J Forensic Sci, July 2014, Vol. 59, No.

4
doi: 10.1111/1556-4029.12449
PAPER Available online at: onlinelibrary.wiley.com

PHYSICAL ANTHROPOLOGY

Amy Z. Mundorff,1,2 Ph.D.; Eric J. Bartelink,2 Ph.D.; and Turhon A. Murad,2 Ph.D.

Sexual Dimorphism in Finger Ridge Breadth

Measurements: A Tool for Sex Estimation from
Fingerprints

ABSTRACT: Previous research has demonstrated significant sexual dimorphism in friction ridge skin characteristics. This study uses a novel
method for measuring sexual dimorphism in finger ridge breadths to evaluate its utility as a sex estimation method from an unknown finger-
print. Beginning and ending in a valley, the width of ten parallel ridges with no obstructions or minutia was measured in a sample of 250 males
and females (N = 500). The results demonstrate statistically significant differences in ridge breadth between males and females (p < 0.001),
with classification accuracy for each digit varying from 83.2% to 89.3%. Classification accuracy for the pooled finger samples was 83.9% for
the right hand and 86.2% for the left hand, which is applicable for cases where the digit number cannot be determined. Weight, stature, and to
a lesser degree body mass index also significantly correlate with ridge breadth and account for the degree of overlap between males and
females.

KEYWORDS: forensic science, dermatoglyphics, friction ridge breadth, friction ridge impressions, fingerprints, sexual dimorphism, sex
estimation

Friction ridge skin, located on the fingertips, palms, soles, and
toes, is different from the skin covering the rest of the human
body (1). Friction ridge skin lacks hair and oil glands, although
it is covered with sweat glands (2). Ridged skin is not exclusive
to humans alone—this characteristic is found in all primates and
some marsupials. In humans, dermatoglyphs (specific character-
istics of ridged skin) begin to form around 10.5–16 weeks esti-
mated gestational age (EGA) with mature primary and secondary
ridges by week 24 EGA (3). These fingerprint patterns, visible
to the naked eye, are typically described as “continuous and dis-
continuous alternating ridges and sulci” (4, p. 10), similar to the
markings or stripes on a zebra.
Dermatoglyphics, the study of friction ridge impressions, has
been a research area in forensic science, medicine, and physical
anthropology since the late 1800s (1). Although the study of der-
matoglyphics also includes palm and sole impressions, most of
the research has been directed toward the study of fingerprints in
large part because they are accessible and easy to measure but
also because they have high heritability values and are phyloge-
netically stable (5,6). A number of studies suggest that environ-
mental factors, such as maternal health, also affect developing
fingerprints, although less so than genetic factors (6,7). Medical
studies have examined correlations between fingerprint measure-
ments and ridge patterns and chromosomal disorders (8),
1
Department of Anthropology, The University of Tennessee, 250 South
Stadium Hall, Knoxville, TN 37996-0720.
2
Department of Anthropology, California State University, 400 West 1st
Street, Chico, CA 95929-0400.
Received 4 Feb. 2013; and in revised form 28 May 2013; accepted 22
June 2013.
schizophrenia (9), male pattern baldness (10), obesity (11), and
blood groups (12), among a host of other conditions.
In forensic science, the unique structure and individuality of
fingerprints was first suggested by W. Herschel in 1859 and later
confirmed by H. Faulds in 1880. Working independently in
India and Japan, respectively, they influenced Galton’s 1892
text. However, it was J. Vucetich’s efforts in Argentina that led
to the use of fingerprints to identify unknown individuals and as
an official means of identifying criminals (13). Holt (5) notes
three important factors about fingerprints that make them valu-
able for personal identification. They are as follows: (i) charac-
teristics are not affected by age once an individual reaches
maturity; (ii) postnatally, their characteristics are also not
affected by environment; (iii) impressions are highly variable
and have patterns that are easily classified. Because fingerprint
impressions are inexpensive and easy to obtain, there are abun-
dant databases available for research.
Presently, the Integrated Automated Fingerprint Identification
System (IAFIS) is the largest fingerprint database in the world,
comprising fingerprints from criminals and terrorists along with
U.S. military and federal employees. It provides automated search
capabilities of latent and patent prints, electronic storage, and
electronic transfer of fingerprints between agencies (13). When
attempting to match a fingerprint recovered from a crime scene to
a potential suspect, that suspect must already have fingerprints on
file. However, when a latent print is recovered but is not on file,
analysis of friction ridge impressions can still provide valuable
information, such as an estimation of the individual’s sex.
Dermatoglyphic research on sexual dimorphism has focused
on pattern and metric variation among a variety of geographic
populations. For example, Acree (14) found a statistically signifi-
cant difference between male and female ridge density in two

© 2014 American Academy of Forensic Sciences 891

892 JOURNAL OF FORENSIC SCIENCES
American populations. Using the same technique, Nayak et al.
(15) and Nithan et al. (16) also found statistically significant
sexual dimorphism in ridge density among different Indian popu-
lations. These studies concluded that ridge thickness and the
depth of the ridge furrow are the two most important characteris-
tics for determining ridge density (14–16).
While numerous studies confirm significant sexual dimor-
phism within a group, some also report that the degree to which
sexual dimorphism is expressed varies between populations (17–
19). For example, studies on Turkmenians, Chuvasians, and
Eastern European, African, Middle Eastern, and Yeminite-Jews,
all demonstrate significant sexual dimorphism, but their levels of
expression vary (6, 20–26). A study comparing males and
females in two northern Indian populations reported significant
sexual dimorphism within and between both populations (19).
Kobyliansky and Micle (25) suggest this variation in the expres-
sion of dermatoglyphic sexual dimorphism may result, in part,
from prenatal environmental factors.
On the other hand, Sanna and colleagues (27) analyzed 41
quantitative dermatoglyphic variables from two Sardinian lin-
guistic groups. They focused on intersex differences within each
group and intrasex differences between the two groups. Specific
variables, such as total finger ridge count (TFRC) and ulnar
ridge count (URC), demonstrate significant differences between
males and females in both linguistic groups. While this similar
pattern of sexual dimorphism was found in both groups, intrasex
variation between groups was not significantly different (27).
When comparing African American and Euro-American males
and females, Acree (14) found no significant intergroup varia-
tion. Salazano and Benevides (28) also found no significant dif-
ferences in the levels of expressions when comparing three
Brazilian groups, each with different levels of genetic admixture
with African populations. Murad (29,30) found similar results
measuring palm prints from North Alaskan Eskimos. He docu-
mented that while male and female Eskimos differed in ridge
count, females were more alike than males across the five vil-
lages examined in the study. These data were used to discuss
mating patterns (29,30).
Numerous studies have attempted to develop sex estimation
methods based on friction ridge density. Specifically, females
are hypothesized to have finer ridge detail leading to a higher
ridge count compared to males. Calculating the mean ridge den-
sity for each subject based on the mean of all ten fingers, Acree
(14) concluded that ridge density in females is significantly
higher than for males. Using these results, Acree establishes a
“threshold of gender differentiation” (14, p. 40). Applying Ba-
yes’ theorem, it was demonstrated that a print with a ridge count
in a 5 9 5 mm2 of ≤11 is most likely male while counts ≥12
are most likely female (14). Because these results were sup-
ported for both ancestral groups in the study, Acree argued that
this could be used for a presumptive indication of sex when
examining an unknown print (14). Similarly, Gungadin suggests
“a ridge count of ≤13 ridges/25 mm2 is more likely to be of
male origin” and a “ridge count of ≥14/25 mm2 is more likely
to be of female origin” (31, p. 6). However, Gungadin notes that
these results are likely population-specific, suggesting that more
research is needed on different populations. In two other studies,
the opposite pattern was found with males having a greater num-
ber of ridge counts (32,33).
Fingerprint analyses have been applied to human identification
and population studies for decades; yet, to date, no study has
been able to provide a reliable method to predict an individual’s
sex based on fingerprint measurements (11,17,34–36). This study
examines metric sex variation in finger ridge breadth to evaluate
whether an isolated print can be accurately classified as male or
female.
Methods and Materials
Fingerprint cards were obtained from a law enforcement
agency in Butte County, California. The sample comprises the
standard ten-print cards from 500 individuals (250 males and
250 females). Each individual card included the standard ten
rolled impressions along with an additional set of ten flat finger-
prints. The rolled fingerprints were collected by inking the finger
and rolling it from one edge to another (1,2,37). The flat set
comprises four fingers simultaneously printed in a flat, pressed
manner with the thumb printed separately. Thus, each individ-
ual’s data sheet consisted of two prints for each digit. Additional
information on the print card included the individual’s sex, self-
reported race, stature recorded in feet and inches, and weight
recorded in pounds. Although fingerprints do not change their
pattern or structure during life, the fingertips tend to increase in
size as an individual grows toward maturity (38, p.139). To
avoid the bias of age-related variation, the study sample was
limited to adults, age 18 years or older.
The present study applies a novel approach for obtaining fin-
ger ridge density that differs from previous ridge counting meth-
ods. Early literature on ridge breath analysis is dominated by
studies counting ridges across a 1.0 cm line, beginning at a trira-
dius. This approach and its various modifications follow the
Henry classification method (HCM) originally developed by Sir
Francis Galton (2,5,26,38–45). This method does not allow for
every finger to be measured because pattern type must first be
considered. In order to be measured, a print must have a pattern
with a triradius (1,39–41); arch patterns do not have a triradius
and thus are not analyzed. More recent studies follow a similar
procedure, which entails counting the number of ridges in a
5 9 5 mm2 within a predetermined location, usually selected
from the upper portion of the fingertip radial boarder. This repre-
sents the number of ridges/25 mm2 (14–31).
The current study applied a method based on measuring ten
parallel ridges as a standardized increment, which avoids count-
ing ridges in a prescribed area. Pattern type was considered inde-
pendent of measurement; therefore, every finger was measured.
If a fingerprint recovered from a crime scene happens to have an
arch pattern, other established methods would not permit mea-
surement of ridge breadth because an arch pattern does not con-
tain a triradius (1,5,39–42,46). Also significant to forensic
investigations, partial prints (where the pattern may not be dis-
cernible) can be measured as long as there are ten uninterrupted
parallel ridges present.
Each individual’s demographic information recorded on the
print card was covered to ensure a blind analysis. An Olympus
SZ4045TR Zoom Stereo Microscope and a unislide by
VELMEX, Inc. (Bloomfield, NY) were used to view the finger-
prints. The ridge breadth measurements were recorded using an
Acu-Rite III Digital Readout System. The fingerprint data sheets
were viewed under a microscope to allow the analyst (AZM) to
locate ten parallel ridges with no obstructions or intervening
minutiae. Minutiae included ridge endings, bifurcations, and
short ridges. Other obstructions and anomalies such as scars
were also avoided. The ten ridges were aligned perpendicular to
the x-axis through the eyepiece. The measurement was taken at
the first location noted to have ten parallel ridges with no minu-
tiae causing a break in the perpendicular line.
 MUNDORFF ET AL. . SEXUAL DIMORPHISM IN FINGERPRINTS 893

Measurements began in a “valley,” halfway between two

Interdigit Comparison
ridges. After counting across ten ridges, the measurement ended
again mid-valley (Fig. 1). The increment machine was scrolled Patent and latent prints recovered at a crime scene are typi-
perpendicular to the orientation of the ridgelines, measuring the cally classified by digit, although this information is not always
ridges to the nearest 0.0001 mm. The majority of measurements able to be determined. To evaluate whether significant differ-
were taken from the rolled prints; however, measurements were ences exist between fingers, mean ridge breadth was compared
taken from flat impressions in cases where the rolled prints were between fingers for both hands (Table 3). The ANOVA results
unreadable. If both types of prints were unreadable across ten indicate significant interfinger differences in mean ridge breadths
ridges or if the finger was not present, no measurement was for each hand (p < 0.001). A Bonferroni post hoc test found sta-
recorded. The individual’s print measurements and associated tistically significant differences between all fingers (p < 0.01)
demographics (e.g., stature and weight) were entered into IBM except left and right digits II and III and digits IV and V.
SPSS 20.0 for the analysis (IBM SPSS. Inc., Chicago, IL). The
mean values for each hand and for all ten fingers combined were
Sex Comparisons
calculated and included as separate variables.
The main objective of this study is to evaluate whether Table 4 presents descriptive statistics for mean ridge breadth
enough sexual dimorphism exists in finger ridge breadths to by sex for each finger. For all ten digits, males have a higher
have utility as a sex estimation method from an unknown finger- mean ridge breadth compared to females (>1.0 mm). This differ-
print. To achieve this goal, male and female ridge breadths were ence is statistically significant for all ten fingers (p < 0.001)
statistically compared. A sectioning point was then created to (Table 4). The data indicate that there are statistically significant
examine the correct classification accuracy for each finger and levels of sexual dimorphism in human finger ridge breadth. A
for each hand (pooled measurements). Factors, such as body size
(e.g., weight, stature, and body mass index (BMI)) may also
TABLE 1––Descriptive statistics for mean ridge breadth (in mm) by digit
influence ridge breadth measurements (e.g., larger individuals
(sexes pooled).
are expected to have greater ridge breadths). This was evaluated
by examining the correlation between ridge breadth, weight (in Mean SD Mean SD N
kg), stature (in cm), and BMI. A final consideration, interdigit
variation, was examined to determine whether significant differ- L1 5.18 0.85 R1 5.36 0.82 500
L2 5.08 0.77 R2 5.24 0.77 497
ences exist between fingers on each hand. This is an important L3 4.96 0.79 R3 5.05 0.81 498
consideration because patent and latent prints from crime scenes L4 4.77 0.77 R4 4.82 0.72 498
cannot always be correctly classified by digit and side (47,48). L5 4.80 0.71 R5 4.84 0.71 492

Results
Table 1 presents descriptive statistics for mean ridge breadth TABLE 2––Paired sample t-test comparison between left and right digits
(sexes pooled).
by digit for the total sample. Digit I corresponds with the thumb,
digit II is the index finger, digit III is the middle finger, digit IV Mean Correlation
is the ring finger, and digit V is the pinky. Mean ridge breadth difference t-value df p-value coefficients p-value
decreases from digits I–IV, with nearly identical mean values for
L1-R1 0.18 6.400 499 <0.001 0.73 <0.001
digits IV and V. Mean ridge breadth is greater for right fingers L2-R2 0.16 6.263 496 <0.001 0.73 <0.001
than for left fingers, and these differences are statistically signifi- L3-R3 0.09 4.092 497 <0.001 0.79 <0.001
cant for digits I–III and when all digit measurements are pooled L4-R4 0.05 1.939 497 0.053 0.73 <0.001
for each hand (p < 0.001; Table 2). Correlation coefficients com- L5-R5 0.04 1.404 491 0.161 0.73 <0.001
puted for left and right digits show a strong positive relationship L-R 0.10 4.493 4980 <0.001 0.75 <0.001
combined
(digits I, II, IV, and V, r = 0.73, p < 0.001; digit III, r = 0.79,
p < 0.001; digits combined, r = 0.75, p < 0.001) (Table 2).
TABLE 3––ANOVA and Bonferroni post hoc test comparison of finger ridge
breadths between digits.

Left hand Right hand

F-ratio p-value F-ratio p-value

33.209 <0.001 50.582 <0.001

Bonferroni p-value Bonferroni p-value

L1 vs. L2 0.003 R1 vs. R2 0.031
L1 vs. L3 <0.001 R1 vs. R3 <0.001
L1 vs. L4 <0.001 R1 vs. R4 <0.001
L1 vs. L5 <0.001 R1 vs. R5 <0.001
L2 vs. L3 1.000 R2 vs. R3 0.237
L2 vs. L4 <0.001 R2 vs. R4 <0.001
L2 vs. L5 <0.001 R2 vs. R5 <0.001
L3 vs. L4 0.001 R3 vs. R4 <0.001
L3 vs. L5 <0.001 R3 vs. R5 <0.001
FIG. 1––Example of measurement beginning and ending in a valley across L4 vs. L5 1.000 R4 vs. R5 1.000
ten parallel ridges.
894 JOURNAL OF FORENSIC SCIENCES

TABLE 4––Descriptive statistics and independent sample t-test comparison of male and female ridge breadths (in mm).

Males Mean SD N Females Mean SD N Mean difference t-value df p-value

L1 5.84 0.61 249 L1 4.53 0.48 250 1.32 26.906 468.767 <0.001
L2 5.64 0.60 249 L2 4.51 0.46 247 1.12 23.449 465.161 <0.001
L3 5.57 0.56 248 L3 4.34 0.43 249 1.23 27.669 462.447 <0.001
L4 5.35 0.56 248 L4 4.20 0.48 250 1.14 24.514 483.912 <0.001
L5 5.31 0.52 248 L5 4.28 0.47 244 1.03 23.190 490.000 <0.001
R1 5.94 0.65 249 R1 4.78 0.50 250 1.16 22.241 468.056 <0.001
R2 5.78 0.58 249 R2 4.70 0.53 250 1.08 21.833 492.236 <0.001
R3 5.65 0.60 249 R3 4.45 0.48 250 1.20 24.818 472.406 <0.001
R4 5.33 0.53 248 R4 4.31 0.47 250 1.02 22.789 496.000 <0.001
R5 5.35 0.51 248 R5 4.31 0.44 247 1.04 24.214 480.266 <0.001
Combined 5.06 0.61 2485 R5 4.96 0.51 2487 1.13 71.261 4796.888 <0.001

sectioning point (SP) was created for each finger based on the use of the combined fingers sectioning points will be especially
weighted midpoint for the samples, and correct classification helpful for sex estimation and shows classification accuracy sim-
accuracy was computed using linear discriminant function analy- ilar to most of the individual digits.
sis following the “leave-one-out” cross-validation procedure.
Values above the SP are classified as males and those below the
Correlation with Weight, Stature, and BMI
SP are classified as females (Tables 5–6). Overall classification
accuracy varies from 83% to 89%. On the right hand, all digits To evaluate the influence of body size, correlation coefficients
show classification accuracy above 80%, varying between 83% were computed between ridge breadth and weight, stature, and
and 87%. The left hand shows a similar pattern, with all digits BMI (Table 8). Weight shows a slightly higher correlation with
showing classification accuracy above 80%, and varying ridge breadth (r = 0.41–0.48, p < 0.001) than stature (r = 0.36–
between 84% and 89%. Despite the moderately high classifica- 0.43, p < 0.001). The correlation with BMI is the lowest
tion accuracy, overlap between the sexes in body size (e.g., (r = 0.152–0.226, p ≤ 0.001) and only explains a small amount
weight and stature) results in a number of larger females mis-
classifying as males and a number of smaller males misclassify-
ing as females. Due to the degree of overlap, these sectioning TABLE 6––Sex classification of left finger ridge breadth measurements using
points will be most useful in cases where fingerprints are either the sectioning point.
very small or very large (i.e., far away from the sectioning
point). Sectioning point N % Correct
In many cases, analysts are unable to determine which specific L1
digit is represented by a fingerprint. Sectioning points were cre- Total 5.173022 499 88.0
ated separately for each hand, pooling the measurements for all Males 249 87.1
digits (Table 7). For the right hand, correct classification accu- Females 250 88.8
racy is 81.3% for males and 86.4% for females (overall = 83.9). L2
Total 5.077046 496 83.9
Similarly, for the left hand, correct classification accuracy is Males 249 80.3
83.7% for males and 88.7% for females (overall = 86.2). The Females 247 87.4
L3
Total 4.953738 497 89.3
Males 249 86.3
TABLE 5––Sex classification of right finger ridge breadth measurements Females 297 92.4
using the sectioning point. L4
Total 4.772066 498 86.9
Sectioning point N % Correct Males 248 83.5
Females 250 90.4
R1 L5
Total 5.347292 499 84.6 Total 4.801931 492 86.2
Males 249 81.1 Males 248 86.7
Females 250 88.0 Females 244 85.7
R2
Total 5.225916 499 83.2
Males 249 81.5
Females 250 84.8 TABLE 7––Sex classification of left and right ridge measurements using the
R3 sectioning point for all fingers combined (pooled by side).
Total 5.230654 499 86.6
Males 249 84.7 Sectioning point N % Correct
Females 250 88.4
R4 Right hand
Total 4.807156 498 84.5 Total 5.05898 2490 83.9
Males 248 84.3 Males 1243 81.3
Females 250 84.8 Females 1247 86.4
R5 Left hand
Total 4.833628 495 86.3 Total 4.95801 2482 86.2
Males 248 84.3 Males 1242 83.7
Females 247 88.3 Females 1240 88.7
 MUNDORFF ET AL. . SEXUAL DIMORPHISM IN FINGERPRINTS 895

TABLE 8––Correlation coefficients between finger ridge breadth, weight, stature, and BMI.

Weight Correlation coefficients p-value Stature Correlation coefficients p-value BMI Correlation coefficients p-value
L1-Weight 0.43 <0.001 L1-Stature 0.39 <0.001 L1-BMI 0.155 0.001
L2-Weight 0.43 <0.001 L2-Stature 0.43 <0.001 L2-BMI 0.156 <0.001
L3-Weight 0.48 <0.001 L3-Stature 0.4 <0.001 L3-BMI 0.226 <0.001
L4-Weight 0.43 <0.001 L4-Stature 0.41 <0.001 L4-BMI 0.182 <0.001
L5-Weight 0.47 <0.001 L5-Stature 0.42 <0.001 L5-BMI 0.206 <0.001
R1-Weight 0.41 <0.001 R1-Stature 0.4 <0.001 R1-BMI 0.152 0.001
R2-Weight 0.44 <0.001 R2-Stature 0.37 <0.001 R2-BMI 0.199 <0.001
R3-Weight 0.42 <0.001 R3-Stature 0.39 <0.001 R3-BMI 0.187 <0.001
R4-Weight 0.41 <0.001 R4-Stature 0.36 <0.001 R4-BMI 0.164 <0.001
R5-Weight 0.47 <0.001 R5-Stature 0.38 <0.001 R5-BMI 0.213 <0.001

of ridge breadth variation. The data clearly show that larger indi-
viduals (both heavier and/or taller) tend to have greater ridge
breadth measurements than smaller individuals, which explains
why some females misclassify as males and vice versa.
Discussion
Patent and latent fingerprints have long been used to identify
missing persons, criminals, and other unidentified individuals.
The ability to reliably discern male from female fingerprints
could significantly reduce search parameters and time, possibly
leading to expedited identification of an individual. This study
demonstrates statistically significant differences between male
and female friction ridge breadth for each of the ten fingers and
for each hand when all finger ridge breadth measurements are
pooled (p < 0.001).
Fingerprint sexual dimorphism is expressed by higher mean
ridge breadth values for males as compared to females
(>1.0 mm), and classification accuracy using a sectioning point
varied between fingers from 83.2% to 89.3%. Bimanual differ-
ences similar to those found by Salazano and Benevides (28)
show directional asymmetry favoring the right hand. Classifica-
tion accuracy was similar to individual digits using sectioning
points aggregating the right fingers (86.4 females and 81.3
males), and even more so when the left fingers are aggregated
(88.7 females and 83.7 males). The relatively high classification
accuracy of the pooled finger sample indicates that it is not nec-
essary to know which digit is represented by a latent or patent
print to apply this method.
Despite successful classification rates, significant correlations
between ridge breadth and stature, weight, and body mass index
confound these results. While fingertips, and thus fingerprints,
continue to grow in size until an individual reaches maturity
(5,38), the study sampled only ridge breadth measurements from
adults. Therefore, the slightly higher correlation between ridge
breadth and weight, rather than with either stature or BMI, sug-
gests that increases in weight after adulthood influences ridge
breadth as well. Although the relationship between weight and
ridge breadth is moderate, it is sufficient to explain the misclas-
sification of larger females and smaller males. The relationship
between weight fluctuation and ridge breadth needs to be
explored further, which could be evaluated in a longitudinal
study on living individuals.
The use of measurements from an unknown fingerprint to esti-
mate an individual’s sex is analogous to biological anthropolo-
gists’ use of measurements from different skeletal elements to
estimate the sex of skeletal remains. Over the last century,
numerous methods have been developed to predict an individ-
ual’s sex based on metric analyses of bones. These studies have
included different aspects of nearly every skeletal element; for
example, the femur (49–52), the humerus (53–57), the cranium
(58–61), and the hand (62–64). And like different hands and dif-
ferent individual digits, different skeletal elements vary in their
expression of sexual dimorphism. This study demonstrates that
finger ridge breadth measurements have levels of sexual dimor-
phism that are similar to long bones, allowing them to be a rea-
sonably accurate predictor of an individual’s sex. Interestingly,
fingerprint ridge breadth seems to be more sexually dimorphic
and, therefore, more accurate for sex estimation, than the under-
lying hand and finger bones (e.g., metacarpals and phalanges)
(63, 65–67).
Conclusion
The results of this study confirm that a significant degree of
sexual dimorphism exists in finger ridge breadth measurements,
with males having larger mean values than females. The section-
ing points provide a useful tool for forensic scientists to narrow
down the possible sex of an unknown individual who is not
already in a missing person or criminal database. If an unknown
fingerprint cannot be identified to specific digit, or only a partial
print has been recovered, this method can be used to predict the
sex of an individual. Limitations of this study include the dis-
covery of significant correlations between finger ridge breath
measurements and weight and height. Further research should be
conducted to better understand the relationship between weight
and finger ridge breath measurements.
Acknowledgments
The authors would like to posthumously thank Larry Estes of
the Butte County California Sherriff’s Department who gra-
ciously provided the data for this study, and David Mercer for
his assistance with the data entry. We also thank the anonymous
JFS reviewers for their comments and suggestions for improving
this manuscript.
References
1. Galton F. Finger prints. New York, NY: Da Capo Press, 1965 [1892].
2. Cummins H, Midlo C. Finger prints, palms and soles: an introduction to
dermatoglyphics. New York, NY: Dover Publications, 1961.
3. Wertheim K. Embryology and morphology of friction ridge skin. In:
McRoberts A, editor. The fingerprint sourcebook. Washington, DC:
National Institute of Justice, 2011(3):1–26.
4. Montagna W, Parakkal P. The structure and function of skin, 3rd edn.
New York, NY: Academic Press, 1974.
5. Holt S. The genetics of dermal ridges. Springfield, IL: Charles C. Tho-
mas, 1968.
896 JOURNAL OF FORENSIC SCIENCES
6. Karmaker B, Kobyliansky E. Sexual dimorphism in two types of dermat-
oglyphic traits in the Turkmenian populations of Russia: principal com-
ponent. Anthropol Anz 2009;67(3):253–68.
7. Wertheim K, Maceo A. The critical stage of friction ridge and pattern
formation. J Forensic Identif 2002;52(1):35–85.
8. Young RS, Reed T, Hodes ME, Palmer CG. The dermatoglyphic and
clinical features of the 9p trisomy and partial 9p monosomy syndromes.
Hum Genet 1982;62(1):31–9.
9. Bagga A. Dermatoglyphics of schizophrenics. New Deli, India: Mittal
Publications, 1989.
10. Oladipo GS, Akanigha BE. Dermatoglyphic patterns in androgenic alope-
cia in a south eastern Nigerian population. J Exp Clin Anat 2005;4
(2):44–7.
11. Oladipo GS, Afolabi EO, Esomonu C. Dermatoglyphic patterns of
obese versus normal weight Nigerian individuals. Biomed Int 2010;1:
66–9.
12. Bharadwaja A, Saraswat PK, Aggarwal SK, Banerji P, Bharawaja S. Pat-
tern of finger-prints in different ABO blood groups. J Indian Acad
Forensic Med 2004;26(1):6–9.
13. http://www.fbi.gov/about-us/cjis/fingerprints_biometrics/iafis/iafis
(accessed October 2012).
14. Acree MA. Is there a gender difference in fingerprint ridge density?
Forensic Sci Int 1999;102:35–44.
15. Nayak VC, Rastogi P, Kanchan T, Lobo SW, Yoganarasimha K, Nayak
S, et al. Sex differences from fingerprint ridge density in the Indian pop-
ulation. J Forensic Leg Med 2010;17:84–6.
16. Nithan MD, Manjunatha B, Preethi DS, Balaraj BM. Gender differentia-
tion by finger ridge count among South Indian population. J Forensic
Leg Med 2011;18:79–81.
17. Jantz RL. Sex and race differences in finger ridge-count correlations. Am
J Phys Anthropol 1977;46(1):171–6.
18. Karmaker B, Yakovenko K, Kobyliansky E. Sexual dimorphism: asym-
metry and diversity of 38 dermatoglyphic traits in five endogamous pop-
ulations of West Bengal, India. Coll Antropol 2001;25(1):167–87.
19. Singh G. Determination of gender differences from fingerprint ridge den-
sity in two Northern Indian population of Chandigarh region. J Forensic
Res 2012;3:145.
20. Karmaker B, Kobyliansky E. Sexual dimorphism in the Turkmenian pop-
ulation in two types of dermatoglyphic traits: discriminant analysis. Coll
Antropol 2009;33(4):1007–14.
21. Karmaker B, Yakovenko K, Kobyliansky E. Quantitative digital and pal-
mar dermatoglyphics: sexual dimorphism in the Chuvashian population
of Russia. J Comp Hum Biol 2008;59:317–28.
22. Karmaker B, Yakovenko K, Kobyliansky E. Sexual dimorphism in the
Chuvashian population of Russia in two types of dermatoglyphic traits:
principal component analysis. Coll Antropol 2008;32(2):467–77.
23. Kobyliansky E, Micle S. Dermatoglyphic sexual dimorphism in East
European Jews. Bull Mem Soc Anthropol Paris 1989;1(1–2):13–36.
24. Kobyliansky E, Micle S. Dermatoglyphic sexual dimorphism in North
African Jews. Int J Anthropol 1988;3:77–91.
25. Kobyliansky E, Micle S. Dermatoglyphic sexual dimorphism in
Middle Eastern Jews. Bull Mem Soc Anthropol Paris 1987;4(4):271–89.
26. Micle S, Kobyliansky E. Dermatoglyphic sexual dimorphism in Yemen-
ite Jews. Bull Mem Soc Anthropol, XIV Serie 1987;4(2):95–113.
27. Sanna E, Usai E, Floris G. Inter-sex and intra-sex differences in quantita-
tive digital dermatoglyphics of Sardinian-speaking groups. Anthropol
Anz 2004;62(2):157–67.
28. Salazano FM, de Benevides FR, Sa F. Fingerprint quantitative variation
and asymmetry in Brazilian whites and blacks. Am J Phys Anthropol
2005;40:325–8.
29. Murad TA. A multivariate analysis of the sexual dimorphism in palmer
dermatoglyphics of North Alaskan Eskimos. Am J Phys Anthropol
1976;44:195.
30. Murad TA. Variation in the palmer dermatoglyphics of North Alaskan
Eskimos. Curr Anthropol 1980;21:672–3.
31. Gungadin S. Sex determination from fingerprint ridge density. Int J Med
Update 2007;2(2):4–7.
32. Reddy GG. Finger dermatoglyphics of the Bagathas of Araku Valley
(A.P.), India. Am J Phys Anthropol 1975;42(2):225–8.
33. Plato CC, Cereglino JJ, Steinberg FS. The dermatoglyphics of American
Caucasians. Am J Phys Anthropol 1975;42(2):195–210.
34. De Forest PR, Gaensslen RE, Lee HC. Forensic science: an introductory to
criminalistics. New York, NY: McGraw-Hill Book Company, 1983;330–59.
35. Jantz RL, Chopra VP. A comparison of dermatoglyphic methodologies
in population studies. Am J Phys Anthropol 1983;60(1):61–7.
36. Kamali MS, Mavalwala J, Khaneqah AA, Bhanu BV. Quantitative der-
matoglyphic traits as measures of population distance. Am J Phys
Anthropol 1991;85(4):429–50.
37. Cowger JF. Friction ridge skin: comparison and identification of finger-
prints. New York, NY: Elsevier Press, 1983.
38. Loesch D. Quantitative dermatoglyphics: classification, genetics, and
pathology. Oxford, U.K.: Oxford University Press, 1983.
39. Cummins H, Waits WJ, McQuitty JT. The breadths of epidermal ridges
on the finger tips and palms: a study of variation. Am J Anat 1941;65–
69:127–50.
40. Bonnevie K. Studies on papillary patterns of human fingers. J Genet
1924;15:1–113.
41. Mi M, Budy A, Rashad M. A population study of finger dermal patterns
and ridge counts. Prog Clin Biol Res 1982;84:285–93.
42. Hall J, Kimura D. Dermatoglyphic asymmetry and sexual orientation in
men. Behav Neurosci 1994;180(6):1203–6.
43. Buchi E. Finger ridge counts in samples from the tribal populations of
India with application of a 3-factor theory. In: Mavalwala J, editor. Der-
matoglyphics: an international perspective. Chicago, IL: Aldine,
1978;91–101.
44. Leguebe A, Vrydagn S, Ducros J. Finger ridge-count correlations: race,
sex, and hand differences. J Hum Evol 1981;10:453–66.
45. Demarchi DA, Giordano AR, Marcellino AJ. Dermatoglyphics in
Araucanian Indians from Patagonia and multivariate comparisons
with other Argentinean Aboriginal populations. Hum Biol 1997;69
(2):227–39.
46. Sangam MR, Krupadanam K, Anasuya K. A study of finger prints: bilat-
eral asymmetry and sex difference in the region of Andhra Pradesh. J
Clin Diagn Res 2011;5(3):597–600.
47. Holt S. The correlations between ridge-counts on different fingers. Ann
Eugen 1951;16:287–97.
48. Holt S. The correlations between ridge-counts on different fingers esti-
mated from a population sample. Ann Hum Genet 1959;23(4):459–60.
49. Pearson K, Bell J. The study of the long bones of the English skeleton.
Part 1: the femur (chapters 1 to 4). Draper’s Company Research Mem-
oirs, Biometric Series X. Cambridge, U.K.: Cambridge University Press,
1917/1919.
50. Seidmann RM, Stojanowski CM, Doran GH. The use of the superior-
inferior neck diameter as a sex assessor. Am J Phys Anthropol
1998;107:305–13.
51. Brown RP, Ubelaker DH, Schanfield MS. Evaluation of Purkait’s trian-
gle method for determining sexual dimorphism. J Forensic Sci 2007;52
(3):553–6.
52. Albanese J, Eklics G, Tuck A. A metric method for sex determination
using the proximal femur and fragmentary hipbones. J Forensic Sci
2008;53(6):1283–8.
53. Dwight T. The size of the articular surfaces of the long bones as charac-
teristic of sex; an anthropological study. Am J Anat 1904\1905;IV:19–
36.
54. Washburn SL. Sex differences in the pubic bone. Am J Phys Anthropol
1948;6:199–207.
55. Iscan MY, Loth SR, King CA, Shihai D, Yoshino M. Sexual dimor-
phism in the humerus: a comparative analysis of Chinese, Japanese and
Thais. Forensic Sci Int 1998;98(1–2):17–29.
56. Dittrick J, Suchey JM. Sex determination of prehistoric central California
skeletal remains using discriminant analysis of the femur and humerus.
Am J Phys Anthropol 1986;70(1):3–9.
57. Albanese J, Cardoso HFV, Saunders SR. Universal methodology for
developing univariate sample-specific sex determination methods: an
example using the epicondylar breadth of the humerus. J Archaeol Sci
2005;32(1):143–52.
58. Giles E, Elliot O. Sex determination by discriminant function analysis of
the crania. Am J Phys Anthropol 1963;21:53–68.
59. Holland TD. Sex determination of fragmentary crania by analysis of the
cranial base. Am J Phys Anthropol 1986;70:203–8.
60. Steyn M, Iscan MY. Sexual dimorphism in the crania and mandibles of
South African whites. Forensic Sci Int 1998;98(1–2):9–16.
61. Ousley SD, Jantz RL. FORDISC 3.0 personal computer forensic discri-
minant functions. Knoxville, TN: University of Tennessee, 2005.
62. Scheuer JL, Elkington NM. Sex determination from metacarpals and the
first proximal phalanx. J Forensic Sci 1993;38(4):769–78.
63. Falsetti AB. Sex assessment from metacarpals of the human hand. J
Forensic Sci 1995;40(5):774–6.
64. Case DT, Ross AH. Sex determination from hand and foot bone lengths.
J Forensic Sci 2007;52(2):264–70.
 MUNDORFF ET AL. . SEXUAL DIMORPHISM IN FINGERPRINTS 897

65. Burrows AM, Zanella VP, Brown TM. Testing the validity of metacarpal Additional information and reprint requests:
use in sex assessment of human skeletal remains. J Forensic Sci 2003;48 Amy Z. Mundorff, Ph.D.
(1):1–4. Department of Anthropology
66. Smith SL. Attribution of hand bones to sex and population groups. J The University of Tennessee
Forensic Sci 1996;41(3):469–77. 250 South Stadium Hall
67. Lazenby RA. Identification of sex from metacarpals: effect of side asym- Knoxville, TN 37996
metry. J Forensic Sci 1994;39(5):1188–94. E-mail: amundorff@utk.edu