By what perverse logic should masked boobies produce two eggs, yet never rear more than one chick invariably branded with the mark of Cain? Nelson argues forcefully that clutches of two eggs represent an adaptation for greatly increased success in raising one chick. The causes of death in eggs and young hatchlings are numerous—siblings intent upon murder being only one of many dangers to which booby flesh is heir. Eggs crack or roll from the nest; tiny hatchlings easily overheat. The second egg may represent insurance against death of the first chick. A healthy first chick cancels the policy directly, but the added investment may benefit parents as a hedge worth the expense of producing another egg (they will, after all, never need to expend much energy in feeding an unnecessary second chick). At Kure Atoll in the Hawaiian Archipelago, for example, clutches of two eggs successfully fledged one chick in 68 percent of nests examined during three years. But clutches of one egg fledged their single chick only 32 percent of the time.
Evolutionary biologists, by long training and ingrained habit, tend to discuss such phenomena as the siblicide of boobies in the language of adaptation: how does a behavior that seems, at first sight, harmful and irrational really represent an adaptation finely honed by natural selection for the benefit of struggling individuals? Indeed, I have (and somewhat uncharacteristically for me) used the conventional language in this essay, for Nelson’s work persuades me that siblicide is a Darwinian adaptation for maximizing the success of parents in rearing the largest number of chicks permitted by prevailing abundances of food.
But I am most uncomfortable in attributing the basic behavioral style, which permits siblicide as a specific manifestation, only to adaptation, although this too is usually done. I speak here of the basic mode of intelligence that permits siblicide to work: the sailor’s system (of my opening paragraph) based on yes-no decisions triggered by definite signals. John Alcock, for example, in a leading contemporary text (Animal Behavior: An Evolutionary Approach, 1975) argues over and over again that this common intellectual style is, in itself and in general, an adaptation directly fashioned by natural selection for optimal responses in prevailing environments: “Programmed responses are widespread,” he writes, “because animals that base their behavior on relatively simple signals provided by important objects in their environment are likely to do the biologically proper thing.”
(On the overwhelming power of natural selection, no less a personage than H.R.H. Prince Philip, duke of Edinburgh, has written in the preface to Nelson’s popular book on birds of the Galápagos: “The process of natural selection has controlled the very minutest detail of every feature of the whole individual and the group to which it belongs.” I do not cite this passage facetiously to win an argument by saddling a position I do not accept with a mock seal of royal approval, but rather to indicate how widely the language of strict adaptation has moved beyond professional circles into the writing of well-informed amateurs.)
As I argued for siblicide and guano rings, I am prepared to view any specific manifestation of my sailor’s intellectual style as an adaptation. But I cannot, as Alcock claims, view the style itself as no more than the optimized product of unconstrained natural selection. The smaller brain and more limited neural circuitry of nonhuman animals must impose, or at least encourage, intellectual modes different from our own. These smaller brains need not be viewed as direct adaptations to any prevailing condition. They represent, rather, inherited structural constraints that limit the range of specific adaptations fashioned within their orbit. The sailor’s style is a constraint that permits boobies to reduce their broods by exploiting a behavioral repertoire based on inflexible rules and simple triggers. Such a system would not work in humans, for parents will not cease to recognize their babies after a small and simple change in location. In human societies that practice infanticide (for ecological reasons often quite similar to those inducing siblicide in boobies), explicit social rules or venerated religious traditions—rather than mere duplicity by removal—must force or persuade parental action.
Birds may have originally developed their brain, with its characteristic size, as an adaptation to life in an ancestral lineage more than 200 million years ago; the sailor’s style of intelligence may be a nonadaptive consequence of this inherited design. Yet this style has set the boundaries of behavior ever since. Each individual behavior may be a lovely adaptation, but it must be fashioned within a prevailing constraint. Which is more important: the beauty of the adaptation or the constraint that limits it to a permissible path? We cannot and need not choose, for both factors define an essential tension that regulates all evolution.
The sources of organic form and behavior are manifold and include at least three primary categories. We have just discussed two: immediate adaptations fashioned by natural selection (exploitation by older booby siblings of their parents’ intellectual style, leading to easy dispatch of nest mates); and potentially nonadaptive consequences of basic structural designs acting as constraints upon the pathways of adaptation (the intellectual style of yes-no decisions based on simple triggers).
In a third category, we find definite ancestral adaptations now used by descendants in different ways. Nelson has shown, for example, that boobies reinforce the pair-bond between male and female through a complex series of highly ritualized behaviors that include gathering objects and presenting them to mates. In boobies that lay their eggs upon the ground, these behaviors are clearly relics of actions that once served to gather material for ancestral nests—for some of the detailed motions that still build nests in related species are followed, while others have been lost. The egg-laying areas of masked boobies are strewn with appropriate bits of twigs and other nesting materials that adults gather for their mutual displays and then must sweep out of the guano ring to lie unused upon the ground. I have emphasized these curious changes in function in several other essays (see 4 and 11) for they are the primary proof of evolution—forms and actions that only make sense in the light of a previous, inherited history.
When I wonder how three such disparate sources can lead to the harmonious structures that organisms embody, I temper my amazement by remembering the history of languages. Consider the amalgam that English represents—vestiges, borrowings, fusions. Yet poets continue to create things of beauty. Historical pathways and current uses are different aspects of a common subject. The pathways are intricate beyond all imagining, but only the hearty travelers remain with us.
4 | Quick Lives and Quirky Changes
POSTHUMOUS TRIUMPH IS HOLLOW, however abstractly rewarding. Nanki-Poo refused Ko-Ko’s inducement to undergo a ceremonious public beheading rather than a private suicide: “There’ll be a procession—bands—dead march—bells tolling…then, when it’s all over, general rejoicings, and a display of fireworks in the evening. You won’t see them, but they’ll be there all the same.” And I never could figure out why America’s premier nineteenth-century anthropologists J. W. Powell and WJ McGee made a bet about who had the larger brain—to be settled by autopsy when the joy of victory could no longer be savored.
Nonetheless, I just made a dumb bet with a female jogging enthusiast: that no woman would win the Boston marathon in my lifetime. I’d rather lose, but expect I won’t. Still, if superior average speed of running males is among the few insignificant but genuinely biological differences between human sexes, I can only respond to charges of gloating (for the abstraction I represent, but not, alas, for me and my huffing eight-minute miles) with a statement of genuine regret; how gladly would I trade this useless advantage for the most precious benefit of being female—several extra years of average life.
I do not know whether shorter male life is a generality in nature—and whether we should therefore add to smaller average size (see essay I) another biological strike against machismo—but I just learned (with thanks to Martin L. Adamson) about an instructive extreme case.
In 1962, James H. Oliver Jr. traced the life cycle of a mite that parasitizes the cocoons of earthworms. Bo
th males and females of Histiostoma murchiei pass through an egg and three juvenile stages before molting into an adult. In addition, the female intercalates one additional stage—euphoniously named the hypopus—between the second and third pre-adult phases. Females develop at a leisurely pace for such a small creature. Discounting the hypopus, the passage from egg to adult, through stages held in common with males, takes one to three weeks. The additional hypopus may extend female life greatly—for these mites find and infest other cocoons only during the hypopal stage (males always stay at home). The hypopus may, first of all, remain dormant for long periods within the skin of the previous juvenile stage, awaiting (so to speak) favorable conditions for emergence and movement to another cocoon. When the hypopus does emerge, it may then live for a long time, moving about in its own cocoon (and sometimes becoming dormant again) or moving out in search of a new home.
Males, by contrast, race through the same stages (minus the hypopus) with a celerity that should inspire Bill Rodgers as he trudges up Heartbreak Hill next Patriot’s Day. “Adult males,” Oliver writes, “have been observed copulating with their mother within 3 to 4 days after being laid as eggs”—and they die soon after this bout of incestuous joy. Why this outstanding difference in life-span between the sexes? And what has it to do with the Oedipal habits of these mites? A further look at the unusual reproductive biology of these parasites seems to provide the answer.
When a hypopus finds a new cocoon, it lays two to nine eggs within two days after molting into an adult—and without benefit of fertilization. All these eggs develop into males—the only source of potential husbands as well. What better evolutionary rationale for rapid male development could we hope to find? The females of most species must seek their husbands. These mites make them from scratch and then wait. Males of Histiostoma murchiei are little more than sources of sperm; the sooner they can perform, the better.
Two days after its incestuous mating, the female begins to lay eggs again and may continue for two to five days, producing as many as 500 offspring—all female this time.
In solving one problem—the differential speed of development between sexes—we have only encountered a more curious question: how can this system work in the first place; how can an unmated female, alone in a new cocoon, produce a generation of husbands, and why are the offspring of her next reproductive bout all female?
The answer to this broader question lies in the unfamiliar style of sex determination in these mites. In most animals, both males and females have paired chromosomes, and the status of one pair determines the sex of its bearer. Human females, for example, have two large sex chromosomes (designated XX), while males have one large (X) and one small (Y) chromosome in their determining pair. All unfertilized egg cells carry a single X, while sperm carry either an X or a Y. We each owe our sex to the good fortune of one sperm among the millions per ejaculate. Animals with paired chromosomes in both sexes are called diploid.
Some animals use a different system of sex determination. Females are diploid, but males have only one chromosome for each female pair and are called haploid (for half the diploid number). Males, in other words—and ironic as this may seem—develop from unfertilized eggs and have no fathers. Fertilized eggs produce diploid females. Animals using this system are called haplodiploid (because males are haploid and females diploid).
Histiostoma murchiei is haplodiploid. Hence, the unmated female in a new cocoon raises a generation of males from unfertilized eggs, and a subsequent generation of females from the resulting incest.
Haplodiploidy, a fascinating phenomenon rich in implication, has circulated through these essays in various contexts for years. It helped to explain the origin of social systems in ants and bees (see essay 33 in Ever Since Darwin), and it underlay the habits of a male mite who fertilizes several sisters within his mother’s body, and dies before “birth” (essay 6 in The Panda’s Thumb). It also circulates widely through the animal kingdom. Haplodiploid species have been found in rotifers, nematodes, mites, and in four separate orders of insects—the Thysanoptera (thrips), the Homoptera (aphids, cicadas, and their allies), the Coleoptera (beetles), and the Hymenoptera (ants, bees, and wasps). These groups are not closely related and their presumed common ancestors are diploid. Thus, haplodiploidy has arisen independently—and often many times—within each group. Although most of these groups contain only a few haplodiploid species amidst a host of ordinary diploids, the Hymenoptera, with more than 100,000 named species, are exclusively haplodiploid. Since vertebrates only include some 50,000 species, as Oliver reminds us, our chauvinistic impression that haplodiploidy is curious or rare should also be revised. At least 10 percent of all named animal species are haplodiploid.
Within the last decade, haplodiploidy has figured most prominently in the news (both general and scientific) for its role in an ingenious Darwinian explanation of an old biological mystery—the origin of sociality in Hymenoptera, particularly the existence of sterile “worker” castes, invariably female, in ants and bees. Since sociality evolved several times within the Hymenoptera, the invariant system of sterile female castes demands a general explanation. The larger problem is even more puzzling: why, in a presumably Darwinian world filled with organisms acting only for their personal reproductive success, should large numbers of females “forego” their own reproduction to help their mother (the queen) raise more sisters?
The ingenious explanation relies upon the peculiar asymmetries of genetic relationship between sexes in haplodiploid animals. In both diploids and haplodiploids, mothers pass half their genetic material (one set of chromosomes in each egg cell) to each offspring. They are therefore equally related (by half of their genetic selves) to both sons and daughters. A female in diploid species also shares approximately half her genes with both brothers and sisters. But a female in haplodiploid species shares three-quarters of her genes with sisters and only one-quarter with brothers for the following reason: Consider any gene (one copy on a single chromosome) in sisters. What is the probability that a brother will share it? If the gene is on a paternal chromosome, then the brother has zero probability of sharing it, for he has no paternal chromosomes. If the gene is on a maternal chromosome, then he has a 50 percent chance of sharing it with his sister—because he either received the same chromosome from his mother, or the other member of the pair. Thus, summing over all genes, the relationship between brother and sister is the average between zero (for paternal genes of sisters, necessarily absent in brothers) and 50 percent (for maternal genes)—or 25 percent.
What then is the probability that a sister will share the same gene? If it is a paternal gene, the sister must share it since fathers have only one set of chromosomes and they pass their entire genetic program to each daughter. If it is a maternal gene, the chance is 50 percent by the same argument advanced for brothers. The total relationship between sisters is therefore the average between 100 percent (for paternal genes) and 50 percent (for maternal genes)—or 75 percent.
Females are therefore more closely related to their sisters (by three-quarters) than either to their mothers (by one-half) or to their own potential offspring (also by one-half). If the Darwinian imperative leads organisms to maximize the numbers of their own genes in future generations, then females will do better by helping their mother raise sisters (as sterile workers do) than by producing their own offspring. Thus, the asymmetry of genetic relationship in haplodiploids may explain both why worker castes of social Hymenoptera are invariably female, and why sociality in this style has evolved many times among the Hymenoptera, but not among the much larger array of diploid organisms. (As always, our complex world provides an exception—the diploid termites, relatives of cockroaches, who at least include both males and females in their worker castes.)
This explanation of an old mystery has so intrigued biologists that a subtle reversal of causality has crept into some accounts. The very existence of haplodiploidy is linked with force and elegance to the evolution of sociality, an
d we are almost led to believe that this mode of sex determination arose “for,” or at least in the context of, the marvelous social organization of ants and bees. Yet a moment’s explicit reflection assures us that this cannot be so, for two reasons.
First, all hymenopterans are haplodiploid, but only a few lineages within the group have developed complex social systems (most hymenopterans are asocial or minimally social wasps). The common ancestor of living hymenopterans must have been haplodiploid, but it was certainly not fully social, since the complex society of highly derived bees and ants has evolved as a phyletic afterthought in several independent lineages. Causality must run in the other direction. Haplodiploidy does not exist “for” sociality unless the future can control the past. Rather, haplodiploidy arose for other reasons and then permitted, by good and unplanned fortune, the later evolution of this wonderfully complex and successful mode of sociality. But what other reasons?—which brings me, finally, to the point of this essay, to the main reason for my fascination with Histiostoma murchiei, and, more immediately, to the second item.
Second, when we consider the usual ecological context of haplodiploidy in a broad range of animals that may have evolved it directly (and not merely co-opted it for another use), an interesting pattern emerges. Histiostoma murchiei shares a mode of life with the mites that die before birth, and with many other haplodiploid animals in distantly related groups: all are “colonizers,” species that survive by seeking rare but rich resources and then reproducing as fast as they can when uncommon fortune rewards their search (the vast majority of Histiostoma’s hypopi die before finding a fresh earthworm cocoon). Haplodiploidy provides several advantages in this chancy approach to survival. Successful colonization does not require two separate migrations of a male and a female, or even that a single migrating female be fertilized before her search for a new resource begins. Any unmated female, even a juvenile, becomes a potential source of new colonies, since she can make a generation of males all by herself and then mate with them to begin a generation of females—the strategy evolved by Histiostoma.