Among birds, alloparental care almost always entails provisioning. Costa Rican magpie jays who have never reproduced themselves bring back beakful after beakful of food to fledglings who flit to conspicuous perches beside their nest and beg for it. Avian helpers often provide more food than the chicks’ own parents do. Some allomaternal feeding involves reciprocal arrangements, especially in mammals with co-suckling. Among cooperatively breeding mice, lions, elephants, or brown hyenas, a co-mother will allow the young of a co-resident mother (who may be her sister or mother) to join her own young at her teats, freeing each female in turn to forage and ensuring shorter gaps between snacks for pups.19
In the case of house mice, females able to set up nestkeeping with a sister enjoy significantly higher reproductive success than either those who choose an unrelated female or those who rear their young alone. Cooperative as this arrangement sounds, sometimes pregnant female house mice kill several of their partner’s pups, with the effect of increasing the amount of milk on offer when their own young are born. Both females still gain from cooperating, but the killer benefits more, at the expense of her partner.20 In other cases, helping is more of a one-way street. Subordinate wolf, wild dog, or meerkat females who have never (and may never) conceive sometimes undergo a “pseudopregnancy,” with a swollen belly and mammary glands. Then once the alpha female’s pups are born, these lactating nulliparas are used as wet nurses, secreting milk for the alpha’s pups. One wild dog who had never given birth herself spontaneously began to lactate ten days after the alpha female’s pups were born, and this allomother suckled them more than their own mother did.21 It is not known why this happens, but by becoming a wet nurse the subordinate may increase her chance of being tolerated in the group. And eventually, she may have an opportunity to conceive.
Among cooperatively breeding canids, wolves, coyotes, red foxes, silver-backed jackals, Semyen foxes, Indian dholes, or—my personal favorites—African wild dogs, allomothers (and also mothers) consume and partially digest prey, then return to the den to regurgitate this special “formula” into the eager mouths of pups. Youngsters who until then were nourished entirely on breast milk rush forward to lick the donor’s muzzle. The lactating mother may be fed regurgitated meat as well.22 Even less appetizing, but every bit as important, are the caecotrophes (partially digested fecal pellets) that naked mole rat alloparents excrete for nearly weaned pups. Along with the preprocessed nutrients, pups ingest endosymbiotic gut flora needed to digest cellulose in the mole rat’s staple diet of fibrous underground tubers.23
In cooperatively breeding canids like these African wild dogs, adults return from hunting to regurgitate predigested meat into the mouths of eager pups. (Chris Johns/National Geographic Image Collection)
The importance to immatures of being provisioned during this highly vulnerable weaning phase is huge, for weanlings are still too small to compete successfully for food with older group members. Across cooperatively breeding species, alloparents continue to subsidize small but rapidly growing young long after they have been fledged or weaned. The ornithologist Tom Langen was the first to systematically quantify prolonged dependence among cooperatively breeding birds. Analyzing data for 261 species of passerine birds, Langen discovered that species did not differ in how long they incubated their eggs or fed nestlings. The duration of postfledging provisioning, however, was twice as long (just over 50 days) in the cooperatively breeding species compared with species without help (20 days). Duration of postfledging dependence for only-occasionally-cooperative bird species fell neatly between these two extremes (30 days).24
It is not yet clear whether animals that grow up slowly are more likely to evolve cooperative breeding, or whether cooperative provisioning permits prolonged dependence and with it a longer preadult life phase.25 Most likely it is a bit of both, since these traits could coevolve. What is apparent is that young who are protected from starvation by cooperatively breeding parents have the luxury of growing slowly and can use the extra time to master complex subsistence skills. Like children learning to make a living, the crested magpie jays that Langen studied have to learn to recognize and catch appropriate insect prey, and to identify and gather palatable berries. In other words, these beguiling jays must learn how to become hunters and gatherers in their own right.26
The correlation between cooperative breeding and long post-weaning dependence is not as well documented for mammals as it is for birds. Still, we know that alloparental provisioning offers valuable learning opportunities at the same time that it also subsidizes longer learning phases for immatures.27 Young lions, wild dogs, and other social carnivores rely on game brought back by older group members to keep from starving while they gradually, awkwardly, master such arts as stalking and downing highly mobile, elusive, and often dangerous prey. The only way weaned but still inexperienced immatures survive their early bungling is through the generosity of other group members, who allow youngsters privileged access to carcasses.28
Among some cooperative breeders, provisioning by alloparents goes a step further. In addition to providing immatures opportunities to learn subsistence techniques for themselves, alloparents actually act as mentors. The best-documented instances of animal teaching occur among pied babblers, a species of ant, and meerkats—animals with lots of alloparental care but few brains and even less general learning. In the ant case, mentors merely reflexively guide naive nestmates to food. Meerkat alloparents actually help pups learn by preprocessing prey for them to practice with.
In response to begging calls from pups, meerkat helpers bring small prey and then remain nearby to supervise how pups handle the meal. The most striking case of monitoring involves scorpions. Even though scorpions have stingers that can deliver dangerous neurotoxins, they account for some 5 percent of the meerkat’s diet. When pups are very young, helpers kill scorpions before delivering them. As pups mature, the helper delivers live scorpions but first disables them by removing their stinger. Gradually, as the pups gain experience, helpers deliver intact scorpions. Should the scorpion scamper off, the helper recaptures it and hands it back to the pup. As Cambridge University researchers Alex Thornton and Katherine McAuliffe point out, teaching in meerkats “can be based on simple mechanisms without the need for intentionality and the attribution of mental states.”29 Nevertheless, there is little question that these alloparents exhibit a powerful urge to respond to the needs of youngsters. In some species, alloparents take dedication even further: They forgo breeding careers altogether in order to help rear the young of others.
SHERMAN’S “EUSOCIAL CONTINUUM”
In eusocial (truly social) animals, alloparents routinely put survival of the group or hive ahead of their individual interests. To qualify as eusocial, organisms must meet three criteria: (1) they must live in groups with overlapping generations; (2) they must provide alloparental care; and (3) they must divide reproductive labor to such a degree that many (or even all) helpers never breed. In the most extreme cases, helpers belong to sterile castes.30 They not only never breed, they are anatomically unequipped to do so.
In the view of the Cornell University zoologist Paul Sherman, animals with shared care can best be understood by locating them along a continuum. At one end are groups where many or most members eventually breed, and at the other end are groups where successful reproduction is concentrated—or “skewed”—to favor the ovaries of just a few females, perhaps even a single especially fecund female, as in the case of a honeybee queen. At the skewed end of the continuum, nonreproductives completely subordinate their direct reproductive interests to those of the group. Many entomologists regard eusociality as a distinct category, but here I follow Sherman, treating eusocial societies as points along a continuum with varying degrees of reproductive skew.31
Social insects such as ants, termites, and the more highly organized species of bees and wasps, along with a rare mammalian case, the naked mole rat, live in large colony societies with the kind of extreme reproductive skew tha
t qualifies them as eusocial. Unlike cooperatively breeding birds, eusocial alloparents do all of the provisioning. Worker ants lug prey back to the hive, then gently place helpless larvae atop their food source. Or the larvae rock their heads and beckon with dancing mandibles to induce alloparents to regurgitate nutritious syrup into their waiting maws. Bee larvae are either fed directly this way or else “bottle-fed” from specially constructed overhead wax pouches filled with pollen and honey.32
Eusocial species with extremely skewed reproduction are distinguished from other cooperative breeders by their typically larger group sizes and their unusually strict, often lifelong, division of labor. Honeybee colonies provide a good example. The grubs that are fed a special concoction called royal jelly develop into queen bees who devote their long lives to producing most or all of the colony’s young, while hard-working nonbreeders tend them. Among some eusocial insects like fire ants, workers are permanently sterile. In others, a few workers, should they live long enough or be so lucky, may get a chance to breed. But the distinguishing feature of eusocial insects is that helpers are not just biding their time or waiting out adverse conditions until they manage to breed themselves. Rather, they spend their entire lives tending and feeding the offspring of one or several superfecund females—often their own mother or sister. Untold numbers actually give their lives for the cause. Per capita death rates for workers defending or provisioning colonies in which they themselves have never bred are staggering.33
Such rigid division of labor goes way beyond the allomaternal dedication found in cooperatively breeding birds or mammals, with one exception. The exceptional case is the naked mole rat, the only vertebrate with a breeding caste and morphological differences between castes.34 Efforts to resolve the puzzle of eusociality in social insects led to the development of the first rigorous theoretical explanation for the evolution of cooperative breeding, known as kin selection.
Looking more like a bad dream than a mammal, naked mole rats (Heterocephalus glaber), with their hairless, wrinkled hides and protruding teeth adapted for tunneling through desert hard-pan, come closer than any mammal known to the skewed reproductive success characteristic of eusocial insects. Fewer than 5 percent of mole rats ever have an opportunity to breed. Females who manage to dominate other group members and achieve breeding status undergo massive morphological changes, including lengthening of the lumbar vertebrae, permitting the “queen” (the bulgingly pregnant female above) to produce large litters. Even more remarkably, male and female mole rats who achieve breeding status develop significantly more brain cells than subordinates, especially in the hypothalamus. Differences in brain morphology between breeding and nonbreeding females are more pronounced than any differences between the sexes. (Jennifer Jarvis)
HAMILTON’S RULE EXTENDS BEYOND KIN SELECTION
Owing to peculiar asymmetries in the genetics of haplodiploid insects, full sisters in ants, bees, and wasps share three quarters of their genes by common descent, instead of the one half typical of full siblings. In 1964 this extra dollop of genetic relatedness caught the attention of the evolutionary theorist and wasp specialist William D. Hamilton. Hamilton hypothesized that a higher-than-usual degree of genetic relatedness between the queen and her sisters in species like honeybees made it especially advantageous for workers to opt out of reproducing themselves, since they could increase their genetic representation in succeeding generations indirectly by investing in their superfecund sister’s young instead of directly in their own. Rather than breed oneself, why not help the queen? She not only carried the same genes by common descent, but as long as she was protected and provisioned by her kin, she could remain safely inside the hive, using her specialized anatomical equipment to pump out eggs at a rate of 5 or 6 a minute, as many as 2,500 in a day. By contrast, a solitary bee trying to breed on her own would be hard put to reproduce at all, much less produce a vast number of offspring likely to survive.
Put this way, altruistic worker bees participate in a win-win scenario benefiting all hive members. It makes perfect evolutionary sense for individuals to behave cooperatively in ways that enhance the reproductive success of relatives with whom they share such a high proportion of genes by common descent. Hamilton termed the combined effects of an animal’s behavior on his or her own direct reproductive success plus the indirect effects on the fitness of close kin “inclusive fitness.”
The logic behind such kin selection is summarized in a deceptively simple expression: C < Br. According to what has become widely known as Hamilton’s rule, altruistic helping should evolve whenever the cost to the helper (designated as C) is less than the fitness benefits (B) obtained from helping another individual who is related by the value of r.
Hamilton’s rule is widely accepted today. Almost all evolutionary biologists assume that without sufficiently close genetic relatedness and an appropriate ratio of benefits to costs, caretaking and other cooperative propensities that do not directly increase the helper’s own reproductive success would not have evolved. By now, however, especially close degrees of relatedness between the helper and the helped such as are found in the haplodiploid social insects or, for that matter, among chimeric marmosets, seem more nearly special circumstances that lower the threshold for the maintenance of high levels of cooperation through evolutionary time rather than an essential condition without which they could not persist.35 For one thing, many eusocial creatures are neither haplodiploid nor chimeric. Termites are a case in point. They are eusocial even though workers are not necessarily super-related to the queen. Not only is close kinship less essential than was at first assumed, but “helping behaviors” themselves are not always quite as altruistic as they first appear.
Even though kinship is not essential for the persistence of cooperation, clearly it matters. The neural and physiological underpinnings for helpful behaviors first evolved in the context of mother-infant relationships and subsequently became extended to others in groups of closely related animals. Degree of relatedness often makes a difference in whether helpers help at all, as well as in how far individuals will go to help. The more alloparental assistance matters for fitness, the more likely kinship is to make a difference.36 Studies of the nondescript brown birds called dunnocks provide one of the best-documented examples.
As with many cooperatively breeding birds (and similar to many traditional human societies), dunnocks have very flexible breeding systems. A female may breed either monandrously (with just one mate) or polyandrously (with several males), just as males may breed with either one or several females. Over the course of their lifetimes, the same individuals may mix and match these various permutations, but so far as caretaking goes, relatedness still matters. When females mate with several males, possible fathers calibrate the amount of food they bring back to chicks according to when and how often they copulated, and hence according to that male’s probability of paternity.37 Such male propensities help explain why some cooperatively breeding females who find themselves short on helpers engage in extrapair copulations with other males in their group, trading copulations for help, as has been reported for African superb starlings (and of course some humans).38
Whether dunnocks, brown hyenas, or Hadza foragers, it is a reasonable bet that helpers provide more food to the infants they feel more closely related to.39 But in the early years of the Hamiltonian era, kin revelations seemed so powerful that they overshadowed other considerations. Today, with more information available, it is increasingly apparent that once the neural and physiological underpinnings for helping behavior were in place, helpers did not necessarily have to be close kin. Researchers are paying more attention to other reasons, besides genetic relatedness, that explain why helpers help in any particular situation. Male superb fairy wrens of Australia, who help rear chicks that they are only occasionally related to, provide a spectacular segue into this topic.
Tiny, wag-tailed, insectivorous birds, constantly hopping about on the ground and flitting from spot to spot, superb f
airy wrens can be hard to get a good look at. Even so, it is difficult to miss the stunning flashes of blue from male feathers that catch and reflect light like the avian equivalent of iridescent blue Morpho butterflies. Superb (and they really are) fairy wrens are typically found in groups with a single breeding female assisted by one to four males—the territory owner plus younger males, often sons of the breeding female who help defend the territory as well as protect and provision her chicks. Because territories are in such short supply and a fairy wren without a territory has little chance of surviving long, females driven out of their natal groups by their mothers are compelled to take the first opening on offer, rather than holding out for the best and the brightest mate. But no matter. As it turns out, the owner of her territory only fathers a fraction of her offspring anyway.
Once DNA testing became standard issue in ornithological toolkits, researchers were stunned to discover that the vast majority (over 75 percent) of fairy wren chicks were sired by outside males. Female promiscuity notwithstanding, all care was provided by males in the mother’s group. When the Australian ornithologists Michael Double and Andrew Cockburn attached tiny radio transmitters to females, they discovered that just before daylight, fertile females were flying off for quick liaisons, then returning just as quickly to the territory where their mate and other helpers remained.40 In a paper fetchingly titled “Pre-Dawn Infidelity: Females Control Extra-Pair Mating in Superb Fairy Wrens,” Double and Cockburn hypothesized that females unable to choose the male that best suited them when selecting a territory subsequently take matters into their own wings. Her partner makes the best of his cuckolded lot by helping rear her chicks anyway. After all, some unknowable fraction of her offspring is still likely to be sired by him.
Mothers and Others Page 21
