Received: 21 September 2017 Revised: 29 April 2018 Accepted: 10 May 2018

DOI: 10.1002/ajhb.23138

American Journal of Human Biology

ORIGINAL RESEARCH ARTICLE

Sexual dimorphism in the ontogeny of second (2D) and fourth

(4D) digit lengths, and digit ratio (2D:4D)
John T. Manning1 | Bernhard Fink2,3

1
Applied Sports, Technology, Exercise, and
Medicine (A-STEM), Swansea University,
Swansea, United Kingdom, University of Swansea
2
Department of Behavioral Ecology, University of
Goettingen, Germany
3
Hanse-Wissenschaftskolleg, Institute for
Advanced Study, Delmenhorst, Germany
Correspondence
Bernhard Fink, Department of Behavioral
Ecology, University of Goettingen, Kellnerweg 6,
37077 Goettingen, Germany.
Email: bfink@gwdg.de
Abstract
Objective: Sex differences are often reported in digit lengths and digit ratio
(2D:4D). However, the ontogeny of these sex differences and their interrelation-
ships are less well known.
Methods: We considered sex differences in the lengths of the 2nd (2D) and 4th
(4D) digit and 2D:4D in children aged 2 to 18 years (Sample I, n = 680) and adults
aged 18 to 30 years (Sample II, n = 89,246). Digit length was determined by direct
experimenter-measurement (Sample I) and direct self-measurement (Sample II).
The data were tested with two-factor ANOVA's (sex and year-group).
Results: In both samples, there were significant main effects of sex and year-
group, and a significant interaction effect on digit length. Digit length was posi-
tively related to age in both samples. Boys had longer digits than girls but only
after 13 years. Men had longer digits than women and the dimorphism increased
from 18 to 30 years. There were significant sex differences in 2D:4D (males <
females), but no significant effect of age and no interaction effect of age and sex
on 2D:4D in children or adults.
Conclusions: Between 2 and 30 years, the lengths and the sexual dimorphisms of
2D and 4D are dependent on age. In contrast, 2D:4D is not age-dependent. We dis-
cuss our findings in the context of the ontogeny of digits and in the light of recent
claims on the presence of static allometry in 2D and 4D lengths.

1 | INTRODUCTION and can be found in populations of fully grown adults. How-

Digit length, specifically the 2nd (2D) and 4th (4D) digits,
and digit ratio (2D:4D) are often reported to be sexually
dimorphic. On average men have longer 2D and 4D than
women, and 2D:4D is also sexually dimorphic such that
men have shorter 2D relative to 4D than women (male 2D
and 4D > female 2D and 4D, Garn, Hertzog, Poznanski, &
Nagy, 1972; male 2D:4D < female 2D:4D, Manning, 2002).
The sex difference in 2D:4D is present if one measures it as
a ratio (Manning, Scutt, Wilson, & Lewis-Jones, 1998), or
as the distal extent of the 2nd and 4th digits when compared
to the 3rd digit (Peters, Mackenzie, & Bryden, 2002). These
sex differences in digit length and 2D:4D are well known
ever, the patterns of digit growth and the sexual dimor-
phisms of digit length and 2D:4D together with their
interrelations during ontogeny are less well known. It is
these ontogenetic patterns that are the focus of this article.
The markedly sexually dimorphic nature of digit length
in human adults is not found in children. Indeed, at some
stages in development, girls have mean digit lengths that
exceed that of boys (Garn et al., 1972). This can be seen in
data from the Fels Longitudinal Study of growth, matura-
tion, and body composition (Roche, 1992). The Fels project
was founded in 1929 as a serial study of growth involving
more than 1,000 participants in the USA. As part of the
study, longitudinal standards for phalanx lengths of children

Am J Hum Biol. 2018;30:e23138. wileyonlinelibrary.com/journal/ajhb © 2018 Wiley Periodicals, Inc. 1 of 7

https://doi.org/10.1002/ajhb.23138
2 of 7 American Journal of Human Biology
have been published (Poznanski, 1974). The participants
were white American children of Northern European ances-
try living in Ohio. Measurements were made from serial
X-rays of the hands that were taken from age 2 to 18 years.
To ensure continuity the standards comprised, where possi-
ble, the same participants such that for each sex the sample
sizes were n = 150 at age 4, n = 124 at age 9, n = 78 at
18, and n = 30 to 85 at intermediate ages. Focusing on mean
lengths of the phalanges, Garn et al. (1972) and Poznanski
(1974) reported longer phalangeal lengths in girls compared
to boys until age 14 and increasingly longer male phalanges
thereafter. With regard to 2D:4D, there is evidence for sex-
ual dimorphism (males < females) and this appears to be
independent of age. In samples in which digit lengths were
directly and experimenter-measured (n = 800, age 2 to
25 years; Manning et al., 1998) and directly and participant-
measured (the BBC internet study, n > 255,000, age
≥ 12 years, Manning, 2010), there was no strong support for
a relationship between 2D:4D and age.
Thus, there is evidence that 2D and 4D grow substan-
tially during human ontogeny and that the sex differences in
their lengths change during this period of growth. However,
it is not clear when such growth ceases and when the sexual
dimorphism in length becomes static. In many populations,
the onset of puberty has been accelerated while continued
growth has lifted its endpoint age well into the twenties
(Sawyer et al., 2012). When data are reported, the 10 to
24 year age range is increasingly being divided into three
categories: 10 to 14 years (early adolescence), 15 to 19 years
(late adolescence), and 20 to 24 years (young adulthood) to
appropriately examine the extent of changes in growth rates
that take place during these years. In the period between
20 to 24 years, young women are typically fully developed
physically while young men continue to gain height, weight,
and muscle mass (Sawyer et al., 2012). Thus, there may be
changes in digit length in young adulthood that are particu-
larly marked in men, leading to an increase in sexual dimor-
phism in 2D and 4D. In addition, it is of interest to establish
whether the apparent stability of 2D:4D is present through-
out the period of ontogenetic change in the length and sexual
dimorphism of digits.
To examine age- and sex-dependent changes in digit
length and 2D:4D, we ideally need to consider large samples
in order to encompass the variation present at each stage of
growth. Moreover, it is desirable to measure digit length
directly rather than indirectly (e.g., from photocopies or
scans of the hand). In comparison to the former, the latter
method tends to increase the length of the 4th digit relative
to the 2nd and this effect is greater for males than females
(Hönekopp & Watson, 2010; Ribeiro, Neave, Morais, &
Manning, 2016). One direct measurement protocol that has
the potential to recruit large numbers of participants is that
MANNING AND FINK
of self-reported digit measurements in on-line surveys. The
BBC internet survey recruited n > 255,000 participants
(Reimers, 2007) who reported the lengths of 2D and 4D
(Manning, Churchill, & Peters, 2007). However, it is inad-
visable to ask children to self-measure their digit lengths.
Thus, we considered two different data sets in children
and adults: (i) direct experimenter-measured digit lengths in
children and adolescents from ages 2 to 18 years, and
(ii) direct self-measured digit lengths in adults aged 18 to
30 years taken from the BBC internet survey. This meant we
considered a range of sex differences in both digit lengths
(2D and 4D) and 2D:4D in two data sets that together cover
age-groups 2 to 30 years. We focus on right hand measures
given previous reports on larger sex differences in right hand
2D:4D than left hand 2D:4D (Manning et al., 1998, and, see
Hönekopp & Watson, 2010 for meta-analysis).
With regard to digit length, we predicted that length and
sexual dimorphism would be dependent on age and that nei-
ther would be static after the age of 18 years. With regard to
2D:4D, we predicted that its magnitude and its sexual dimor-
phism would not be significantly age dependent, i.e., it
would be essentially static throughout ontogeny.
2 | METHODS
It has become established in recent years that, in addition to
sex differences, there are significant ethnic differences in
2D:4D. These are best seen in the three major ethnic groups:
“Whites” (of European descent), “Blacks,” and “South-East
Asians.” Whites tend to have high mean 2D:4D, while
Blacks and South-East Asians have low 2D:4D (Manning,
2002; Manning et al., 2007). Whites make up the largest eth-
nic group in the BBC survey. Therefore, we restricted both
our samples to Whites only.
2.1 | Sample I
The participants were British children and adolescents aged
2 to 18 years from the North-West of England. The data were
collected in 1995 as part of a study on fluctuating asymmetry
in children (Wilson & Manning, 1996). There were 40 partici-
pants (20 boys) from each year-group with a total of 680 indi-
viduals in the sample. Recruitment was from preschool groups,
play centers, primary schools, comprehensive schools, and a
sixth form college. Permission to measure was obtained from
parents and the educational institutions, and where appropriate
from the participants (≥ 16 years) themselves. The protocol of
the study was approved by the local ethical committee.
Measurements of digit length were made directly from
the fingers using Vernier calipers measuring to 0.1mm. Mea-
surement landmarks were from the ventral crease of the digit
proximal to the palm to the tip of the digit. Twenty
MANNING AND FINK American Journal of Human Biology 3 of 7

FIGURE 1 Cross-sectional development data of 2nd (2D) and 4th digits (4D), and digit ratio in boys and girls (Sample I) and adults (Sample II). Figures 1a
and 1b visualize mean digit length [(2D + 4D)/2)] across age, Figures 1c and 1d the sexual dimorphism of mean digit length, and Figures 1e and 1f the digit
ratio (2D:4D)

participants were re-measured after a period of time had
elapsed (at least 24 hours).
2.2 | Sample II
The participants were from the BBC internet study (see
Reimers, 2007 for details of the study). Participants self-
measured finger lengths and reported them online via drop-
down menus. Finger lengths were reported in millimetres
(values between 10 and 100 mm) for left and right index
and ring fingers. The instructions to measure and (self-)
report finger lengths were as follows: “Hold your right
hand in front of you. Look at where your ring finger joins
the palm of your hand. Find the bottom crease. Go to the
4 of 7 American Journal of Human Biology MANNING AND FINK

middle of this crease. Put the 0 of your ruler exactly on TABLE 1 Two-factor (sex and age) ANOVA analyses for dependent
the middle of the bottom crease. Make sure the ruler runs variables right 2D, right 4D, and right 2D:4D in 680 White children and
adolescents from 18 year-groups (2 years, 3 years, …, 18 years)
straight up the middle of your finger. Measure to the tip of
your finger (not your nail) in millimetres.” We considered Right 2D df F P
data from year-groups 18 to 30 years. In comparison to Sex 1, 646 7.08 < .01
experimenter-measured data, online studies tend to have
Age 16, 646 336.95 < .0001
high rates of random error in self-reported morphological
measurements. To eliminate outliers, and in accordance Sex × age 16, 646 6.11 < .0001

with earlier reports of the BBC data set, we restricted our Right 4D
sample of 2D:4D to a range of 0.80 to 1.20 (Manning
Sex 1, 464 24.89 < .0001
et al., 2007).
Age 16, 646 294.52 < .0001

Sex × age 16, 646 5.87 < .0001

3 | RE SUL TS
Right 2D:4D

Sex 1, 464 11.36 < .001

3.1 | Sample I
Age 16, 646 1.49 .097
3.1.1 | Repeatability
The repeated measures of first and second 2D:4D gave an Sex × age 16, 646 0.69 .081

intra-class coefficient of 0.81, F (1,19) = 9.77, P < .0001.
Therefore, we assumed the differences in between-individual
measurements of 2D:4D were significantly greater than the
within-individual measurement error.
3.1.2 | Descriptive statistics
There were 680 participants (340 males) in the total data set
and 40 participants (20 boys) in each year-group (see Sup-
porting Information Table S1 in the Appendix for details of
means by sex and age). With regard to 2D and 4D, the
means (± SE) were as follows: 2D for males 5.964 ±
.066 cm, and for females 5.887 ± .056 cm; 4D for males,
6.045 ± .065 cm, and for females 5.894 ± .055. For 2D:4D
the means (± SE) were as follows: for males, .987 ± .003,
and for females 1.000 ± .003.
Visual inspection indicated that mean digit lengths
[(2D + 4D)/2)] increased with age (Figure 1a). There
appeared to be little evidence of a sex difference in mean
digit length until age 14 to 18 years when males > females
(Figure 1c). However, there was no evidence of an age
dependent pattern for 2D:4D (Figure 1e).
3.1.3 | Statistical analysis
We performed three two-factor ANOVA's with indepen-
dent variables sex (male, female) and year-group (2 years,
…, 18 years), and dependent variables 2D or 4D, or
2D:4D. There were significant main effects of sex and age,
and a significant interaction effect of sex and age on 2D
length (Table 1). Fisher's PLSD test showed that male–
female 2D had a significant mean difference of .076 cm,
critical difference .056 cm, P < .01 (i.e., males had signifi-
cantly longer 2D than females). There were 135 pairwise
comparisons (2 years, 3 years, …., 17 years, 18 years), all
There were 40 children (20 males) per year group.
were negative and 126 of these were significant. Between
2 and 12 years, there were eight year-groups that showed
longer 2D in females compared to males. From 12 to
18 years male 2D became progressively longer than
female 2D.
For 4D there were similar effect patterns to that of 2D.
There were significant main effects of sex and age, and a
significant interaction effect of sex and age on 4D length
(Table 1). Fisher's PLSD test showed that male–female 4D
had a significant mean difference of .151 cm, critical differ-
ence .059 cm, P < .0001 (i.e., males had significantly longer
4D than females). Of the 135 pairwise comparisons, all were
negative and 125 of these were significant. Between 2 and
12 years, there were eight year-groups that showed longer
4D in females compared to males (4D males–4D females).
From 12 to 18 years, male 4D became progressively longer
than female 4D.
The pattern of effects for 2D:4D differed from that of
2D and 4D. There was a significant main effect of sex on
2D:4D, but no significant effects of age and no significant
interaction effect of sex and age (Table 1). Fisher's PLSD
test showed that male–female 2D:4D was = –.013, critical
difference .008, P < .001 (i.e., males had significantly
lower 2D:4D than females). Of the 135 pairwise compari-
sons, 117 were not significant and of the remaining
18, there were 14 negative and 4 positive. There was no
noticeable pattern of change in 2D:4D or male 2D:4D–
female 2D:4D. That is, the age-dependent patterns of digit
length and sex differences in digit length were not found
in 2D:4D.
MANNING AND FINK American Journal of Human Biology 5 of 7

TABLE 2 Two-factor (sex and age) ANOVA analyses for dependent negative and 58 of these were significant. All year-groups
variables right 2D, right 4D, and right 2D:4D in 89,245 young White adults showed longer 2D in males compared to females and the
from 13 year-groups (18 years, 19 years, …, 30 years)
overall pattern was for an increase in the sex difference
Right 2D df F P from 18 to 30 years.
For 4D, there were similar effect patterns to that of 2D.
Sex 1, 89220 4477.72 < .0001
There were significant main effects of sex and age, and a
Age 12, 89220 30.96 < .0001
significant interaction effect of sex and age on 4D length
Sex × age 12, 89220 1.81 < .05 (Table 2). Fisher's PLSD test showed that male–female 4D
Right 4D
had a significant mean difference of .736 cm, critical dif-
ference .029 cm, P < .0001 (i.e., males had significantly
Sex 1, 89220 5307.41 < .0001
longer 4D than females). Of the 78 pair-wise comparisons,
Age 12, 89220 30.79 < .0001 75 were negative and 60 of these were significant. All
Sex × age 12, 89220 1.76 < .05 year-groups showed longer 4D in males compared to
females and this sex difference increased from 18 to
Right 2D:4D
30 years.
Sex 1, 89220 945.07 < .0001 The pattern of effects for 2D:4D differed from that of 2D
Age 12, 89220 1.43 .146 and 4D. There was a significant main effect of sex on
2D:4D, but no significant effects of age, and no significant
Sex × age 12, 89220 0.98 .470
interaction effect of sex and age (Table 2). Fisher's PLSD
test showed that male–female 2D:4D was = –.011, critical
difference .001, P < .0001 (i.e., males had significantly
3.2 | Sample II lower 2D:4D than females). Of the 78 pairwise comparisons,
3.2.1 | Descriptive statistics 62 were not significant and of the remaining 16 all were
There were 89,245 participants (46,402 males) in the data positive.
set (see Supporting Information Table S2 in the Appendix
for details of means by sex and age). With regard to 2D, the
4 | DISCUSSION
means (± SE) were as follows, for males, 7.179 ± .006 cm,
and for females 6.519 ± .007 cm. With regard to 4D, the The results of the present study show that across an age range
means (± SE) were as follows, for males, 7.315 ± .007 cm, of 2 to 30 years, digit length (2D and 4D) increased substan-
and for females 6.578 ± .008. For 2D:4D, the means (± SE) tially. During this period of growth, the sex difference in digit
were as follows: for males, .984 ± .0002, and for females length varied across year-groups. That is, males did not have
.994 ± .0003. significantly longer digits than females before the age of
Visual inspection indicated that mean digit lengths 13 years, but thereafter the sexual dimorphism in digit length
[(2D + 4D)/2)] increased with age, although the growth (males > females) became progressively greater. In contrast,
spurt was much lower than for boys and girls up to age 2D:4D showed sex differences (males < females) indepen-
18 (Figure 1b). The sex difference (males > females) in dent of age.
mean digit length was consistent across age groups (Fig- One point of importance in these considerations is the
ure 1d). Like for boys and girls (Sample I), there was no mode of measurement of digit length. In our data, digit
evidence of an age dependent pattern for 2D:4D length is measured from soft-tissue markers, and the data
(Figure 1f). are cross-sectional rather than longitudinal. Measurements
of phalangeal growth are available from X-rays taken in
3.2.2 | Statistical analysis the Fels study of longitudinal growth in children aged from
We performed three two-factor ANOVA's with indepen- 2 to 18 years (for review see Roche, 1992). The Fels data
dent variables sex (male, female) and year-group (18 years, shows that the length of the tubular bones (phalanges) of
…, 30 years), and dependent variables 2D or 4D, or the digits increases rapidly with age and girls tend to have
2D:4D. There were significant main effects of sex and age, longer (not shorter) phalanges than boys (Garn et al.,
and a significant interaction effect of sex and age on 2D 1972). During early puberty, the sex differences in phalanx
length (Table 2). Fisher's PLSD test showed that male– length reverses as the digits of boys grow faster than those
female 2D had a significant mean difference of .660 cm, of girls. The point at which male and female phalanges
critical difference .019 cm, P < .0001 (i.e., males had sig- lengths are approximately equal is reached at 13 years.
nificantly longer 2D than females). There were 78 pair- Thereafter, the sex difference increases as phalanges
wise comparisons (18 years, …., 30 years), 75 were growth continues at a faster pace in boys relative to girls
6 of 7 American Journal of Human Biology
(Poznanski, 1974). Analyses of longitudinal growth of the
phalanges in the Fels study of children, young adults, and
their parents have emphasized that phalanges lengths and
sex differences in phalanges lengths change rapidly, but
the ratios of bone-to-bone length remain relatively stable
(Garn et al., 1972; Garn, Burdi, Babler, & Stinson, 1975;
Poznanski, 1974). These data add support to our findings
that digit length and digit sexual dimorphism are strongly
age-dependent, whereas the ratios of phalangeal lengths
are not.
Our data are relevant to claims that the apparent sexual
dimorphism in 2D:4D can be simply explained as a by-
product of an allometric shift in digit ratio with digit length
(Kratochvíl & Flegr, 2009; Lolli et al., 2017). These claims
are based on the assumption of static allometry, i.e., that the
samples they consider consist of individuals that are fully
grown. Kratochvíl and Flegr (2009) reported data from three
samples: 238 children of an elementary school; 297 univer-
sity students; 117 “members of the adult population.” Lolli
et al. (2017) considered the same data with the exclusion of
the elementary school children. However, no details of age
were given in either study. Our findings indicate that the
assumption of static allometry is not appropriate for children
and young adults recruited from elementary school and uni-
versity. With regard to the latter, Flegr, Lindová, Pivoñková,
and Havlicek (2008) was cited as the source of this sample.
They gave means and ranges for age as: female students,
21.05 years, 18–28 years and male students, 20.94, 18–27
years. The sample of the “adult population” may also have
contained individuals in the 18 to 30 year group. Static
allometry describes the phenotypic effects of the expression
of different alleles (e.g., those that influence digit length)
across individuals in a population that has ceased to grow.
During ontogenetic change it is to be expected that different
alleles are expressed with regard to traits such as digit
length. Samples that mix individuals whose digits are grow-
ing with those that are not are inappropriate for the analysis
of static allometry. The contention that sexual dimorphism
in 2D:4D is significantly influenced by an allometric shift in
digit ratio with digit length remains controversial (see a con-
sideration of Kratochvíl & Flegr, 2009 by Forstmeier, 2011;
Manning, 2010).
In conclusion, we have considered the sex difference in
2D:4D across two samples that have sex- and age-dependent
differences in the lengths of 2D and 4D. We have found that
2D:4D is sex-dependent (males < females). Moreover, the
sex difference in 2D:4D was independent of age-related
changes in both digit length and sex differences in digit
length. We consider the stability of 2D:4D during periods of
growth of 2D and 4D to be a remarkable example of homeo-
stasis of form. With regard to the contention that sex differ-
ences in 2D:4D are the result of a by-product of an
MANNING AND FINK
allometric shift in digit ratio with digit length (Kratochvíl &
Flegr, 2009; Lolli et al., 2017). We suggest that these claims
are based on an incorrect assumption of static allometry in
samples made up of individuals who are likely to be actively
growing.
OR CID
Bernhard Fink http://orcid.org/0000-0003-2739-5236
REFERENCES
Flegr, J., Lindová, J., Pivoñková, V. R. A., & Havlicek, J. A. N. (2008). Brief
communication: Latent toxoplasmosis and salivary testosterone concentra-
tion – important confounding factors in second to fourth digit ratio studies.
American Journal of Physical Anthropology, 137(4), 479–484.
Forstmeier, W. (2011). Women have relatively larger brains than men: A com-
ment on the misuse of general linear models in the study of sexual dimor-
phism. The Anatomical Record: Advances in Integrative Anatomy and
Evolutionary Biology, 294(11), 1856–1863.
Garn, S. M., Hertzog, K. P., Poznanski, A. K., & Nagy, J. M. (1972). Metacarpo-
phalangeal length in the evaluation of skeletal malformation. Radiology,
105(2), 375–381.
Garn, S. M., Burdi, A. R., Babler, W. J., & Stinson, S. (1975). Early prenatal
attainment of adult metacarpal-phalangeal rankings and proportions. Ameri-
can Journal of Physical Anthropology, 43(3), 327–332.
Hönekopp, J., & Watson, S. (2010). Meta-analysis of digit ratio 2D:4D shows
greater sex difference in the right hand. American Journal of Human Biol-
ogy, 22(5), 619–630.
Kratochvíl, L., & Flegr, J. (2009). Differences in the 2nd to 4th digit length ratio
in humans reflect shifts along the common allometric line. Biology Letters,
5(5), 643–646.
Lolli, L., Batterham, A. M., Kratochvíl, L., Flegr, J., Weston, K. L., &
Atkinson, G. (2017). A comprehensive allometric analysis of 2nd digit
length to 4th digit length in humans. Proceedings of the Royal Society
B: Biological Sciences, 284(1857), 20170356.
Manning, J. T. (2002). Digit ratio: A pointer to fertility, behavior, and health.
New Brunswick, NJ: Rutgers University Press.
Manning, J. T., Scutt, D., Wilson, J., & Lewis-Jones, D. I. (1998). The
ratio of 2nd to 4th digit length: A predictor of sperm numbers and
concentrations of testosterone, luteinizing hormone and oestrogen.
Human Reproduction, 13(11), 3000–3004.
Manning, J. T. (2010). Digit ratio (2D:4D), sex differences, allometry, and finger
length of 12–30-year olds: Evidence from the British Broadcasting Corpora-
tion (BBC) Internet study. American Journal of Human Biology, 22(5),
604–608.
Manning, J. T., Churchill, A. J., & Peters, M. (2007). The effects of sex,
ethnicity, and sexual orientation on self-measured digit ratio (2D:4D).
Archives of Sexual Behavior, 36(2), 223–192.
Peters, M., Mackenzie, K., & Bryden, P. (2002). Finger length and distal finger
extent patterns in humans. American Journal of Physical Anthropology,
117(3), 209–217.
Poznanski, A. K. (1974). The hand in radiologic diagnosis. Philadelphia: W.B.
Saunders.
Reimers, S. (2007). The BBC internet study. General methodology. Archives of
Sexual Behavior, 36(2), 147–161.
Ribeiro, E., Neave, N., Morais, R. N., & Manning, J. T. (2016). Direct versus
indirect measurement of digit ratio (2D:4D): A critical review of the litera-
ture and new data. Evolutionary Psychology, 14(1), 147470491663253.
Roche, A. F. (1992). Growth maturation and body composition: The Fels Longi-
tudinal Study 1929–1991. Cambridge Studies in Biological Anthropology.
Cambridge: Cambridge University Press.
Sawyer, S. M., Afifi, R. A., Bearinger, L. H., Blakemore, S. -J., Dick, B.,
Ezeh, A. C., & Patton, G. C. (2012). Adolescence: A foundation for future
health. Lancet (London, England)), 379(9826), 1630–1640.
MANNING AND FINK American Journal of Human Biology 7 of 7

Wilson, J., & Manning, J. T. (1996). Fluctuating asymmetry and age in chil-
dren: Evolutionary implications for the control of developmental instabil-
ity. Journal of Human Evolution, 30(6), 529–537.
How to cite this article: Manning JT, Fink B. Sexual
dimorphism in the ontogeny of second (2D) and
fourth (4D) digit lengths, and digit ratio (2D:4D). Am
SUPPORTING INFORMATION J Hum Biol. 2018;30:e23138. https://doi.org/10.1002/
Additional supporting information may be found online in ajhb.23138
the Supporting Information section at the end of the article.