Stephan J Schoech. American Scientist. Volume 86, Issue 1. Jan/Feb 1998.
If you are in the right habitat in southcentral Florida, you can frequently watch groups of Florida scrub-jays. If you make a hissing sound-a technique that birders call pishing-you might attract half a dozen of these birds, which resemble the common blue jay in size and shape, minus the crest. The blue wings and tail combined with a mostly blue head, a gray-brown triangular “cape” on a blue-gray back and a streaked blue-gray “necklace,” or breast band, identify the scrub-jay. The pishing calls in the birds from the local territory, where they work together to raise offspring, an approach called cooperative breeding.
In cooperatively breeding species, one or more helpers assist a breeding pair in rearing young. Alexander Skutch, an American who has observed birds in Costa Rica for more than 50 years, first described cooperative breeding in 1935. Since then, the overwhelming majority of research in this field has examined the ultimate, or evolutionary, factors behind this approach to reproduction. In brief, many of those studies suggest that young birds of species that have relatively strict habitat requirements often remain at home when all of the available nesting territories are occupied. Remaining at home sets the stage for helping to rear closely related nestlings (most are full siblings). If the helping effort results in survival for more young that share genes with the helper, the nonbreeding helper may gain some measure of evolutionary success by helping to get some of its genes into the population, where they have a chance of surviving and reproducing. This is termed indirect fitness, and it might be thought of as making the best of a bad situation.
Few studies have addressed the proximate, or causal, mechanisms that facilitate cooperative behavior in birds, although many investigators have suggested that physiological mechanisms could help explain either delayed breeding or helping behavior. Biologists are just beginning to explore that area, largely because many of the techniques required to examine the reproductive or energetic physiology of freeliving animals in the field have only recently been developed.
The Florida scrub-jay may be the most studied cooperatively breeding species in the world. In 1969, Glen Woolfenden of the University of South Florida, John Fitzpatrick of Cornell University and subsequently their colleagues and students began studying a population of these birds at Archbold Biological Station. In 1987, my colleague, Ron Mumme of Allegheny College, started working with a population immediately adjacent to Woolfenden and Fitzpatrick’s. Fortunately, Mumme chose me as his field assistant in 1989, and I have been wintering in southern Florida ever since.
The Florida scrub-jay lives only on the state’s peninsula. On average, a group consists of three birds, but it can range from two to eight. They work together to defend a territory that averages about nine hectares. Although the helpers (nonbreeding birds) do not assist in building nests, incubating eggs or brooding nestlings, they perform many tasks, including defending the territory, acting as sentinels and defending against predators, and the vast majority also provide food for the nestlings. More than 50 percent of the helpers are one-year-old birds, but a few of them may be as old as seven, and male and female birds perform helping duties with equal likelihood. In general, helpers are the offspring of the breeding pair from previous years. Florida scrub-jays are monogamous and-with the exception of an occasional “divorce”pairs remain together until one member dies. To switch from helper to breeder, males generally stay on their home territory and either “bud” a small section on its periphery, inherit it when the breeding male dies or find an opportunity when a neighboring males dies. Females usually disperse more widely in search of breeding opportunities.
Although some populations have adapted to suburban living, the majority live in dry oak scrub dominated by vegetation that is usually less than two meters tall. Suitable scrub habitat is maintained by fire. Fire suppression leads to taller oaks, which seem to favor blue jays that apparently exclude scrub-jays. The combination of fire suppression and increased human habitation of the peninsula has made the Florida scrubjay a threatened species.
Readers may be most familiar with cooperative breeders-including social ants, bees, wasps and naked-mole rats-in which only one female, the queen, can breed. Other females are not capable of breeding, so they help instead. Consequently, my first question was: Are helper scrub-jays physiologically capable of reproducing? As I shall show, a combination of hormonal measurements together with observations of these birds indicates that at least some of the male and female helpers could reproduce, given the proper circumstances. Second, I wondered if a hormonal mechanism facilitates helping behavior. The hormone prolactin influences parental behaviors, including nest building and other aspects of caregiving, in many animals. Prolactin, therefore, was a likely candidate for mediating helping behavior in scrub-jays.
Dissecting the Delay
If helpers could not reproduce, differences might exist in the reproductive axes of breeders and helpers. The reproductive axis consists of the hypothalamus (a region of the brain), the pituitary (an endocrine organ immediately below the brain) and the gonads (testes or ovaries). The components of the hypothalamo-pituitary-gonadal axis communicate with one another via endocrine or neuroendocrine secretions. For example, the hypothalamus produces a neurohormone called gonadotropin-releasing hormone, which travels to the anterior pituitary. In response, the pituitary produces and releases luteinizing hormone and follicle stimulating hormone into the blood system. During puberty or at the onset of a breeding season, luteinizing hormone and follicle stimulating hormone initiate gonadal growth and maturation. Later, these hormones maintain gonadal function, including the production of sperm and the development of ovarian follicles. The gonads are the primary producers of the sex steroid hormones, including the best-known ones, testosterone and estradiol, an estrogen. These hormones affect gamete development and maturation, as well as secondary sex characteristics, such as coloration of bare skin (a chicken’s comb or a turkey’s wattles) and specialized courtship plumage in herons and egrets. Testosterone and estradiol can also induce sexual behaviors, including courtship behavior, through their effects on the brain.
During three years of fieldwork, I collected more than 400 blood samples to compare hormone levels of breeders and nonbreeding helpers in search of a physiological basis for reproductive inactivity among helpers. To examine the performance of a helper’s hypothalamus and pituitary, I measured levels of luteinizing hormone. My results show that the luteinizing-hormone levels in both male and female helpers are statistically equivalent to those of breeders. The relatively high levels seen in female breeders during the time when they are nest building, copulating and laying eggs reflects their participation in these activities. Overall, the similarity in luteinizing-hormone levels indicates that a helper’s hypothalamus and pituitary are fully functional.
The condition of a helper’s gonads can be inferred from the levels of sex steroids. Male breeders had higher levels of testosterone than male nonbreeding helpers, but the seasonal pattern of testosterone secretion in helpers mirrors that of breeders. A male breeder’s high level of testosterone during nest building and egg laying might be attributed to participating in those activities, but male helpers show similar increases even though they do not participate at that time. The increased levels of testosterone in helpers may reflect heightened interactions with jays of either sex or solely be a response to the suite of environmental cues that herald the onset of spring and summer. Nevertheless, a male helper’s lack of a mate or its young age might explain the failure of helpers to achieve the same testosterone levels as breeders.
Beyond secreting testosterone, can a male helper’s testes also produce sperm? A few observations suggest that they can. In my population during 1993 and 1994, 10 one-year-old males bypassed the helper stage and ascended directly to breeder-meaning that they occupied and defended a territory with a female. In the end, eight of those pairs built nests and seven of the females laid eggs. Of the seven clutches, three hatched, three were lost to predators and two were infertile. Although the rate of infertile clutches is higher than that of experienced breeders, whether the infertility of the eggs resulted from the absence of viable sperm as opposed to a problem with the female’s ovarian follicle or was the result of unsuccessful copulations owing to the inexperience of the young male cannot be determined.
In contrast to the males, the females’ levels of their primary sex-steroid hormone, estradiol, did not differ statistically between breeders and helpers. That suggests that a female helper’s ovaries function, at least in terms of hormone production and secretion. Also in contrast to the males, the seasonal changes in a female helper’s levels of estradiol do not parallel those of breeders. Female breeders secrete the highest levels of estradiol during the prenesting period, whereas helpers’ highest levels are between the nestbuilding and incubating periods, which may reflect their dispersal in search of a breeding territory. In 1993, I captured 17 female helpers during the nest-building, egg-laying and incubation periods. Of those, the four that were captured away from their home territories had estradiol levels nearly 10 times higher than the 13 helpers that were captured on their home territories. The high levels of estradiol might have prompted dispersal, or the dispersing females may have interacted with potential mates, which could also induce the high levels of estradiol. Paradoxically, the few dispersing females that I sampled in 1994-the only other year that I captured female helpers away from home-did not have elevated estradiol levels.
Additional evidence indicates that helpers are reproductively capable and that they delay breeding only because they lack the opportunity. In each of three years-1989,1994 and 1997-my colleagues and I found two females sharing incubation duties at a nest with seven eggs. Because the maximum clutch size for a scrub-jay is five, both females had clearly produced eggs with the resident male. Moreover, Woolfenden and Fitzpatrick reported a few cases of breeding by one-year-old females. In addition, some males and females remain helpers until they are two, three or older, long after they can breed. Finally, if a female or male breeder’s mate dies, they occasionally revert to helping, which provides further evidence that helping is not solely a function of being too young to breed. As mentioned earlier, some one-yearold males can also breed.
Body Condition and Stress
Based on their studies of austral blackbirds, brown-and-yellow marshbirds and bay-winged cowbirds in Argentina, Gordon Orians of the University of Washington and his colleagues hypothesized that insufficient food resources might explain delayed breeding in those species. This hypothesis predicts that helpers will be in poorer condition and weigh less than breeders and that supplemental food should increase the number of helpers that become breeders. Considerable evidence shows that young birds-and helpers of most species are young birds-are generally less adept at foraging than their elders.
To test the first prediction, I weighed jays over the three years of my study, and the helpers weighed less than same-sex breeders. Before drawing any conclusions, I factored in the effect of a general size difference. Using wingchord dimensions-essentially the length of a bird’s wing, a measurement that is often used as an indicator of size-I found that the size of helpers is 97.3 percent that of breeders. That figure corresponds closely with the body-mass data that show that male helpers weigh 98.0 percent as much as male breeders. Statistical analysis confirmed that the size difference explains the weight differences. So, rather than being in poorer condition, the male helpers are just relatively small, young birds that are not yet fully grown.
The differences between the body masses of female breeders and helpers cannot be explained solely by a size difference. The female helpers wing-chord measurements show that an average helper is 97.5 percent as large as a breeder, but the average helper’s body mass is only 94.1 percent that of an average breeder. A female breeder’s higher body mass may be explained by seasonal variation. At the onset of the breeding season, female breeders and helpers weigh about the same amount, suggesting that they are in equally good condition and that helpers have foraged as efficiently as breeders during the winter months. As the breeding season progresses, the difference in weight develops when female breeders increase their body mass by 11 percent in a matter of weeks as they undergo physiological and anatomical changes in preparation for laying.
In 1993, I tested the second prediction of Orians’s hypothesis by providing nine groups of jays with twice-a-day supplements of dried dog food, peanuts and meal worms. These intelligent birds soon learned that I came bearing snacks, and they would usually be waiting for me at the food station. Nevertheless, the supplemental food failed to increase the proportion of helpers that switched to breeders. That was the case for both males and females-suggesting, again, that opportunity, not body condition, determines whether a helper becomes a breeder.
Another potential physiological explanation for helping comes from Jerram Brown of the State University of New York at Albany, who proposed the psychological-castration hypothesis, which suggests that the presence of a breeding pair “forces” helpers to remain reproductively quiescent. Although Brown did not invoke an endocrine mechanism, John Wingfield of the University of Washington and I-along with our colleagues-postulated independently that dominance interactions could activate the stress, or adrenal, axis, which could suppress breeding. Abundant evidence from many animals shows that numerous stressors, including social conflict, can increase adrenal secretion of the stress hormone corticosterone. (Some groups of animals produce the slightly different steroid molecule, cortisol, and yet others produce both.) Elevated corticosterone levels can inhibit the reproductive axis at many levels, which could force helpers to remain reproductively inactive.
As is true for most cooperatively breeding species, Florida scrub-jay helpers are subordinate to the samesex breeder on the territory. Nevertheless, I found that helpers and breeders have statistically equivalent levels of corticosterone throughout the breeding season. These findings agree with Wingfield and his colleagues’ data from white-browed sparrow weavers, the only other cooperatively breeding bird species in which basal levels of corticosterone have been assessed. Although Nora Mays and her colleagues at the University of Arizona measured corticosterone and found no differences between breeders and helpers in the cooperatively breeding Harris’ hawk, the difficulty of trapping this raptor made measuring basal levels impossible.
I would argue that there are no fundamental physiological differences between breeders’ and helpers’ reproductive axes that provide a mechanism that explains the lack of reproduction in helpers. Even though younger jays differ to a small degree in their reproductive axes, ontogeny by itself does not explain delayed reproduction. Similarly, despite differences in body mass and body condition, these are as likely an effect of a helper’s reproductive quiescence rather than its cause. Florida scrub-jay helpers are merely breeders in waiting. Given the opportunity, they are fully capable of breeding, but given the constraints under which they live, they are forced to bide their time.
Parenting, Helping, and Prolactin
Although I have not found a clear-cut physiological mechanism that causes delayed breeding in Florida scrub-jays, a hormonal mechanism could still be responsible for scrub-jays performing parental and helping behavior. In many animals, the pituitary hormone prolactin has been associated with parental behavior, which made it a natural candidate for examination in scrub-jays. From scrub-jays in my study population, I found the following seasonal profile of prolactin secretion: low levels early in the breeding season and the highest levels when the birds are caring for eggs and nestlings. In addition, females have higher levels of prolactin than males, and breeders have higher levels than helpers.
Before trying to interpret those data, I shall explain a fundamental truism about the relation between hormones and behavior. When asked whether a hormone’s presence causes a behavior or the behavior causes the hormone’s presence, I often answer somewhat flippantly, “Yes!” In other words, a hormone may increase the likelihood that a behavior will occur, and engaging in the behavior often induces increased secretion of the same hormone.
How does this relate to the prolactin profiles of Florida scrub-jays? As is true of other species for which there are prolactin data, the member of the breeding pair that provides the greater portion of care of the eggs and young invariably has higher levels of prolactin. This is thought to be partially attributable to the greater levels of exposure to the stimuli of the nest, eggs and young that a primary caregiver experiences. Given that scrub-jay female breeders are the sole incubators of eggs and brooders of nestlings, it is not surprising that they have the highest prolactin levels. Similarly, helpers probably have lower levels of prolactin because they are excluded from the nest area by the breeders until after the young have hatched, thereby eliminating their exposure to the potential stimuli of a nest and eggs.
Perhaps more interesting, the seasonal profile of prolactin secretion provides temporal evidence for a link between the hormone and parental and helping behaviors. To better examine this relationship, during 1993 and 1994 I watched 27 nests to quantify the number of nest visits and how much food each group member contributed to nestling care. Not surprisingly, given their greater degree of investment in the nestlings, breeders contributed more than helpers. There were, however, no differences by sex.
To correlate an individual’s parental or helping behavior with its prolactin level, I captured as many of the animals as possible within one day of the nest watch and measured their levels of the hormone. Then, I compared that level with a feeding score, a quantification of how much a bird fed the nestlings. Scrub-jays carry food to the nest in their bill, throat sac or both. The amount of food an individual brought to the nest was scored numerically based on the degree of throat-sac distension as follows: a 1 if there was no visible distention but the bird did transfer food to at least one nestling, a 2 for a noticeable bulge in the throat sac and a 3 if the throat sac was extremely full. When my colleagues and I could not clearly see the throat sac but the individual fed, it was scored as a 1 to err on the side of caution. The number of nest visits and feeding scores were totaled and averaged over the two nest watches and then expressed as feeding score per hour.
The results show a positive correlation between feeding score and prolactin levels when all jays are considered together. Moreover, helpers show the strongest positive correlation between feeding score and prolactin. A closer look at those data reveal that a number of helpers did not feed nestlings, thereby receiving feeding scores of zero. Comparing the prolactin levels of these “nonhelpers” with those of helpers that fed nestlings shows that the latter birds-the helping helpershad significantly higher levels of prolactin. Such correlations between helping and prolactin are exciting and provide some support for the hypothesis that helping behavior is mediated by prolactin.
Paradoxically, when the relation between prolactin and the amount of food delivered by breeders is considered, no correlation was found. Although disconcerting, this is not surprising, because I did not examine all of the parental behaviors that prolactin has been shown to promote in other animals. For example, prolactin has been implicated in distraction displays, defense of young and maternal aggression in birds and mammals. Given the evidence that prolactin may mediate a diverse array of parental behaviors, I would like to study a broader range of behavioral measures in the future.
In fact, many more experiments remain to confirm prolactin’s role in helping. Historically, attributing a function to a hormone requires three steps. First, a correlation between the hormone and a behavior is noted. Second, the hormone is removed-often by excision of the source tissue or through more modern pharmacological methods that obviate the need for surgical excision-and then investigators check to see if the potentially hormone-mediated behavioral or physiological response has also been removed. Third, replacement of the hormone should reinstate the behavioral or physiological response. Additionally, if a hormone is thought to cause a behavior, increasing levels of the hormone should also increase the behavior. Although most of these experiments remain to be completed on scrub-jays, my colleagues and I are excited about the potential that this research has in illuminating one of the mechanisms underlying cooperative behavior in Florida scrub-jays.