Boy Milk vs. Girl Milk

At the grocery store, I can purchase a variety of infant formulas. For example, there are formulas for very low birth weight infants, soy-based formulas for infants with dairy allergies, and low-sodium formulas for infants who need restricted salt intake. But should I be able to buy Boy Formula or Girl Formula? The answer is… maybe. Recent research has shown that in deer, monkeys, and humans, mothers make different milk for sons and daughters.

No two mammalian species produce identical milk. This is because for each species, milk synthesis reflects the environment, phylogeny, behavioral care, and developmental needs of the young (Oftedal and Iverson 1995).  In terms of environment, the mother’s diet, the climate, and the community ecology can all be associated with milk composition and volume. Closely related species are more likely to produce similar milks due to shared genes. Behavioral care relates, in part, to how often the infant nurses. Young parked in nests, burrows, dens, or roosts while their mothers forage for food, need rich milk to sustain them until the next feeding.   

 

But the function of mother’s milk is for the development of the young. Young that grow fast need protein, young that grow fat need fat (duh), and young that grow slowly need sugar for behavioral activity. Young also need minerals, vitamins, immunofactors, hormones, and water in milk- and all of these will vary among species as well. In this way milk reflects the developmental priorities of the infant.

But within many mammalian species, males and females develop along divergent trajectories and at differing rates. And when these differences occur during early post-natal life, we can hypothesize that differences in mother’s milk are a contributing factor. An evolutionary perspective provides crucial insights for understanding why there are differences between males and females, and why mothers may bias investment in favor of sons or daughters.

When One Became Two… 
So that Two Could Make One

Asexual reproduction is the process by which offspring come from a single parent and inherit 100% of the parent’s genetic material. It’s a really boring way to say “cloning.” And for billions of years of life on earth, bacterial organisms relied exclusively on asexual reproduction to pass their genes into the next generation. 

by Miles K

But “cloning” has some drawbacks, and about a billion years ago sexual reproduction evolved (bow chikka wow wow!). Now there were two sexes, “pointers” and “setters”, so to speak. Sexual reproduction, because the genes of two parents are combined, minimizes the accumulation of deleterious mutations and produces much more variation among progeny. And its natural selection, acting on heritable variation, that generates adaptations. More variation equals more opportunities for nature to select. In this way sexual reproduction allowed substantial and rapid divergence of organisms, from which plants and animals eventually evolved.

 Pavel Filonov. Animals (1930)

Two Steps Forward, One Step Back

But sexual reproduction also has some disadvantages, because in most sexually reproducing species, one parent ends up doing a majority of the reproductive heavy lifting.

And in mammals, it’s the ladies. 

The time and energy costs of pregnancy and lactation are substantial, much more so than the production of sperm. For mammalian young to survive, there are no real short-cuts, so females have made a substantial commitment from conception until the young are independent. And for each conception, most (but not all) mammalian females are investing in the offspring of a single male. 

In contrast, following copulation, a male’s work is largely done. A single male can impregnate many females in a short period of time- becoming very reproductively successful. Prime examples of such mammals are elephant seals, elk, baboons, gorillas, and Ghenghis Khan. But being the dominant male, with a harem of ladies, is not for the faint of heart. It requires being big, being strong, and/or being intimidating. Basically a male needs to be an all-around bad-ass. And for every bad-ass, there are a whole bunch of ninnies, wusses, and pansies that never sire any offspring. 

What does this mean for milk? Well, natural selection favors adaptations that increase “fitness” - a term that refers to the number of copies of one’s genes that get into subsequent generations. And the biggest contributor to fitness is the number of offspring one produces that survive to reproduce themselves. As a mammalian mother, a daughter is a fairly safe bet; daughters that survive to adulthood are very likely to produce young. In comparison, a son has a greater potential return in number of grand-offspring than does a daughter. But sons are a riskier bet because they have a higher probability of being a zero than a hero. 

Mammalian mothers hedge their bets- and produce both sons and daughters. But in many mammalian species, male infants are born bigger and grow bigger during infancy. Researchers have hypothesized that biases in milk synthesis may contribute to the differences in post-natal growth trajectories between sons and daughters… and data are starting to support, in part, these hypotheses. And even though humans aren’t characterized by the polygynous social organization of red deer or hamadryas baboons, signatures of a polygynous mammalian heritage may still be lurking in our milk.

Iberian Red Deer 
(Cervus elaphus hispanicus)

"Aw yeah, sup ladies"
photo by Carlos Urdiales

The first evidence for sex-biased milk, to the best of my knowledge, was the report of a 5-year study in captive red deer of 91 calves and 60 hinds (19 hinds were sampled two seasons, 6 sampled 3 seasons). Landete-Castillejos and colleagues (2005) found that hinds produced ~12% higher milk yield for sons than that produced for daughters. The mean concentration of protein in milk was also higher for sons. The mean concentration of fat and lactose was the same for sons and daughters, but because hinds produced greater volumes for sons, the total fat yield and total lactose yield for sons was greater. Lastly the combination of greater yield and higher mean protein concentration in milk meant that sons received much higher total protein yield.

The authors controlled for a number of important factors (hind mass and identity, birth timing within the season, and study year), however the analysis seemingly did not control for calf mass.* This is important because male calves are heavier at birth and get proportionally heavier throughout the lactation period than do female calves. By not controlling for calf mass in the statistical model, it’s unclear if there was a sex bias per se or whether milk synthesis was scaling to the mass of the calf. Interestingly, milk yield and total yield of milk fat, lactose, and protein were positively correlated with birth mass and the later three correlated with calf growth. This suggests that milk energy transfer is a function of mass and growth, rather than specifically sex. However, the most suggestive clue of a true sex-bias is the significantly higher protein concentration for sons, because it is so critical for gaining lean tissue mass. The higher total fat and sugar yield for sons was a byproduct of greater total yield for the bigger male calves, but the significantly higher concentration of protein for sons reveals that mothers are making different milk for sons, not just more milk.

*Based on description of general linear models in methods section and details provided in Table 2.

In a follow up study of 46 hind-calf pairs, the same research team revealed that mean concentrations of calcium and phosphorus were higher for daughters, but magnesium concentration was higher for sons (Gallego et al. 2009). Although these differences were statistically significant, the biological significance is unclear because the magnitudes of the differences were actually quite small. Importantly though, in this study, calf birth mass was a covariate in the models so the effect on minerals was not because males were heavier at birth. However, the sex difference in growth and milk yield was not replicated from the earlier report discussed above. Moreover, once milk yield was accounted for, the daily production of minerals in milk did not differ as a function of calf sex. Of particular note, minerals in milk are often associated with the casein micelles of milk proteins. If sons get more protein, why then would the calcium and phosphorus concentrations be lower?  

Taken together, the two red deer studies reveal that there can be differences in milk synthesized for sons and daughters, but that such differences are not always present.

Antarctic Fur Seals 
(Arctocephalus gazella) 
& California Sea Lions 

(Zalophus californianus)

I kind of can't handle the cuteness that is the Antarctic fur seal pup.

Two meta-analyses of wild-living phocids have addressed sex-biased milk production. Lunn and Arnould (1997) aggregated data from multiple field seasons and found no evidence that seal mothers favored sons or daughters in nursing behavior, foraging effort, or milk energy transfer (as measured through the gold standard of hydrogen isotope dilution). Although male pups weighed more than female pups, their growth trajectories were seemingly parallel. The greatest divergence in growth trajectories occurred after weaning during juvenility, suggesting that maternal investment during early development was not a major contributor male sexual dimorphism.

Onoand Boness (1996) also aggregated data from multiple field seasons and found something slightly more exciting in sea lions- milk energy transfer was greater for sons… but this effect disappeared once greater male pup size was included as a model covariate. The authors concluded that milk production in sea lion mothers was a function of pup mass, not pup sex. Since males were bigger at birth the mammary gland synthesized milk to sustain that pup mass and growth trajectory. Most interestingly, though, growth in male pups was seemingly enhanced because they had a lower resting metabolic rate than female pups. Whereas female pups burned more energy on body maintenance, male pups got a “bonus” energy allocation to growth from their lower resting metabolic rate. 

 Selection in action

Because both studies used isotopic methods for inferring caloric transfer via milk, we know little about the composition or volume of milk being produced by mothers of sons vs. mothers of daughters. However, the sea lion study demonstrates that mammalian young are not passive recipients of mother’s milk. Physiological processes within the young may differ between the sexes influencing the utilization of ingested milk.

Rhesus Monkeys 
(Macaca mulatta)

Sex-biased milk synthesis has also been documented in captive rhesus monkeys. At the California National Primate Research Center, mothers in the outdoor breeding colony synthesize different milk for sons and daughters. In the interest of full disclosure, the results reported here come from my own research program, so two things are obvious: I likely have unconscious biases and it’s the most magnificent research ever. 

 by Carl Buell

In an early paper from 2007, I showed in a sample of 106 mother-infant pairs that milk for sons was significantly higher in percent fat but there was an interaction with maternal parity. First-time mothers, known as primiparous mothers, that produced a son had milk that was 7.7% fat, whereas mulitparous mothers (females that had reproduced previously) rearing sons produced milk that was 6.5% fat. In contrast, primiparous and multiparous mothers rearing daughters produced 5.3% and 6.2% milk fat respectively. Primiparous mothers of sons also produced significantly higher protein concentration in milk (2.5% vs. ~2.1% in all other groups). Unfortunately, I failed to co-vary infant mass or analyze milk yield. Milk yield is important because, in general, milk energy density and milk yield are negatively correlated, so richer milk may be counter-balanced by there being less of it.

Why would primiparous mothers differ from multiparous mothers? Primiparous mothers initiate reproduction while they are still growing, so they are simulatenously paying the energetic costs of growth and reproduction. Additionally because they are not yet full size, they have fewer bodily resources to mobilize for milk synthesis. Lastly, infants must attain a threshold size to survive, so first-born infants are proportionally bigger relative to the mother’s size (Hinde 2007 & 2009). This means that primiparous mothers have higher operating costs and fewer resources to rear a relatively bigger baby. No easy feat. 

 photo by David Lewis

Natural selection favors adaptations that maximize lifetime reproductive success, known in evolutionary biology as “life-history theory.” Over-investing at the outset of the reproductive career is unlikely to be favored by natural selection. Investing too much on the first reproductive effort can potentially stunt growth, lengthen the inter-birth interval, and reduce survival. All these things are likely to generate a frowny face from natural selection. One solution for primiparous mothers is to cut corners with daughters, because it shouldn’t be too costly in terms of the number of grad-offspring she’ll produce. And that’s what I found; reduced milk energy content, and first-born daughters were quite small.

But the sons of primiparous mothers would have to compete with the strapping lads of multiparous mothers. A subset of primiparous mothers rearing sons cut bait: first-born sons were much more likely to die than expected from the population mortality rates for either male infants or first-born infants. But for the rest of first-born males, primiparous mothers seemingly pumped them full of fat and protein and at peak lactation they were comparable in body size to the sons of multiparous mothers. To be clear, mothers are not consciously making Sophie’s choice investment “decisions.” Rather physiological adaptations that are sensitive to energy balance and infant characteristics are the underlying mechanisms that contribute to sex-biased milk synthesis.

 Technically Barbary macaques, but this picture is freaking awesome.

In 2009 I published a follow-up analysis. Sons of primiparous mothers, arguably the most interesting group in the 2007 study, were also the least represented in the dataset. I recruited an additional seven primiparous mother-son dyads, increasing the sample size (N=113). I also included measures of milk volume (the amount of milk mothers produced in a standardized period of time) and co-varied infant body mass and maternal body mass in all statistical models. The sex effect, and interaction with parity remained. Primiparous mothers produced 24% richer milk for their sons than did primiparous mothers for their daughters, and 10% richer milk than the milk that multiparous mothers synthesized for their sons.

But interestingly, mothers of daughters made 21% more milk. Because milk yield was measured gravimetrically in grams, and milk energy density was calculated as kcal/g, the product of the quality and quantity of milk could be combined into an “Available Milk Energy” parameter standardized among all mothers. Greater available milk energy was associated with heavier mothers and heavier infants, but infant sex and maternal parity (and the interaction) were no longer significant. Sons are heavier than daughters, and multiparous mothers are heavier than primiparous mothers, so infant sex and maternal parity remained important, but were secondary to maternal and infant body mass. However, much like in the red deer, the milk “recipe” for sons and daughters differed.

 Humans 
(Homo sapiens)

Two preliminary studies indicate that there are sex biases in milk synthesis in human mothers. In 2010, Powe et al. reported that among 25 Boston-area women, mothers of sons produced ~25% higher energy density in milk than mothers of daughters. However there are several caveats to this study- the vast majority of mothers were on their first pregnancy so it may be that this was an infant sex & primiparity effect similar to the one found in the rhesus monkeys. Additionally, the milk energy density calculation the authors used underestimated the energy value of protein by about a third, but we don’t know how much protein concentrations differed between the milk for sons and that for daughters. Lastly the study only investigated milk composition, not milk volume. Regardless of these limitations, the results from this study are the first, to my knowledge, to indicate that there are differences in milk energy density between milk produced for sons and milk produced for daughters in humans.

More recently, Fujita and colleagues (2012) revealed sex-biases in the milk fat concentration among 72 women in rural Kenya. On average, mothers of sons produced significantly higher fat concentrations in milk. However there was an interaction between infant sex and maternal socio-economic status. Relatively wealthier women- women whose household owned more land and more dairy animals (such as camels, cattle, goats, and sheep)- produced milk with higher fat concentrations for sons. But relatively poorer women produced higher fat concentrations for daughters. Interestingly, higher socio-economic status was associated with lower mean fat concentrations in milk, possibly because such mothers were producing more milk. However investigating milk volume among Ariaal people in Kenya presents ethical and cultural challenges. The data from this study suggests that sex-biases in milk synthesis among humans may be sensitive to cultural and economic factors. These results provide support for the Trivers-Willard hypothesis (1973) that mothers in the best condition should favor sons because they will likely have high reproductive success, but that women in poor condition should favor the “safer” investment in daughters. 

 Ariaal women in Kenya

Conclusions

Taken collectively, these studies reveal that mothers make milk differently for males and females. Other parameters, such as infant mass and maternal mass, parity, and access to resources, can also play mediating roles. However to date relatively little research effort has been dedicated to investigating sex-biased milk synthesis. Perhaps because the topic sounds dangerously close to “milk is sexist.”  But that does not absolve us from the responsibility of understanding the many ways in which milk synthesis reflects the developmental priorities of the infant. This is especially relevant for commercial infant formulas or use of donor milk to feed at-risk NICU babies. 

Many, many questions remain. For example, is there biological significance for infant development in the different milk “recipes” for sons and daughters? And to what extent are other constituents, not yet investigated, similar or different in milk produced for sons vs. daughters? How do sons and daughters differently utilize milk from the mother? What are the signals and mechanisms through which the mammary gland “knows” its synthesizing milk for a son or a daughter? Clearly this is not an exhaustive list of the potential questions that lactation science can tackle, among the many others, but does highlight exciting potential directions for research in nutrition, evolutionary biology, anthropology, and animal science.

Citations:

Fujita M, Roth E, Lo YJ, Hurst C, Vollner J, Kendell A. 2012. In poor families, mothers' milk is richer for daughters than sons: A test of Trivers-Willard hypothesis in agropastoral settlements in Northern Kenya. Am J Phys Anthropol. 149(1):52-9.

Gallego L, Gómez JA, Landete-Castillejos T, Garcia AJ, Estevez JA, Ceacero F, Piñeiro V, Casabiell X, de la Cruz LF. 2009. Effect of milk minerals on calf gains and sex differences in mineral composition of milk from Iberian red deer (Cervus elaphus hispanicus). Reproduction. 138(5):859-65.

Hinde K. 2009. Richer milk for sons but more milk for daughters: sex biased investment during lactation varies with maternal life history in rhesus macaques. Am J Hum Biol 21:512-519.

Hinde K. 2007. First-time macaque mothers bias milk composition in favor of sons. Curr Biol 17:R958-R959.

Landete-Castillejos T, Garcia AJ, López-Serrano FR, Gallego L. 2005. Maternal quality and differences in milk production and composition for male and female Iberian red deer calves (Cervus elaphus hispanicus). Behav Ecol Sociobiol. 57:267-274.

Lunn NJ, Arnould JPY. 1997. Maternal investment in Antarctic fur seals: evidence for equality in the sexes? Behav Ecol Sociobiol. 40: 351-362

Oftedal, O. T. & Iverson, S. J. (1995). Phylogenetic variation in the gross composition of milks. In Handbook of milk composition, (ed. R. G.Jensen), pp. 749–789. Academic Press, New York .

Ono KA, Boness DJ. 1996. Sexual dimorphism in sea lion pups: differential maternal investment, or sex-specific differences in energy allocation? Behav Ecol Sociobiol. 38:31-41.

Powe CE, Knott CD, Conklin-Brittain N. 2010. Infant sex predicts breast milk energy content. Am J Hum Biol. 22(1):50-4.

Trivers RL, Willard DE. 1973. Natural selection of parental ability to vary the sex ratio of offspring. Science. 179(4068):90-2.