Yes, Male Mammals Tend To Be Larger Than Females: A Response to Tombak et al.

Despite the authors’ best efforts, their own empirical evidence supports the very “narrative” they claim to falsify.

Reality’s Last Stand is a reader-supported publication. Please consider becoming a paying subscriber or making a one-time or recurring donation to show your support.

Sexual dimorphism, the condition where males and females of the same species exhibit different characteristics beyond their sexual organs, is a well-established phenomenon in the animal kingdom, particularly among mammals. This difference often manifests in body size, with males typically being larger than females, a phenomenon that can be seen in species ranging from whales to lions to humans. The roots of sexual dimorphism are deeply entrenched in evolutionary biology, where different reproductive strategies of males and females often dictate their physical forms. Charles Darwin, one of the earliest observers of these patterns, noted that in many species size dimorphism can be explained by battles among males for dominance and mating opportunities, which lead to the selection of larger body sizes over successive generations.

Despite this, a recent 2024 scientific paper published in Nature Communications by Princeton biologists Kaia Tombak, Severine Hex, and Daniel Rubenstein purports to “refute” the longstanding “‘larger males’ narrative”—an idea supposedly endemic among biologists that they argue is not supported by empirical data.[1]

While scientific ideas always leave the door open to future refutation, asserting that such a longstanding biological concept has been debunked requires robust and exceptionally compelling new evidence. As astronomer Carl Sagan famously stated, “extraordinary claims require extraordinary evidence,” and challenging a centuries-old biological principle is indeed an extraordinary claim.

However, rather than presenting compelling data, the authors manipulate statistics and dismiss their own findings in favor of a conclusion they seem to prefer.[2] Simultaneously, they hypocritically criticize everyone from Charles Darwin to Robert Trivers—two leading researchers in their field—for upholding a “narrative” mired in “androcentrism.” This criticism appeared in their original preprint, though Nature’s peer review process seemingly softened this language to a “general focus on males” in the literature.[3]

Let’s begin with the central claim of the paper that the authors have falsified the idea that males tend to be larger than females in mammals.

The authors claim that there was no size-dimorphism (monomorphism) as often as there were larger males. Even if this were true, if I told you that x > y, and your big rebuttal to my idea was that, actually, x ≥ y, that additional texture would be closer to replication than falsification. In fact, Tombak et al. found that in mammals, males were larger 45 percent of the time, species were monomorphic 39 percent of the time, and females were larger only 16 percent of the time. You read that right: the category of larger males was the most common, and 2.8 times as many mammalian species had larger males than had larger females.

Figures 1 and 2 from Tombak et al. (2024) (CC license)

Sexual dimorphism causes changes in “size” on many vectors (e.g., length, volume, density, mass, strength, and so on),[4] and wider sexual dimorphism is a complex and multivariate set of differences in traits spanning the body and mind.[5] However, Tombak et al. relied on only body-mass and (worse) body-length to call into question not only the larger-males hypothesis, but the basic tenets of the theory of sexual selection. Yet even their seemingly preferred analysis, which focused on the subset of their species for which body-length data were available, only found that 49.9 percent of species were monomorphic in body-length, 28.0 percent had larger-males, and 22.1 percent had larger females. This still suggests that monomorphism constitutes half of all cases at most, and that species with larger males are at least 27 percent more common than those with larger females.

Tombak et al. also found that mammal species with larger males exhibited, on average, 13 percent greater dimorphism than those with larger females, with the most male-dimorphic species being 2.3 times as dimorphic as the most female-dimorphic species. The authors even reported that, “the famously larger females in [order] Lagomorpha were only so in... rabbits and hares... while... pikas... have more male-biased dimorphism.” This sounds more like myth-busting their own “narrative.” Consistently, every metric underscores the trend that males tend to be the larger sex, both in the presence and magnitude of this difference.

Tombak et al. argue that other scientists often find that mammalian species more commonly have larger males because they do not weight their analyses by bio-diversity. In other words, if an order and family of mammals has a greater number of distinct species, their sex-differences—or lack thereof—should carry more weight. The authors criticize this supposed oversight as “taxonomically biased sampling,” contrasting it with their approach of “balanced taxonomic representation.” Put plainly, the authors are saying that if you were to focus especially on the most species-rich bats, rodents, rabbits, and shrews, you could almost conclude that males do not generally tend to be larger than females among mammals.

I am grateful to have learned about the significant prevalence of both monomorphism and species where females are larger among mammalian orders. However, when someone refers to “mammals,” they typically do not mean to focus solely on, for example, rodents. They intend to discuss patterns across a variety of forms, not just closely related species that might obscure the diversity among mammalian orders.

If the issue was biased research by others, one would expect the goal of Tombak et al. to be gathering accurate statistics across all mammalian orders. Instead, they examined only 17 of the 29 mammalian orders (59 percent) to draw their conclusions. Given the excessive number of bat and shrew species, one could similarly skew data to bolster the claim that mammals tend to be blind. Of course, there is no good reason for this—why not consider weighting the results by biomass or the number of living individuals of each species?

In the recent academic tradition of tearing down great men, Tombak et al. position their work in opposition to great theorists such as Charles Darwin and Robert Trivers. They even go so far as to spuriously imply that no one thought males tended to be larger among mammals until Darwin’s The Descent Of Man in 1871. However, it is likely that most people, based on their experiences with fellow humans, farm animals, and pets, have assumed this of most animals since prehistoric times. Yet Darwin was, by all accounts, open-minded and mild-mannered,[6] and Trivers was a radical leftist[7]—hardly the types one would ordinarily accuse of chauvinism.

Although such an example may be present in Tombak et al.’s extensive list of citations, nowhere in the paper do the authors actually make a specific accusation of other scientists wrongfully projecting the larger-males pattern onto a species for which this does not apply (which would be strange indeed if someone studying a particular animal didn’t bother to notice such a basic pattern). In fact, none of the examples they cite as having used biased orders in their estimates have titles that mention “mammals” generally; they each refer to specific orders such as “ungulates” or “primates,” where the claims about the prevalence of larger-male species may well be accurate. When researchers mention “mammals,” they are likely not limiting themselves to a particular order, regardless of its diversity.

Additionally, the ideological framing of Tombak et al.’s research undermines confidence. This approach brings to mind the critique of the “academic left” by Paul R. Gross and Norman Levitt in their book Higher Superstition.[8] The peer review process seems to have replaced the language in which Tombak et al. make a rhetorical plea for “more nuanced language and thought” to the fact-centric “more nuanced investigation.” From the pre-print to the peer-reviewed version, the social constructivist-sounding “assign a species as dimorphic” became “determine rates of dimorphism.” Other examples of ideological language (or outright falsehoods), such as accusations of “biased predictions,” “blind spots,” and claims that “dimorphism was minimal” in dimorphic species, were removed during peer review. Tombak et al. even close by labeling their work as “low-hanging fruit” and implying that many (presumably sexist) biological “narratives” (specifically naming polygyny) remain to be overturned (with the right statistical and rhetorical bag of tricks, of course).

It seems odd to me that classical sexual selection theory, where males compete for access to females and females selectively choose their mates, is viewed as having a “general focus on males.” Even if researchers in sub-fields where males possess conspicuous natural weapons, like horns, find it beneficial to focus on male competition for mates, does this suggest that these researchers are more chauvinistic than, say, bird researchers who deem it more relevant to focus on female mate-preferences among begging males?

However, Tombak et al. do not engage much with the sexual selection theory they take such umbrage with. For instance, they note species where “males are much larger than females in the beginning of the breeding season but statistically the same size by the end of it.” The authors concede that their methodology favors data from the non-breeding season, which is monomorphic, thus obscuring the reality of sexual dimorphism in such species. This approach overlooks the fact that sexual selection theory predicts species that do not mate year-round may exhibit dimorphism only during mating seasons.

On several occasions, Tombak et al. twist their finding that the larger-males pattern correlated with higher variance in male size as somehow evidence against the reality of sexual dimorphism. The problem is that the sexual selection theory they take issue with tends to go hand-and-hand with the greater male variability hypothesis, which posits that males typically show more variance in many traits. While natural selection for size in a relatively stable environment may lead to a stability (or, low variance) in size, sexual selection can drive a continual “red queen” effect resulting in a shift toward more extreme trait values, creating ongoing variance.[9, 10] Traditional sexual selection theory (not the monomorphic hypothesis) then tends to predict that males who compete for mates through size should display greater size variance than females of the same species.[11, 12, 13] That greater variance in size was also found in the females of the larger-female species, but not in the monomorphic species, may indicate that both males and females compete for mates via size in their respective dimorphic species, in broad agreement with sexual selection theory.

When discussing, for example, Trivers’ parental investment theory, we are not merely describing a simplistic, chauvinistic view that females generally invest more in their offspring than males. In fact, Trivers made a careful ecological argument for why this tended to be the case. To simplify, the mother is typically present during her offspring’s birth, which shapes investment patterns. Trivers also used his theory to accurately predict species where the usual patterns (sometimes including which sex is larger) are reversed (because the selection pressures are reversed).[14, 15, 16] While Tombak et al. seem to challenge the entire framework of parental investment theory among mammals due to their false belief that mammals aren’t sexually size-dimorphic, they fail to consider that their particular monomorphic species may have evolved under selection pressures for evenly shared parental investment. Moreover, they overlook instances where species with larger females could indicate greater paternal investment. Such scenarios would actually support Trivers’ theory, which fundamentally addresses differences driven by ecological selection pressures.

Tombak et al. devote little time to theorizing about the selection pressures that shaped monomorphic and larger-female species, much less the majority of species where males are larger, whose provisional explanations they dismiss. To the extent that they discuss others’ ideas, these include major concessions such as sexual selection favoring “agility in combat” in some species, as well as sperm competition in others—explanations which are hardly any less “androcentric” than size-dimorphism, if such things bother you. Tombak et al. position Katherine Ralls as their alternative (rather than additional) hero in the field of biology. However, they themselves quote Ralls stating, “species with little sexual size dimorphism were extremely numerous in the most species-rich mammalian orders.” This implies not that one should over-weight these species-rich orders when speaking of “mammals,” but that one should attempt to explore the ecological theories linking monomorphism with species richness. One might expect that more variance in an order (i.e., species richness) would imply more variance in the traits of the species within it, including size dimorphism. Why else would taxonomists divide the organisms into so many distinct species?

Acknowledging the challenges of achieving a “zero state” in dimorphism, which represents a narrow target, the fact that many species exhibit monomorphism is a significant finding and should not be discounted. However, since evolution is a ruthless game of marginal fitness advantages and the gradual nature of evolutionary change demands this marginal effectiveness, even minor differences are significant. In species with distinct males and females, the concept of perfect monomorphism is almost untenable given how evolution operates: a wing likely began as a mere fraction of a feather, yet it was sufficient to give its bearers a reproductive advantage.[17] Unfortunately, while the current form of the neo-Darwinnian synthesis struggles to predict the precise nature of these differences (due to ecological complexity), it is adept at predicting the direction of their effects (given the acute sensitivity of fitness consequences).

The authors also include data from species with sample sizes as low as N = 18 (nine individuals’ size measurements for each sex), despite acknowledging that small sample sizes likely bias results toward monomorphism. The study suffers from compounded errors due to the small sample sizes across populations within species, species within families (often only one, despite selecting for species-rich families), families within orders, and orders within the class Mammalia. This error is in addition to selection bias, which likely skews the results, as these groups were not chosen randomly. It is unclear why the authors do not provide error bars for their major results, but it is easy to imagine these casting doubt on the accuracy of their conclusion.

In sum, despite the authors’ best efforts, their own empirical evidence tends to support the very “narrative” (though some might call it a basic observation) they claim to falsify. The true take away of Tombak et al. (2024) is that it remains difficult to falsify the hypothesis that males tend to be larger among mammals, even with your thumb on the scale. One gets the sense that a reader will (and is meant to) come away from this paper knowing less about the world than they had before reading it.

Reality’s Last Stand is a reader-supported publication. If you enjoyed this article, please consider upgrading to a paid subscription or making a recurring or one-time donation below. Your support is greatly appreciated.

References

1. Tombak, K. J., Hex, S. B. S. W., & Rubenstein, D. I. (2024). New estimates indicate that males are not larger than females in most mammal species. Nature Communications, 15(1872).

2. Huff, D. (2023). How to Lie With Statistics. Penguin UK. Chicago

3. Tombak, K. J., Hex, S. B. S. W., & Rubenstein, D. I. (2023). New estimates indicate that males are not larger than females in most mammals. bioRxiv.

4. Sell, A., Hone, L.S.E., & Pound, N. (2012). The Importance of Physical Strength to Human Males. Human Nature, 23, 30–44, Table 1.

5. Stewart-Williams, S. (2019). The Ape That Understood The Universe: How The Mind And Culture Evolved, 310. Cambridge University Press.

6. Oatley, K. (2018). Our Minds, Our Selves: A Brief History Of Psychology, 40. Princeton University Press.

7. Pinker, S. (2016). The Blank Slate: The Modern Denial Of Human Nature (2nd ed.), 111. Penguin Books.

8. Gross, P. R., & Levitt, N. (1998). Higher Superstition: The Academic Left and its Quarrels with Science (2nd ed.). The Johns Hopkins University Press.

9. Ridley, M. (1993). The Red Queen: Sex And The Evolution Of Human Nature, 66-67, 138-139, 144-145, 341-342; 133. Harper Perennial.

10. Miller, G. (2000). The Mating Mind: How Sexual Choice Shaped The Evolution of Human Nature, 14, 20. Doubleday.

11. Pinker, S. (2016). The Blank Slate: The Modern Denial Of Human Nature (2nd ed.), 344-345. Penguin Books.

12. Miller, G. (2000). The Mating Mind: How Sexual Choice Shaped The Evolution of Human Nature, 14, 20. Doubleday.

13. Pinker, S. (2005, February 14). Sex Ed. The New Republic.

14. Pinker, S. (2016). The Blank Slate: The Modern Denial Of Human Nature (2nd ed.), 252. Penguin Books.

15. Stewart-Williams, S. (2019). The Ape That Understood The Universe: How The Mind And Culture Evolved, 68-73. Cambridge University Press.

16. Dawkins, R. (2016). The Selfish Gene (4th ed.), 182-184, 192-194, 202-204, 210. Oxford University Press.

17. Dawkins, R. (2021). Flights Of Fancy: Defying Gravity By Design & Evolution, 106, 236-237, 242, 251-252, 254, 256, 258.