Macaque Mothers Favor Their Sons


The Trivers Willard Hypothesis predicts that under certain conditions, individuals will bias their investment in offspring differently depending on the sex of the offspring. It is believed that this can be as extreme as infanticide or as subtle as providing different amounts of breast milk.

A new study by Katherine Hinde finds that macaques may do this. However, I think this may be counterintuitive.

Hinde uses data from 106 rhesus macaques (Macaca mulatta) to show that first time mothers produce richer milk when they have sons compared to when they have daughters. She suggests that “[t]his difference seems to reflect the trade offs between the benefits derived from additional investment in sons and the costs of diverting energy from maternal growth and development.”

The reason that I say that this is counterintuitive is this: Previous studies on female bonded primates have shown relatively little bias in investment … Boyd and Silk showed this to be, essentially, a random relationship so while some studies show a bias, others do not, and if you plot out the degree of bias across a range of female bonded primate species you get a cloud of data consistent with the hypothesis that the Trivers Willard “effect” is not strong with these animals. However, the situation with Hinde’s monkeys may be special.

But first, let’s be clear on the Tivers Willard effect itself. It has specific components, which may be stated more than one way, but that need to be examined else it may be misunderstood.

Blogging on Peer-Reviewed Research

Stated most compactly, individuals who are likely to produce high ranking offspring will differentially invest more resources in the sex with the highest variance in reproductive success.

Its the variance part that has to be understood to get the beauty of this idea. Consider mammals. In virtually all mammals, the number of offspring a female is likely to have is not very variable, while the number of offspring a male may have can, in some species, vary a great deal from zero to a much larger number than any female will have.

Under these conditions, it would be ideal to know in advance if your offspring is going to be one of the better producers, and if so, go for a male offspring. If not, go for the sure bet, the female.

To be precise, Trivers Willard states that under conditions where the rank of the offspring can be “known” to the mother, perhaps simply because she can assess her own rank (and higher ranking moms have higher ranking kids) then selection should favor individuals with some mechanism for assessing rank and then biasing investment in offspring. This has been demonstrated in several species.

Here’s what Hinde has to offer:

Milk samples were collected once during peak lactation from subjects of known age and social rank housed in large outdoor enclosures at the California National Primate Research Center (CNPRC). Analyses of milk constituents were conducted…

As found in red deer, there was a significant main effect of infant sex on milk [quality] however, an interaction effect between parity and sex revealed that sons born to primiparous mothers … get an additional gain in [milk quality] …

Here is a very impressive graph summarizing the results:


The difference is greater with first-time mothers, but in both cases sons receive more nutrition leading to greater body mass. In fact, the effect, while there is a fair amount of variation is rather startling.

I think the reason that you don’t seem to see the Trivers Willard effect in many female-bonded species is because males disperse to new groups and spend a lot of time hanging around before they mate with any females. This may mean that females can’t really do as much for their sons as they can do for their daughters, so they end up not biasing investment very much. This does not, however, mean that biasing investment cannot work. I would like to know if this specific effect — of milk production and body weight — can be measured in other female-bonded primates.

HINDE, K. (2007): First-time macaque mothers bias milk composition in favor of sons. Current Biology, 17(22), R958-R959.

TRIVERS, R. L. & WILLARD, D.E. (1973): Natural selection of parental ability to vary the sex ratio of offspring.. Science, 179, 90-2.


  1. #1 The Ridger
    October 17, 2008

    I can’t answer your wonder, but I can offer an anecdote for what little it may be worth. I had a mare who consistently favored colts over fillies, to the point of letting the colt (1) nurse longer and eat from her grain bucket, while weaning the fillies (2) early and forcing us to crib-feed them so she wouldn’t steal their grain, let alone not let them eat hers. Tiny sample, not worth much, but it immediately came to my mind. And since this is the Internet, I’m sharing 😉

  2. #2 Notagod
    October 18, 2008

    Is it possible that the effect is driven by the infant, not the mother? Maybe if the demand for milk by the infant is greater than the available quantity, the production of milk might shift to overloading the nutrients because the quantity cannot or will not be increased. The stress on the mother’s own needs might be balancing between loss of liquids verses loss of nutrients?

  3. #3 greg laden
    October 18, 2008

    Notagod: Good idea, and the role of the infant as a key actor is often overlooked. The amount of milk produced normally is controlled by the infant in primates which is why, I think, the ‘push’ side of this is assumed to be the mother. But that does not rule out a more complex relationship.

  4. #4 Katie
    November 15, 2008

    Primate infants do not “control” milk production- the frequency and degree of mammary evacuation of milk is what influences milk synthesis- and the frequency and degree of mammary evacuation occurs though infant nursing behavior. However mothers have numerous behavioral tactics to control infant access and duration of nursing. Milk production therefore reflects the behavioral negotiations between mothers and infants.

    As such, milk energy density may have another function unrelated to infant growth which would be particularly advantageous for young, growing mothers; the higher fat concentration in milk is likely to satiate sons more quickly (Jenson 1999) and may therefore reduce infant behavioral demand to nurse, ameliorate mother-offspring conflict, and prevent maternal depletion.

    Jenson RG. 1999. Lipids in human milk. Lipids 34:1243-1271.

  5. #5 Greg Laden
    November 15, 2008


    You are correct. What I should have said is that suckling is a major controlling determinant of milk production.

  6. #6 Katie
    November 15, 2008

    Yeah it gets really tricky to figure this stuff out because mom and infant are constantly negotiating investment throughout lactation, trying to pull it toward their own respective optima and investment can shift moment to moment depending on environmental, maternal, and infant conditions. Not that I would go back and select a different topic, but I wasn’t really prepared for how complicated it would be!