All ETDs from UAB

Advisory Committee Chair

Elizabeth Gardner

Advisory Committee Members

Erica Lawton

Jason Linville

Document Type


Date of Award


Degree Name by School

Master of Science (MS) College of Arts and Sciences


Individual characteristics, the marks that are unique to the tool that created them, are the basis of the comparative science of Firearm and Tool Mark Identification. The individual characteristics are created by microscopic imperfections or burrs in the firearm left during the various machining operations used in its manufacture. The depth, contour, and relative spatial orientation of the resulting striations left on the fired bullets are the subject of microscopic comparison to establish an association with or elimination from a particular firearm. Individual characteristics can be created from the combination of a multitude of factors. In this study, four aspects of firearms examination were investigated: • The origin of individual characteristics and the survivability of reaming marks, • The error rate for bullet comparisons using ten consecutively manufactured barrels, • The effect of damage to bullets on the error rate, • Reproducibility of bullet comparisons among different examiners. It has long been assumed that striae on fired bullets resulted from the last tool to contact the barrel; however, little has been done to prove this assertion. To establish that reaming marks in the barrel survive button rifling, ten barrels were cast after the reaming operation and the reaming marks photographed at 10x total magnification. The process was repeated after button rifling. Comparisons of pre-rifling to post-rifling casts show the reaming marks survive the button rifling process. To establish which area of the barrel imparts the individual characteristics, the bore of one barrel was sanded at increments of one inch, from chamber to muzzle. After each section was sanded, the barrel was test fired and the bullets collected for comparison to a pre-sanding reference bullet. The bullet comparisons indicate that sanding the imperfections at the muzzle created the most drastic change in the markings on the bullets. Removing the reaming marks on the surface of the lands and the corresponding change in individual characteristics deposited on the bullet indicates that pre-rifling processes, such as reaming, have the most impact on individual characteristics. Consecutively manufactured barrels have the highest potential for carryover of striations from one barrel to the next. Previous studies have found examiner error rates to be substantially low; however, little work has been done that meets the rigger of the black box, open set, with double blind elements methodology that has been incorporated into this study. Bullets fired from consecutively reamed and button rifled barrels were used to make and distribute test kits to firearm examiners for comparison. The false identification rate was 0.455% +/- 0.139 (1/220) while the false elimination rate was 1.82% +/- 0.555 (8/440), both at the 95% confidence interval. The total error rate was determined to be 1.36% +/- 0.414 (9/660) for 660 total comparisons. This confirms that trained examiners can determine which bullets were fired in the corresponding consecutively manufactured barrel when they are not aware that the comparisons are for consecutively manufactured barrels, not all unknowns can be matched to a known, and the comparisons include damaged bullets. Damage is often incurred on bullets, distorting or even destroying the markings left by the barrel of the firearm and making comparisons difficult. Damaged bullets were included in this study to determine if they negatively affected comparison error rates and to establish an uncertainty of measurement for the base, land impression, and groove impression measurements. The total error rate for undamaged bullets was 1.11% +/- 0.338 (5/451) with the damaged error rate at 1.91% +/- 0.582 (4/209), both at the 95% confidence interval. The mean base diameter for damaged bullets was determined to be 0.351 inches while the mean for undamaged bullets was 0.352 inches. The uncertainty of measurement for damaged bullets was +/- 0.013 inches from the mean while undamaged bullets deviated +/- 0.007 inches from the mean. This shows that damaged bullets have almost twice as much error and uncertainty of measurement when compared to undamaged bullets, while still remaining relatively low overall. Land (LIMP) and Groove (GIMP) impression widths are a class characteristic that can be used for elimination if sufficient difference is observed. Thirty-seven participants were asked to measure the width of the LIMP and GIMP impressions of 1,110 full metal jacket (FMJ) bullets. The uncertainty of measurement (UoM) for damaged and undamaged bullet LIMP measurements was determined to be 0.056 +/- 0.006 inches and 0.056 +/- 0.004 inches, respectively. Damage to bullets has a negative impact on the accuracy of LIMP and GIMP measurements and widens the associated UoM. The class elimination threshold for LIMP and GIMP measurements was found to be +/- 0.005 inches from the mean with 91.3% (63/69) of responses greater than or equal to the threshold eliminated. Ten of the test kits were evaluated by 20 participants with each test kit being examined by two participants for a total of 300 comparisons. Two examiners analyzing the same test kit came to the same result in 77% (232/300) of the comparisons. Of the 68 differing conclusions between the two examiners for the same test kit, 76% (52/68) reported identification versus inconclusive, 21% (14/68) reported elimination versus inconclusive, and 3% (2/68) reported identification versus false elimination on individual characteristics. This shows that results are reproducible 77% of the time between two examiners.



To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.