Human size evolution: no evolutionary allometric relationship between male and female stature

https://doi.org/10.1016/j.jhevol.2004.07.004Get rights and content

Abstract

In many animal groups, sexual size dimorphism tends to be more pronounced in species with large body size. Similarly, in a previous cross-cultural analysis, male and female stature in humans were shown to be positively allometrically related, indicating a similar relationship where populations with larger stature were more dimorphic. In this study, we re-examine the hypothesis of an allometric relationship between the sexes using phylogenetic methodology. First, however, we tested whether there exist phylogenetic signals in male and female stature. Data on mean stature from 124 human populations was gathered from the literature. A phylogenetic test showed that male and female stature were significantly associated with phylogeny. These results indicate that comparative methods that to some degree incorporate genetic relatedness between populations are crucial when analyzing human size evolution in a cross-cultural context. Further, neither non-phylogenetic nor phylogenetic analyses revealed any allometric relationship between male and female stature. Thus, we found no support for the idea that sexual dimorphism increases with increasing stature in humans.

Introduction

In the animal kingdom, adult males and females often differ in size. In most animal groups, females are usually the larger sex, though in mammals and birds males are most often larger – a pattern that can be attributed to sexual selection through male-male competition in these two groups (Darwin, 1871).

Sexual size dimorphism (SSD) varies considerably between different animal species and taxa. This variation among species appears to follow a general allometric pattern between SSD and body size in related species. In taxa in which males is the larger sex, SSD tends to increase with body size (hyperallometry), i.e. larger species have relatively larger sexual size dimorphism, while SSD decreases with body size in species where females are larger (hypoallometry). This trend was first pointed out by Rensch, 1950, Rensch, 1959) and is hence called Rensch's rule.

Rensch's rule, however, contains no statement on what causes the phenomenon. Instead, a number of functional hypotheses have been proposed to explain the pattern, comprehensively summarized and evaluated in a review by Fairbairn (1997). One potential explanation is that female size changes when larger male size is being selected for as a result of a genetic correlation between the sexes concerning size-controlling genes (Maynard Smith, 1978, Lande, 1980, Lande, 1987, Lande and Arnold, 1983). That a genetic correlation for size-controlling genes is likely to exist in humans is indicated by a high degree of covariance between the sexes in stature (Rogers and Mukherjee, 1992; but see Rice, 1984, Roldan and Gomendio, 1999). Theoretical work shows that this effect is only expected initially, however, as natural selection is expected to reverse the process and eventually return female body size to its natural selection optimum (Lande, 1980, Reeve and Fairbairn, 2001).

Another potential explanation for Rensch's rule is what Fairbairn (1997) terms correlational selection; when male size increases because of sexual selection, larger female size will be selected for as well due to effects of the larger size of the males. In mammals this selection can come about, for example, because of increased demands for giving birth to and nursing larger male offspring (e.g. Lindenfors, 2002).

Even though Rensch's rule has been widely accepted and is supported by a number of studies in different animal groups, there are also several contradicting results (Abouheif and Fairbairn, 1997, and references cited therein). Particularly primates is a well-studied group and several non-phylogenetic analyses support the notion that primates conform to Rensch's rule (e.g. Ralls, 1976, Clutton-Brock et al., 1977, Leutenegger, 1978, Leutenegger and Cheverud, 1982, Gaulin and Sailer, 1984, Reiss, 1986), with the exception of strepserhines (Kappeler, 1990), and with a weaker trend for platyrrhines (Ford, 1994). Since much of the size variation among species can be explained by shared ancestry, however, a comparative method that takes phylogeny into account is necessary (Brooks and McLennan, 1991, Harvey and Pagel, 1991).

Phylogenetic analyses of primate SSD do not show clear-cut results in favour of either the presence or absence of Rensch's rule, however. Lindenfors and Tullberg (1998) found no significant allometric relationship between male and female size, while an allometric relationship was found in studies by Cheverud et al., 1986, Abouheif and Fairbairn, 1997, Plavcan and van Schaik (1997) and Smith and Cheverud (2002).

These discrepant results can possibly be explained as an effect of different sample sizes, however. For example, as both theoretical work (Maynard Smith, 1978, Lande, 1980, Lande, 1987, Lande and Arnold, 1983) and empirical evidence (Lindenfors and Tullberg, 1998) indicate that sexual selection on male body size induces a correlated response in female body size, it should follow that co-variation between body size and body size dimorphism should be found only in comparisons between clades differing in degree of sexual selection, while comparisons within clades sharing a common mating system should show no such co-variation. Thus, by excluding species randomly in the phylogeny, hence excluding many species within clades sharing a common mating system, one excludes the exact variation that would diffuse the evidence for the presence of Rensch's rule. Note that this problem would not only occur if sexual selection is the mechanism behind Rensch's rule, but should apply equally to any cause behind Rensch's rule that contains a phylogenetic component.

Most studies in support of Rensch's rule have been carried out on an interspecific level. However, Rensch claimed that the rule also should apply to “subspecies of a species” (Rensch, 1959, p. 159), thus implying that it ought to be possible to also trace effects of Rensch's rule in comparisons between populations.

Human SSD is most commonly measured as the male to female stature (height) ratio. In every population of human adults ever studied, mean stature in males has been greater than in females (Eveleth, 1975). The average SSD in a cross-cultural sample has previously been reported to be approximately 1.07 (Gaulin and Boster, 1985). Different human populations vary somewhat in SSD, however. For example, in a population with a relatively high SSD, like the Mountain Ok (Eveleth and Tanner, 1990), the SSD will be about 8 percent higher than in a population with low SSD, like Assyrians (Field, 1952). If differences in human SSD are consistent with Rensch's rule, then populations with above average mean stature should be more likely to display a high SSD.

To test this, Wolfe and Gray (1982) collected and compared mean heights of men and women in various human populations and found support for an allometric relationship between male and female stature, thus indicating that Rensch's rule applies also to interpopulation comparisons on stature dimorphism. Their conclusion was later criticised by Gaulin and Boster (1985), who claimed that the feeble support for allometry found by Wolfe and Gray (1982) was an artefact of too small sample sizes for some of the populations.

Instead, Gaulin and Boster (1985) argued that cross-cultural differences in SSD are mainly a function of within-population sample size, and that the degree of dimorphism in humans actually is very consistent. However, variation in SSD between populations is clearly present in the sample used in the present study (Appendix I). Further, in e.g. the examples mentioned above of populations with large differences between recorded SSD – Mountain Ok and Assyrians – sample size is over 100 subjects per sex.

None of the two above mentioned studies on human sexual stature allometry used a comparative phylogenetic method, however, or any other method appropriate to correct for errors arising as a consequence of populations sharing a common ancestry. In a previous study on human SSD, though not testing for the presence of Rensch's rule, Holden and Mace (1999) found that sexual stature dimorphism showed a highly significant association with phylogeny, thus suggesting that there should be phylogenetic signals in both male and female stature. If human populations have more similar body sizes the more genetically related they are, comparative methods that to some degree incorporate genetic relatedness between populations are a necessity when analyzing human size evolution in a cross-cultural context.

Here, we investigate the presence of phylogenetic signals in male and female stature, and test the possibility of an allometric relationship between male and female stature in humans using both non-phylogenetic and phylogenetic approaches.

Section snippets

Materials and methods

Data on male and female mean stature were collected from a variety of published sources (Appendix I) for 124 of the populations included in Cavalli-Sforza et al. (1994). Data from different sources were considered as belonging to the same ethnic group if they had the same name and location, if a matching synonymous name could be found in Grimes (1992), or could otherwise be deduced to be the same from e.g. Murdock (1967).

Populations with data on fewer than 14 individuals of each sex were

Results

Mean SSD in the whole sample, analyzing tip values directly, was 1.072 both when including and excluding Europeans. An average calculated incorporating phylogenetic information as suggested by Garland et al. (1993) gave a mean of 1.069, both with and without Europeans.

Major axis regressions on population data directly, without taking phylogeny into account, showed a strong correlation between male and female stature data (Europeans included: Male Stature = 1.030·Female Stature−0.066, p < 0.001; R2 = 

Discussion

In this study we show that the relationship between average male and female stature in human populations follow a pattern that is not significantly different from isometry. Hence, the level of sexual dimorphism cannot be seen as a function of stature. Furthermore, we show that there are clear phylogenetic signals in the data on human stature, indicating that comparative methods taking genetic ancestry into account are a necessity when analyzing human size evolution.

Since there was no support

Acknowledgements

We wish to thank Birgitta S. Tullberg and three anonymous reviewers for comments on a previous draft of this manuscript. This work was supported by the Swedish Research Council (through PL).

References (99)

  • S.P. Blomberg et al.

    Testing for phylogenetic signal in comparative data: behavioral traits are more labile

    Evolution

    (2003)
  • J.A. Birkbeck et al.

    Nutritional status of British Columbia IndiansII. Anthropometric measurements, physical and dental examinations at Ahousat and Anaham

    Can. J. Public Health

    (1971)
  • D.R. Brooks et al.

    Phylogeny, Ecology, and Behavior - A Research Program in Comparative Biology

    (1991)
  • L.L. Cavalli-Sforza

    Anthropometric data

  • L.L. Cavalli-Sforza et al.

    The History and Geography of Human Genes

    (1994)
  • K.C. Chen

    Taiwan Aborigines: A Genetic Study of Tribal Variations

    (1967)
  • J.M. Cheverud et al.

    A phylogenetic autocorrelation analysis of sexual dimorphism in primates

    Am. Anthropol.

    (1986)
  • T.H. Clutton-Brock et al.

    Sexual dimorphism, socionomic sex ratio and body weight in primates

    Nature

    (1977)
  • J. Comas

    Anthropometric studies in Latin American Indian populations

  • E. Crognier

    Natural selection and physical adaptation to a biotype of tropical Africa

  • C. Darwin

    The Descent of Man and Selection in Relation to Sex

    (1871)
  • P.B. Eveleth

    Differences between ethnic groups in sex dimorphism of adult height

    Ann. Hum. Biol.

    (1975)
  • P.B. Eveleth et al.

    Worldwide Variation in Human Growth

    (1990)
  • P.B. Eveleth et al.

    Worldwide Variation in Human Growth

    (1976)
  • D.J. Fairbairn

    Allometry for sexual size dimorphism: pattern and process in the coevolution of body size in males and females

    Annu. Rev. Ecol. Syst.

    (1997)
  • J. Felsenstein

    Phylogenies and the comparative method

    Am. Nat.

    (1985)
  • J. Felsenstein

    Contrasts for a within-species comparative method

  • Field, H. (1952). The anthropology of Iraq. Papers Peabody Mus. 46, nos....
  • Field, H. (1970). Contributions to the physical anthropology of the peoples of India. Florida: Field Research...
  • S.M. Ford

    Evolution of sexual dimorphism in body weight in platyrrhines

    Am. J. Primatol.

    (1994)
  • A. Froment et al.

    Climate-associated anthropometric variation between populations of the Niger bend

    Ann. Hum. Biol.

    (1984)
  • P. Ganguly

    Progressive decline in stature in India: a study of sixty population groups

  • T. Garland et al.

    Phylogenetic analysis of covariance by computer-simulation

    Syst. Biol.

    (1993)
  • T.J. Garland et al.

    Procedures for the analysis of comparative data using phylogenetically independent contrasts

    Syst. Biol.

    (1992)
  • S.J. Gaulin et al.

    Human marriage systems and sexual dimorphism in stature

    Am. J. Phys. Anthropol.

    (1992)
  • S.J.C. Gaulin et al.

    Sexual dimorphism in weight among the primates: the relative impact of allometry and sexual selection

    Int. J. Primatol.

    (1984)
  • J. Gillin

    The Barama River Caribs of British Guiana

    Papers Peabody Mus

    (1936)
  • W.W. Greulich

    The growth and developmental status of Guamanian school children in 1947

    Am. J. Phys. Antropol.

    (1951)
  • J.P. Gray et al.

    Height and sexual dimorphism of stature among human societies

    Am. J. Phys. Anthropol.

    (1980)
  • R.L. Hall

    Sexual dimorphism for size in seven nineteenth century northwest coast populations

    Hum. Biol.

    (1978)
  • P.H. Harvey et al.

    The Comparative Method in Evolutionary Biology

    (1991)
  • B.H. Heath et al.

    Growth and somatotype patterns of Manus children, territory of Papua and New Guinea: application of a modified somatotype method to the study of growth patterns

    Am. J. Phys. Anthropol.

    (1971)
  • J. Hiernaux

    La diversité humaine en Afrique subsaharienne

    (1968)
  • J. Hiernaux

    Variabilité du dimorphisme sexuel de la stature en Afrique Subsaharienne et en Afrique

  • C. Holden et al.

    Sexual dimorphism in stature and women's work: a phylogenetic cross-cultural analysis

    Am. J. Phys. Anthropol.

    (1999)
  • Hooton, E.A., Dupertuis, C.W., Dawson, H. (1955). The physical anthropology of Ireland. Papers. Peabody. Mus. 30, nos....
  • A. Hrdlička

    The Pueblos

    Am. J. Phys. Anthropol.

    (1935)
  • Cited by (87)

    • Three-dimensional morphometric analysis of cervical vertebral endplate anatomy: A systematic literature review

      2022, Interdisciplinary Neurosurgery: Advanced Techniques and Case Management
    • History as evolution

      2021, The Handbook of Historical Economics
    • Sexual dimorphism variation in Fordisc samples

      2020, Sex Estimation of the Human Skeleton: History, Methods, and Emerging Techniques
    • Biological Normalcy

      2023, Annual Review of Anthropology
    View all citing articles on Scopus
    View full text