although statisticalinformation about NIBIN is clearlynot subject to this prohibition,BATFE has adopted a policy oftreating every request for informa-tion with extreme caution, toensure that no information that iscovered by the Tiahrt amendmentsis disclosed to the public.Accordingly, BATFE can take weeksor months to respond to even themost routine questions that seekonly information that BATFE isclearly permitted to release.

9 Michigan State Police, NewsRelease, January 22, 2004.

10 Telephone Interview with DetectiveMike Bonciemino, NYPD, January9, 2004.

11 Braga, A, and Pierce, G., “LinkingCrime Guns: The Impact ofBallistics Imaging Technology onthe Productivity of the BostonPolice Department’s BallisticsUnit,” 49 J. Forensic Sci. No. 4,Paper ID JFS2003205, availableonline at www.astm.org (July,2004).

12 Telephone Interviews with DoreenHudson (February 2, 2004) and BillMoore (March 12, 2004), LAPDScientific Investigation Division.

13 Personal communication, OaklandPolice Chief Richard Word, June 25,2003.

Endnotes

23

14 The Maryland State Police found, for exam-ple, that cartridge cases submitted by Glock,one of the largest manufacturers of hand-guns sold in America, was shipping hand-guns with cartridge cases that had not beenfired by the handgun with which they wereshipped. Maryland State Police ForensicSciences Division, “Maryland-IBIS(Integrated Ballistics Identification System),”September, 2003. A study by an examiner inArizona also found that: 1) 11 of 15 Glocksexamined for the study were shipped withcartridge cases that had not been fired fromthe gun with which they were shipped; and2) six of the 15 Glocks were shipped with twocartridge cases that had been fired from twodifferent guns. Tew, Jon D., “IncorrectManufacturer-Supplied Test Fire CartridgeCases,” 35 Assoc. Firearms Toolmark Exam. J.195 (Spring, 2003).

15 For example, an out-of-state firearm distribu-tor may remove the cartridge cases prior todelivering the firearm to the New York dealer.

16 California Department of Justice, “Feasibilityof a California Ballistics IdentificationSystem,” January, 2003.

17 “Joint Statement [of National RifleAssociation and NRA-Institute for LegislativeAction] on Ballistic ‘Fingerprinting,’” October17, 2002, accessed on the World Wide Web athttp://www.nraila.org/News/Read/Releases.aspx?ID=2053 on May 5, 2004.

18 Trace data reported by the Bureau of Alcohol,Tobacco and Firearms (which became BATFEin 2003) indicates that 38 percent of crimeguns recovered in Baltimore and traced to alicensed gun dealer in 2000 came from out-side of Maryland. When the crime was com-mitted by a person aged 18–24, the percent-age of crime guns traced to out-of-state deal-ers jumps to 45 percent. Bureau of Alcohol,Tobacco and Firearms, “Crime Gun TraceReports, 2000—Baltimore, MD,” pp. 15–16(July, 2002). In New York City, this problem iseven more severe. A whopping 85 percent ofall traced crime guns in New York City in2000 came from out of state, with a handfulof southern states providing most of the sup-ply. For gun crimes involving young peopleaged 18–24, out-of-state sources provided93% of the crime guns. Bureau of Alcohol,

Tobacco and Firearms, “Crime Gun TraceReports, 2000—New York, NY,” pp. 15–16(July, 2002).

19 California Department of Justice, “Feasibilityof a California Ballistics IdentificationSystem,” January, 2003.

20 Ibid., Attachment A, p. 8-5.21 There are currently five terminals in

Maryland and eight in New York. CybeleDaley (FTI), personal communication, April27, 2004.

22 BATFE claims that the inclusion of serialnumbers in reference databases runs afoul ofa prohibition on the expenditure of BATFE’sfunds “in connection with consolidating orcentralizing, within the Department ofJustice, the records, or any portion thereof, ofacquisition and disposition of firearms main-tained by Federal firearms licensees.” P.L.108-199. Because licensed dealers maintainrecords that include the serial numbers offirearms, BATFE views reference databases as“consolidating or centralizing” a “portion” ofthe records maintained by licensees.

23 Telephone interview with Detective MikeBonciemino, NYPD, January 9, 2004.

24 California Department of Justice, “Feasibilityof a California Ballistics IdentificationSystem,” January, 2003, Attachment B, pp.14–15.

25 The following discussion draws heavily oninformation we received during a visit to theNew Hampshire facilities of NanoVia onDecember 3, 2003, and in subsequent con-versations. We thank NanoVia for allowing usto visit their facilities and for educating usabout microstamping.

26 Telephone interview with Lucien Haag,February 5, 2004.

27 Currently, firearm serial numbers are notstandardized. Consequently, firearms fromdifferent manufacturers can have the sameserial number, and authorities cannot tellmuch about the gun from the serial numberalone. In addition, roughly 10 percent ofcrime gun trace requests fail because ofproblems with the serial number submittedby the requesting agency. A standardizedserial number would reduce these errors.

\CRACKING THE CASE: The Crime-Solving Promise of Ballistic Identification

24 � Endnotes

The Educational Fund to Stop Gun Violence1023 15th St. NW, Ste. 600Washington DC 20005

Phone: 202-408-7560Fax: 202-408-0062

�

the crime-solving promise of ballistic identification

Documents