Content uploaded by Emil O. W. Kirkegaard
Author content
All content in this area was uploaded by Emil O. W. Kirkegaard on Mar 28, 2022
Content may be subject to copyright.
MANKIND QUARTERLY 2022 62:3 484-497
484
Europeans Have Larger Testes than Sub-Saharan Africans
but Lower Testosterone Levels
Edward Dutton*
Asbiro Univerisity, Łódź, Poland
Helmuth Nyborg**
Aarhus University, Denmark (1968-2007, retired)
Emil Kirkegaard***
Ulster Institute for Social Research, London, UK
* Asbiro University, ul. Milionowa 12, 93-193 Łódź, Poland;
Email: e.c.dutton@dunelm.org.uk
** Email: helmuthnyborg@hotmail.com
*** Email: emil@emilkirkegaard.dk
Species and subspecies differ substantially in the size of their testicles. A
study has found differences in average testis size when comparing
Europeans and Northeast Asians. Other studies have found differences in
testosterone levels between Blacks, Whites and Northeast Asians. We
sought to replicate and extend these findings in a dataset of 4,462 US
soldiers (the Vietnam Experience Study). We find that those of Sub-
Saharan African descent (Blacks, n = 525) average smaller testes than
those of European descent (Whites, n = 3,654), d = 0.24, while Blacks
average higher testosterone levels, d = 0.11. These patterns are not
explained by differences in age, weight, or height. We discuss these
findings in terms of evolutionary theory of sperm competition and mating
systems. These differences are thought to relate to life history speed
differences, especially regarding mating patterns. A more polygynous
mating pattern is associated with smaller testes and high testosterone,
whereas a more competitive/free mating pattern is associated with larger
testes, helping with semen production and thus sperm competition. These
DUTTON, E., et al. TESTIS SIZE IN EUROPEANS AND SUB-SAHARAN AFRICANS
485
findings are consistent with how individual level differences in ejaculate
quality relate to life history speed.
Keywords: Testicles; Testis; Race; Testosterone
Comparing various mammalian species, it has been found that large testicle
size, in relation to body size, positively correlates with high copulatory frequency
and with the extent to which females are likely to mate with multiple males during
any given ovulatory cycle (Kenagy & Trombulak, 1986). Among primates, those
whose females will normally only mate with one male during each cycle, such as
gorillas, copulate less and have smaller testes in relation to their body size than
chimpanzees whose females, living in larger groups, will mate with many males
during their cycle (Harcourt et al., 1981). This has also been found when
comparing other primate species (Kappeler, 1997). It is widely accepted that
larger testes, in relation to body size, develop in order to allow the male to produce
more semen, elevating his chances in a system of sperm competition (Schillaci,
2006). The relationship between testis size, sperm competition and evolution has
been explored in a number of studies. For example, among the horned beetle,
males with smaller horns will compensate for this by developing larger testes
(Simmons & Emlen, 2006). A number of studies have found that ejaculate quality
in human males is negatively associated with social assertiveness and
psychoticism, both markers of high testosterone (see DeLecce et al., 2020).
In humans, larger testes correlate with male hormone levels and with semen
quantity (Takihara et al., 1987). It has also been found that a European sample
averaged larger testes than samples of Hong Kong Chinese (Diamond, 1986).
Many studies have recorded that there are consistent differences in average
testosterone levels when comparing Blacks with Whites (Ellis & Nyborg, 1992)
and with East Asian males. It has been noted that European-descended males
have, on average, higher free testosterone levels than Chinese males when
controlling for age (Xu et al., 2014). A meta-analysis of this issue found that after
adjustment for age, Black men have a modest but significantly higher free
testosterone level than White men of between 2.5% and 4.9% (Richard et al.,
2014). Studies that have not found this have often involved unrepresentative
samples, such as university students (e.g., Maestripieri et al., 2014), which are
problematic due to race differences in higher education participation in Western
countries (Herrnstein & Murray, 1994). Studies in other mammals have noted that
free testosterone strongly correlates with testicle size (Preston et al., 2011).
Thus, all else being equal, we would expect humans of Sub-Saharan African
descent to have larger testes than humans of European descent. This would be
in line with Europeans having larger testes than Northeast Asians, and Sub-
MANKIND QUARTERLY 2022 62:3
486
Saharan Africans being highest in free testosterone and Northeast Asians being
lowest. Accordingly, we set out to investigate human sub-population differences
in testosterone levels and in testis size and the relationship between the two
variables.
Method
Data source
We used archival data from the Vietnam Experience Study (VES;
https://www.cdc.gov/nceh/veterans/default1c.htm; http://ves.emilkirkegaard.dk/).
The VES is a US military dataset based on a sample of 4,462 enlisted men (3,654
Whites, 525 Blacks, 200 Hispanics, 49 Amerindian/Native Americans, and 34
Asians). They were inducted in the military between 1965 and 1971, and a follow-
up wave was conducted in 1985-1986. The purpose of the study was to examine
the possible health effects of exposure to the Agent Orange chemical weapon
during the Vietnam War. About 60% of the sample served in the Vietnam War
while the rest were controls who served elsewhere (such as in Korea).
A very wide variety of health measures were taken during the second wave
due to varied reports in the media about the health effects of the war. These
measurements included psychiatric interviews (DIS), self-rating (MMPI), medical
history, general physical examination, blood and urine analyses, sensory acuity,
and intelligence (19 tests). For a subsample, there were tests of semen and some
measures of numbers of offspring fathered by the men, the purpose of which was
to look for fertility issues and possible birth defects in the children. The physical
examination included measures of body size, including testis sizes (variables
GP01N05 and GP01N06), and the blood analyses included measurement of the
testosterone level (variable LD6861).
Measurements
The testes were in the beginning examined using a caliper to measure the
long axis of both testes. However, after 1st January 1986, by which time
approximately half of the veterans had been examined, the caliper was replaced
by Prader’s orchidometer (Prader, 1966) in order to improve precision. The
medical examiner stood in front of the subject and palpated (examined by touch)
one testis with one of his hands. At the same time, he went through a string of
testis-shaped beads with his other hand until he found one which was the closest
in size to the testis he was palpating. The Prader orchidometer consists of a string
of 12 1-25 ml beads (shown in Figure 1). This was then repeated for the second
testis.
DUTTON, E., et al. TESTIS SIZE IN EUROPEANS AND SUB-SAHARAN AFRICANS
487
Figure 1. The Prader orchidometer beads used to measure testis size (Karaman
et al., 2005).
We converted all data to Z scores to put them on the same scale without
having to deal with the conversion between the units from the two measurement
methods. In doing so, we found a small number of data errors, which we re-coded
as missing data due to uncertainty about how to correct them. We verified that
our data conversion worked as intended by comparing the correlation between
the right and left testis. In the unadjusted data, these correlated about .80 and
that was also the case in the corrected and merged data. The appendix provides
more information about the re-coding. Although Prader’s orchidometer method
MANKIND QUARTERLY 2022 62:3
488
may seem bizarre, it has been found by comparison to ultrasound measurement
to be quite precise (Mbaeri et al., 2013; Oehme et al., 2018; Sakamoto et al.,
2008). Accordingly, it is accepted as a precise instrument and it cannot, thus, be
averred that there may be subjective differences between the results obtained by
different medical examiners.
Blood specimens were taken in the morning before breakfast, following an
overnight fast commencing at 7pm. Plasma testosterone concentration was
determined (in nanograms/deciliter, or ng/100ml) using a standard double
antibody radioimmunoassay system (Leeco Diagnostics, Inc), and monitored with
bench and blind repeat quality control procedures. The coefficient of variation for
these repeats was between 4.49 and 9.72 with a median of about 5 (4 tests). This
value is close and possibly superior (Owen et al., 2013) to values from modern
equipment (Xu et al., 2018). The documentation for the measurement of
testosterone can be found on page 158ff of the Laboratory Methods and Quality
Control report (supplement A).
Results
Descriptive statistics for the main variables are shown in Table 1. Table 2
shows the correlations between the numerical variables.
The sample was on average borderline overweight (mean BMI = 26) with a
fairly representative mean age of 38. The two testis measurements were very
strongly correlated (r = .80), indicating reliability of the measurements. Average
testis size was weakly related to height and weight (r = .10 and .09), and less so
to BMI (r = .05). It was not related to testosterone level (r = .03, p > .01). Table 3
shows summary statistics for average testicle size and testosterone levels by
race.
Table 1. Descriptive statistics for the main variables; MAD = median absolute
deviation, SD = standard deviation.
N
Mean
Median
SD
MAD
Min
Max
Skew
Kurtosis
Testis mean
4413
0.00
-0.36
1.00
0.92
-4.1
4.6
0.19
0.73
Testosterone
(ng/dl)
4462
679.6
657
234.6
215.0
53
1950
0.82
1.49
Age (years)
4462
38.3
38
2.5
3.0
31
49
0.14
-0.01
Height (cm)
4462
176.2
176
6.75
5.9
150
202
-0.01
0.13
Weight (kg)
4461
83.6
82
15.4
13.3
47
190
1.22
3.77
BMI
4460
25.9
25.5
3.74
3.3
16.2
51.2
1.02
2.34
DUTTON, E., et al. TESTIS SIZE IN EUROPEANS AND SUB-SAHARAN AFRICANS
489
Table 2. Correlations between numerical variables; Testost. = testosterone.
Testis
right
Testis
left
Testis
avg.
Testost.
Age
Height
Weight
Testis left
0.80***
1
Testis avg.
0.95***
0.95***
1
Testost.
0.02
0.03
0.03
1
Age
0.00
0.00
0.00
-0.19***
1
Height
0.10***
0.09***
0.10***
-0.02
0.01
1
Weight
0.09***
0.09***
0.09***
-0.32***
0.06***
0.43***
1
BMI
0.05***
0.05***
0.06***
-0.34***
0.06***
0.04*
0.81***
* p < .01, ** p < .005, *** p < .001
Table 3. Summary statistics for testis size and testosterone level by race; SE =
standard error, MAD = median absolute deviation, SD = standard deviation.
White
Black
Hispanic
Asian
Native
Testis size average
N
3616
518
197
34
48
Mean
0.04
-0.21
-0.03
-0.75
0.11
SD
0.99
0.98
1.07
1.04
0.95
Median
-0.36
-0.36
-0.36
-0.67
-0.05
MAD
0.92
0.92
0.92
0.84
0.92
Min
-4.10
-4.10
-2.66
-2.75
-2.85
Max
4.62
3.37
4.62
2.13
1.51
SE of mean
0.02
0.04
0.08
0.18
0.14
Testosterone (ng/dL)
N
3654
525
200
34
49
Mean
676.0
701.4
680.7
727.7
678.1
SD
230.5
248.5
260.0
255.1
256.3
Median
654.5
679
635.5
686
618
MAD
215.7
228.3
197.9
210.5
220.9
Min
53
122
83
362
175
Max
1915
1846
1950
1450
1405
SE of mean
3.8
10.8
18.4
43.8
36.6
MANKIND QUARTERLY 2022 62:3
490
The sample sizes for Asians and Native Americans are very small, and so of
limited interest here. They are included for the possible use in future meta-
analyses. The results show that Whites average slightly larger testes than Blacks
and Hispanics. For testosterone, the relationship is reversed. Blacks have the
highest level and Hispanics are again intermediate. However, the relationships
are quite small and for the Hispanic comparisons, not statistically reliable. Thus,
we focus only on the White-Black comparisons. Standardized effect sizes [with
95% confidence intervals] for testis size are: White-Black 0.25 d [0.16, 0.34],
White-Hispanic 0.068 d [-0.075, 0.21]. For testosterone levels: White-Black -0.11
d [-0.20, -0.017], White-Hispanic -0.02 d [-0.16, 0.12]. Thus, the effect sizes can
be described as small to medium in strength. Because testosterone is on a ratio
scale, we can also compute the percentage differences, which are 3.9% for
White-Black, and 2.9% for White-Hispanic comparisons, respectively. Our finding
for the White-Black difference is closely in line with that of a 2014 meta-analysis
which found Blacks had 2.5 to 4.9% higher levels (Richard et al., 2014).
The relationships found above, however, could be due to some confounding
with age, height, and weight. To examine this, we used linear regression to
include these as covariates. Results are shown in Table 4.
Table 4. Regression results for testis size. Z = standardized to mean = 0,
standard deviation = 1. Race reference is White.
Model 1
Model 2
B
SE
p
B
SE
p
Intercept
0.04
0.017
0.025
0.07
0.227
0.750
Race: Black
-0.25
0.047
<0.001
-0.24
0.046
<0.001
Race: Hispanic
-0.07
0.073
0.351
0.01
0.074
0.877
Race: Asian
-0.78
0.171
<0.001
-0.67
0.171
<0.001
Race: Native
0.07
0.145
0.606
0.04
0.145
0.787
Age
0.00
0.006
0.877
height_z
0.10
0.015
<0.001
BMI_z
0.05
0.015
<0.001
R2 adj.
0.010
0.022
N
4413
4411
* p < .01, ** p < .005, *** p < .001.
DUTTON, E., et al. TESTIS SIZE IN EUROPEANS AND SUB-SAHARAN AFRICANS
491
The model results show that the Black-White testis size difference cannot be
explained with age, height or BMI differences (0.25 d before, 0.24 d with
covariates). Overall, however, testis size was very poorly predicted by the model,
with 2.2% of the variance explained with the full model. Table 5 shows the
comparable model for testosterone level.
Table 5. Regression results for testosterone level. z = standardized to mean =
0, standard deviation = 1. Race reference is White.
Model 1
Model 2
B
SE
p
B
SE
p
Intercept
0.00
0.017
1
2.60
0.215
<0.001
Race: Black
0.11
0.047
0.02
0.11
0.044
<0.01
Race: Hispanic
0.02
0.074
0.78
0.09
0.070
0.22
Race: Asian
0.22
0.175
0.20
0.09
0.163
0.58
Race: Native
0.01
0.146
0.95
0.10
0.137
0.48
Age
-0.07
0.006
<0.001
height_z
-0.02
0.014
0.23
BMI_z
-0.33
0.014
<0.001
R2 adj.
0.001
0.146
N
4462
4460
* p < .01, ** p < .005, *** p < .001.
As before, the Black-White gaps remain after adjustment for covariates (0.11
d before and after). We find here an inverse relation with age, a -0.07 standard
deviation lower testosterone level for each year of age (standardized beta = -
0.17). The effect size of BMI is surprisingly large at β = -0.33, and this is the chief
reason this model explains variation much better than the testis model (14.6%
explained vs. 2.2%).
Discussion
Our key finding is that Sub-Saharan Africans have slightly higher
testosterone levels than Europeans together with smaller testes. This may
potentially make sense in terms of differences in breeding patterns between these
subgroups. Anatomically modern humans are descended from approximately
twice as many females as males, with approximately 80% of females breeding
but only around 40% of males doing so, these being the more dominant males
MANKIND QUARTERLY 2022 62:3
492
selected by the females as part of polygamous unions (Baumeister, 2010;
Seielstad et al., 1998; Wilder et al., 2004). There are, however, racial differences
in the extent of this imbalance. X chromosome diversity has been shown to be
lower in European and East Asian populations than in African populations (Arbiza
et al., 2014). This would imply that polygamous mating patterns were more
pronounced in Sub-Saharan Africa than they were in Europe or Northeast Asia.
Outside of Africa, it appears, individual males monopolized the females to a lesser
extent, engaged in fewer sexual partnerships and, in essence, moved closer to
monogamy. Congruous with this difference, a large longitudinal study has found
that Black males score much higher on socio-sexuality than White or East Asian
males even within countries: ‘Black men were generally more permissive than
White, Hispanic, and Asian men’ (Sprecher et al., 2013). This is one of many
studies that have found higher levels of socio-sexuality when comparing Blacks
to other groups (see Dutton, 2020, pp.125-128; Johnson et al., 1994; Kogan et
al., 2015; Rushton, 1995).
Testosterone is associated with high levels of mating effort and thus with
socio-sexuality (Edelstein et al., 2011). Accordingly, finding that Blacks have
higher average testosterone levels than Whites would be congruous with
evidence of higher levels of socio-sexuality among Blacks, as well as with
evidence that Blacks are adapted to an ecology which is easier, in terms of basic
needs being met, but also more unstable, resulting in greater effort being put into
copulation as opposed to bonding and nurture. This direction of effort towards
mating, rather than nurture, in an unstable ecology is known as a relatively fast
life history strategy, and many studies have presented evidence for group
differences in average life history speed (see Dutton, 2020, Ch. 6; Figueredo et
al., 2021).
On the other hand, our finding that Blacks have smaller testes than Whites,
with Whites also having larger testes than Northeast Asians, would appear to defy
a simple explanation. One possibility is that Whites have larger testes than
Northeast Asians simply due to being higher in socio-sexuality, meaning that
females will be more likely to be unfaithful, resulting in the development of larger
testes in order to triumph in sperm competition in a context of relatively large,
cooperative social groups in which females are able to have contact with males
other than their main sexual partner. However, it may be that among Sub-
Saharan Africans, there was less of a pressure to develop large and cooperative
social groups, due to the easier ecology, with such cooperation anyway less likely
to be reciprocated due to the relative instability of the ecology. In these smaller
social groups, there would be fewer males around to attempt to cuckold any given
dominant male, with the result that sperm selection pressure would be weaker,
leading to the development of smaller testes among Sub-Saharan Africans
DUTTON, E., et al. TESTIS SIZE IN EUROPEANS AND SUB-SAHARAN AFRICANS
493
despite their having the highest testosterone levels. This is congruous with
individual-level data which found that males who pursue a slower life history
strategy concomitantly reflect the cuckoldry-risk hypothesis. They produce higher
quality ejaculates, allowing them to triumph in the battle of sperm selection
(Barbaro et al., 2019). Similarly, Blacks pursue a faster life history strategy than
Whites but their smaller testicles imply lower ejaculate quality, with sperm quality
and quantity robustly correlating (Liao et al., 2019).
Thus the harsh, stable, and highly competitive ecology of Northeast Asia
would result in the necessity of high levels of nurture and of stable, strongly
cooperative groups, reducing both testosterone levels and testicle size. However,
the plentiful but unstable ecology of Africa would result in high mating effort but
in only small and fission-prone groups, meaning relatively few males in any given
group. This fission tendency in Sub-Saharan African groups has been widely
documented (Norbeck, 1961, p.238; Turnbull, 1968, p.132). This tendency to not
form large polities would reduce the intensity of sperm competition, reducing
testis size. Therefore Europeans, between these two extremes, have the largest
average testis size.
Limitations
One issue with these data is that testes shrink due to various conditions,
specifically malnutrition, alcoholism, and chronic, terminal illness (Handelsman &
Staraj, 1985). However, this would not appear to be relevant to our sample.
Another potential issue is differences in the method by which testes are
measured. However, an analysis of the five main methods has found that they
are all roughly equally reliable, so this is just as reliable as any other method and
there is no reason to be prejudiced against any of the procedures because it
appears bizarre (Chipkevitch et al., 1996).
Finally, it might be averred that there might have been differences between
the White and Black men in our sample, which could have influenced our findings:
for example, differences in physical, physiological, reproductive and
psychological health; in war experience and trauma; in exposure to chemical
agents during the war; in drug use; in military rank; in personal background or in
reproductive history. However, in that our findings are congruous with meta-
analyses on race differences in androgens it seems most improbable that these
confounds are germane. Moreover, with the exception of exposure to chemical
agents, these could be regarded as reflections of testosterone as well as inducers
or suppressors of it.
Conflict of Interest: We declare that there is no conflict of interest.
MANKIND QUARTERLY 2022 62:3
494
Data Availability: The data that support the findings of this study are available
from the American Government but restrictions apply to the availability of these
data, which were used under license for the current study, and so are not publicly
available. Data are however available from the authors upon reasonable request
and with permission of the American Government.
References
Arbiza, L., Gottipati, S., Siepel, A. & Keinan, A. (2014). Contrasting X-linked and
autosomal diversity across 14 human populations. American Journal of Human
Genetics 6: 827-844. https://doi.org/10.1016/j.ajhg.2014.04.011
Barbaro, N., Shackelford, T.K., Jeffery, A.J., Lopes, G.S., Holub, A.M. & Zeigler-Hill, V.
(2019). Life history correlates of human (Homo sapiens) ejaculate quality. Journal of
Comparative Psychology 133: 294-300.
Baumeister, R. (2010). Is there Anything Good about Men? How Cultures Flourish by
Exploiting Men. Oxford: Oxford Univ. Press.
Chipkevitch, E., Nishimura, R., Tu, D. & Galea-Rojas, M. (1996). Clinical measurement
of testicular volume in adolescents: Comparison of the reliability of 5 methods. Journal
of Urology 6: 2050-2053.
DeLecce, T., Lopes, G.S., Zeigler-Hill, V., Welling, L.M., Shackelford, T.K. & Abed, M.G.
(2020). A preliminary but methodologically improved investigation of the relationships
between major personality dimensions and human ejaculate quality. Personality and
Individual Differences 153: 109614.
Diamond, J. (1986). Ethnic differences: Variation in human testis size. Nature 320: 488-
489.
Dutton, E. (2020). Making Sense of Race. Whitefish, MT: Washington Summit.
Edelstein, R., Chopnik, W. & Kean, E. (2011). Sociosexuality moderates the association
between testosterone and relationship status in men and women. Hormones and
Behavior 60: 248-255.
Ellis, L. & Nyborg. H. (1992). Racial/ethnic variations in male testosterone levels: A
probable contributor to group differences in health. Steroids 57: 72-75.
Figueredo, A.J., Hertler, S.C. & Peñaherrera-Aguirre, M. (2021). The biogeography of
human diversity in life history strategy. Evolutionary Behavioral Sciences 15: 10-26.
Handelsman, D. & Staraj, S. (1985). Testicular size: The effects of aging, malnutrition,
and illness. Journal of Andrology 6: 144-151.
DUTTON, E., et al. TESTIS SIZE IN EUROPEANS AND SUB-SAHARAN AFRICANS
495
Harcourt, A.H., Harvey, P.H., Larson, S.G. & Short, R.V. (1981). Testis weight, body
weight and breeding system in primates. Nature 293: 55-57.
Herrnstein, R. & Murray, C. (1994). The Bell Curve: Intelligence and Class Structure in
American Life. New York: Free Press.
Johnson, E.H., Jackson, L.A., Hinkle, Y., Gilbert, D., Hoopwood, T. et al. (1994). What is
the significance of black-white differences in risky sexual behavior? Journal of the
National Medical Association 86: 745-759. PMCID: PMC2607704.
Kappeler, P.M. (1997). Intrasexual selection and testis size in strepsirrhine primates.
Behavioral Ecology 8: 10-19.
Karaman, M.I., Kaya, C., Caskurlu, T., Guney, S. & Ergenekon, E. (2005). Measurement
of pediatric testicular volume with Prader orchidometer: Comparison of different hands.
Pediatric Surgery International 21: 517-520.
Kenagy, G.J. & Trombulak, S. (1986). Size and function of mammalian testes in relation
to body size. Journal of Mammalogy 67: 1-22.
Kogan, S.M., Yu, T., Allen, K.A., Pocock, A.M. & Brody, G.H. (2015). Pathways from
racial discrimination to multiple sexual partners among male African American
adolescents. Psychology of Men and Masculinity 16: 218-228.
Liao, W.B., Zhong, M.J. & Lüpold, S. (2019). Sperm quality and quantity evolve through
different selective processes in the Phasianidae. Scientific Reports 9: 19278.
Maestripieri, D., Klimczuk, A., Traficonte, D. & Wilson, M.C. (2014). Ethnicity-related
variation in sexual promiscuity, relationship status and testosterone levels in men.
Evolutionary Behavioral Sciences 8: 96-108.
Mbaeri, T.U., Orakwe, J.C., Nwofor, A.M., Oranusi, K.C. & Mbonu, O.O. (2013).
Accuracy of Prader orchidometer in measuring testicular volume. Nigerian Journal of
Clinical Practice 16: 348-351.
Norbeck, E. (1961). Religion in Primitive Society. New York: Harper.
Oehme, N.H.B., Roelants, M., Bruserud, I.S., Eide, G.E., Bjerknes, R., Rosendahl, K. &
Juliusson, P.B. (2018). Ultrasound-based measurements of testicular volume in 6- to
16-year-old boys—intra- and inter-observer agreement and comparison with Prader
orchidometry. Pediatric Radiology 48: 1771-1778.
Owen, L.J., MacDonald, P.R. & Keevil, B.G. (2013). Is calibration the cause of variation
in liquid chromatography tandem mass spectrometry testosterone measurement?
Annals of Clinical Biochemistry 50: 368-370.
Prader, A. (1966). Testicular size: Assessment and clinical importance. Triangle 7: 240-
243.
MANKIND QUARTERLY 2022 62:3
496
Preston, B., Stevenson, I., Lincoln, G., Montfort, S., Pilkington, J. & Wilson, K. (2012).
Testes size, testosterone production and reproductive behaviour in a natural
mammalian mating system. Journal of Animal Ecology 81: 296-305.
Richard, A., Rohrmann, S., Zhang, L., Eichholzer, M., Bazaria, S. et al. (2014). Racial
variation in sex steroid hormone concentration in black and white men: A meta-analysis.
Andrology 2: 428-435.
Rushton, J.P. (1995). Race, Evolution and Behavior: A Life History Perspective. New
Brunswick, NJ: Transaction Publishing.
Sakamoto, H., Ogawa, Y. & Yoshida, H. (2008). Relationship between testicular volume
and testicular function: Comparison of the Prader orchidometric and ultrasonographic
measurements in patients with infertility. Asian Journal of Andrology 10: 319-324.
Schillaci, M. (2006). Sexual selection and the evolution of brain size in primates. PLoS
ONE 1(1): e62. https://doi.org/10.1371/journal.pone.0000062
Seielstad, M.T., Minch, E. & Cavalli-Sforza, L.L. (1998). Genetic evidence for a higher
female migration rate in humans. Nature Genetics 20: 278.
Simmons, L.W. & Emlen, D.J. (2006). Evolutionary trade-off between weapons and
testes. Proceedings of the National Academy of Sciences USA 103: 16346-16351.
Sprecher, S., Treger, S. & Sakaluk, J. (2013). Premarital sexual standards and
sociosexuality: Gender, ethnicity, and cohort differences. Archives of Sexual Behavior
42: 1395-1405.
Takihara, H., Cosentino, M.J., Sakatoku, J. & Cockett, A. (1987). Significance of
testicular size measurement in andrology II: Correlation of testicular size with testicular
function. Journal of Urology 137: 416-419.
Turnbull, C. (1968). The importance of flux in two hunting societies. In: R. Lee & I.
DeVore (eds.), Man the Hunter. New Brunswick, NJ: Transaction Publishing.
Wilder, J.A., Mobasher, Z. & Hammer, M.F. (2004). Genetic evidence for unequal
effective population sizes of human females and males. Molecular Biology and
Evolution 21: 2047-2057.
Xu, L., Yeung, S.A., Kavikondala, S., Leung, G.M. & Schooling, C.M. (2014).
Testosterone concentrations in young healthy US versus Chinese men. American
Journal of Human Biology 26: 99-102.
Xu, X., Zhang, X., Zhong, Y., Saad, F., Perez-Patron, M. et al. (2018). Dynamic patterns
of testosterone levels in individuals and risk of prostate cancer among hypogonadal
men: A longitudinal study. Journal of Urology 199: 465-473.
DUTTON, E., et al. TESTIS SIZE IN EUROPEANS AND SUB-SAHARAN AFRICANS
497
Appendix: Testis data recoding
Some values are clearly wrong using the date breakpoint given, i.e., 1st
January 1986. The documentation about the measurements can be found on
page 187 and 268-271 of the medical examination book. Unfortunately, they don’t
go into more details about the methods. However, we note they have the same
errors in their results, i.e., some data that supposedly concern sizes that are far
beyond the possible values. There is at least one White subject with left testis
size given as 20, which is an impossibly high value and which must instead
indicate the volume. This subject was tested on 1985-12-17, i.e., before their
breakpoint data. There is similarly one subject tested on 1985-01-22 with size 15,
which must be volume 15, and one with size 10 on 1985-12-09 which must be
volume 10. There are 4 people tested before June 1985, which seems to be
before data collection started at all. They were likely tested in 1986 and we set
this as the new test date. Other data errors include values that only occur once,
such as one person whose value was given as 17. This might be size 7 (an
unlikely value). To avoid introducing any more errors, these impossible values
were recoded to missing (NA). For more detail, see the output of the R notebook.
Figure A1 below shows the correlations between the left and right testis after data
conversion.
Figure A1. Scatterplot of correlation between left and right testis after data
correction and merging.