8127/frame/fm

b.
Nonbiological Evidence
Because physical match evidence seems so self-evident, the basis for assuming
an individualization of two pieces of an object to each other are not often
discussed. Nevertheless, the random nature of the generation of physical
match evidence predicts its individualizing potential. As we will discuss in
the next chapter, the understanding of how matter divides is so fundamental
that it deserves a place as a tenet of forensic science. The apparent obviousness
of physical matching seems to have precluded much mention of notable cases
involving its use in historical reviews. One of the few examples tells of John
Toms of Lancaster, England, who was convicted of murder based on physical
match evidence (Kind and Overman, 1972). The piece of newsprint that he
had used as wadding in the murder pistol was judged exactly complementary
to the piece remaining in his pocket. Even with physical matching, we are
not without the inescapable fingerprint comparison. In 1949, O’Hara and
Osterburg (1949) wrote of a glass fracture: “The elements in this case which
correspond to the characteristics of a fingerprint are the points of the curves
when they are regarded as being plotted in some system of coordinates.”
Like fingerprints, toolmarks, including their most prominent manifesta-
tion as bullet markings, have always been accepted as inherently individual-
izing evidence. With the exception of Balthazard’s first attempts to
systematize firearms evidence with respect to marks conferred on a cartridge
case by the moving parts of a gun (Thorwald, 1964), the emphasis seems to
* Individualizing potential should not be confused with certain individualization. Whether
or not a piece of evidence may be individualized using DNA depends on many factors,
including the state of the sample (quality and quantity), which loci are analyzed, how many
loci are analyzed, and the population frequencies of the alleles involved. And, as with any
forensic analysis, the veracity of the results rests on the reliability of the analysis and
thoughtful interpretation of the data.
8127/frame/ch02 Page 52 Friday, July 21, 2000 11:51 AM

---

The Evolution of Forensic Science
53
have been directed toward perfecting the mechanics of detection and visu-
alization. From the first documented experience of Henry Goddard in 1835,
through the first attempts at photomicrography by Jesrich in 1898, to Waite
and Goddard’s adaptation of the comparison microscope for bullet analysis
in about 1920 (Thorwald, 1964), no one worried much about the basis for
a conclusion of individuality. Like fingerprints, the markings left on a bullet
by both the barrel and the various moving parts of the weapon are complex
and irregular. Consequently, individualizing potential is assumed, and a rich
oral history is likely to exist around its justification. However, like fingerprint
examiners, the opinions of toolmark and firearm examiners have been
accepted almost without challenge regarding the individualization of an
impression to a tool or a bullet to a gun. A landmark exception to this trend
is presented in a paper published in 
The American Police Journal
in 1930 by
Luke May, in which he presents his methods for toolmark comparison and
some case examples. In 
Wash. v. Clark
he individualized a knife to a cut in a
fir tree. May presents the following calculation to substantiate his conclusion:
Considering only the major marks on this cut, it can be mathematically
determined that no other blade in the world would make a cut like this.
Invoking the law of probabilities, using the algebraic formula for determin-
ing combinations and permutations, with only one-third of the marks here
shown as factors, there would be only “one” chance of there being another
blade exactly like this if everyone of the hundred million people in the
United States had six hundred and fifty quadrillion knives each. Using all
of the marks, and the factors of depth, width, shape, etc., it would be carried
to infinity.
Unfortunately, he gives us no clue as to any statistical studies on which his
numbers might be based, nor does he provide the details of the mathematical
determination or his algebraic formula for determining combinations and
permutations. He also fails to inform us regarding his assumptions about
independence. One is left to wonder just how his rather oddly expressed
frequency of the [population of the U.S.] 
×
[six hundred and fifty quadrillion
knives] has been determined.
It is not as if no work has been performed since May’s time to address
these questions. The study mentioned previously by Biasotti (1959) and
numerous other analyses of consecutively manufactured gun barrels and
other tools (reviewed in Murdock and Biasotti, 1997; Nichols, 1997) are
commendable efforts. Although isolated pockets of work continue (Tulleners
and Giusta, 1998), the greater toolmark and firearms community has shown
relatively little interest in exploiting and continuing this work to provide a
more quantitative presentation of their data.
8127/frame/ch02 Page 53 Friday, July 21, 2000 11:51 AM

---

54
Principles and Practice of Criminalistics
Interestingly, Kirk, in the second edition of 
Crime Investigation
(1974),
mentions that increased mass-production methods were generating tools that
were initially indistinguishable by the impressions they made. He hastens to
add that this situation is quickly “remedied” by use of the tool; nevertheless,
it is one of the few admissions of a limitation to toolmark analysis. And like
fingerprint examiners, firearms and toolmark examiners may soon be called
upon to substantiate their opinions based on experience with scientific stud-
ies and provide statistics supported by databases.