James T. Costa
THE OTHER INSECT SOCIETIES
767pp. Harvard University Press. £38.95.
978 0 674 02163 1
Earwigs make good mothers; but they are not the insects that spring immediately to mind when the phrases “social insects”, “insect sociality” or “parental care” are used. Our thoughts turn instead to the complex, caste-driven societies of the ants, bees, wasps and termites. These are the famous socialites with the good PR, the celebrities of the insect social world, the names in the Hello! magazines of entomology. In this extraordinarily thorough book, James T. Costa sets the record straight and rebalances our view of sociality in insects by dealing with the neglected also-rans. Thirty-five years ago, E. O. Wilson carefully segregated what he termed the eusocial insects, the celebrities, from the many, many other groups that exhibit what Wilson describes as more generalized sociality. Costa draws the same line.
Eusocial insects are defined as forming groups in which members use the same composite nest, exhibit co-operative brood care and overlap of generations, and in which reproductive castes have evolved. Once the eusocial insects are discarded, definitions of what does and does not constitute sociality have been blurred historically. Teleological bunfights have been rife; in his introduction to Costa’s book, Wilson notes archly that in 1918 P. Deegener coined ninety-two “tongue-twisters” to describe the ninety-two different forms of animal sociality that he recognized; this classification is, mercifully, defunct.
The first two chapters of The Other Insect Societies are hard-going but thorough; first, Costa reviews the possible concepts for sociality, their nomenclature and the underlying mechanisms proposed to explain the phenomenon. Elements such as advantage to the colony, the clan or the species have been loosely invoked since Darwin’s day; but the theory of an overwhelming genetic basis for sociality has been underpinned since the 1960s by W. D. Hamilton’s paradigm of kin selection, which separates natural selection into two elements, direct and indirect. The direct applies to the individual and the indirect to the reproductive success of relatives. Reproductive altruism occurs where direct selection is nil and indirect selection greater than nil. It works at the genic level and leads indirectly to the selfish gene hypothesis. J. B. S. Haldane summarized the process succinctly: “I would lay down my life for two brothers or eight cousins”.
But it is now recognized that kin selection may be an inadequate explanation at some levels of social evolution – Wilson and Bert Hölldobler have recently suggested in a paper that kin selection may be unimportant relative to colony-level selection in the early stages of the evolution of eusocial insects. While an element of what we might anthropomorphically term altruism is very often apparent to the group, the group may not be closely related to the carer or carers. The current view, endorsed strongly by Costa, and supported by some recent models, is that there is a much greater interplay between genetic and ecological factors in social evolution than was previously thought. Relatedness within social groups as proof of kin selection may neither be as watertight nor as important as first thought. Indeed, Hamilton was “explicit in noting that ecological elements set the cost and benefit bar”.
Costa lays out his stall with his synthesis of “the ecology of social evolution”. His division of the types of sociality dealt with here into just four categories is refreshingly simple to follow. First come those species that exhibit exclusive maternal and biparental care behaviours. As Costa acknowledges, there is a striking correlation between these behaviours and the pattern of reproduction that D. W. Tallamy and W. P. Brown have termed “semelparous”. Semelparity involves reproductive effort being concentrated in a once-in-a-lifetime event, as distinct from iteroparity, in which reproductive effort is spread throughout the adult lifespan. Usually the semelparous care investment is in just a single clutch; a once-only batch of eggs compels care or defence. Parental care is here more likely to evolve, since the limited possibilities of future reproduction lower the cost of expending energy caring for eggs or young. Costa suggests that semelparity has evolved in response to selective pressures such as predation, or challenges of resource distribution (often food) or other environmental factors. Males are involved in care where predation or parasitism risks are greatest and what Costa terms “paternity assurance” may play a part.
Category two is rare and comprises those species that exhibit exclusive paternal care for the brood. Its driving factors are quite different from those giving rise to maternal or biparental care. All arthropod groups known to exhibit exclusively paternal care are iteroparous, and Costa postulates that male care releases females from reproductive constraints that would otherwise be imposed by necessary care behaviours. Caring fathers make good mates and paternal care evolves by sexual selection, a different evolutionary dynamic from maternal or biparental care. The phenomenon may not, however, necessarily involve monogamy nor demand that the brood cared for be the offspring of the caring male. In many examples, the creche minder mates with multiple females and each case needs to be examined and explained separately. “These males may thus enjoy a promiscuous lifestyle even as they provide exclusive parental care . . . . Tending clutches may become a badge of honour sought by females, in which case it may actually behove a male to acquire a clutch or brood to care for even if it is not his own!”
The third group is the “fortress defenders”, species that show no maternal care, but in which some colony members have evolved defender behaviour to protect an often ephemeral and vital food resource such as a plant gall. Fourthly and finally comes the herd. Insect herds come in two forms, the larval herd and the mixed-family herd, the latter involving different stages and ages. Herds gain advantage from communal living by pooling defensive abilities and the wherewithal to find and utilize food. In mixed-family herds, defence may often be the role of adults.
Costa sees the underlying drives to these social behaviours as primarily ecological:
both biotic and abiotic: for example, predation, parasitism, kleptoparasitism, resource needs and thermal challenges. Naturally, kin selection may play a role in some cases, but clearly not in all. The widespread occurrence of communal oviposition, merging of unrelated groups, and the absence of kin discrimination in many groups suggests that maintenance of strict family structure is unimportant relative to the group-derived benefits.
Preliminaries disposed of, we enter clear blue waters and embark on a remarkable voyage through the natural history of social insects. Costa arranges this part of the work taxonomically, and some 650 pages of the book are devoted to this comprehensive survey. Eighteen chapters deal each with one or more insect orders, though the richness of social life in the true bugs (Hemiptera), beetles (Coleoptera) and moths and butterflies (Lepidoptera) necessitates several chapters for each of these.
We begin with earwigs as good mothers. Costa opens this chapter (as he does the others) with an informal overview of the group, a little history, and key references. Descriptive and interpretive natural history, ecology and taxonomy is a wonderfully rich field of science and involves working from a vast literature spanning a timeline much longer than those of many other biological disciplines and written in many more languages. Thus, with the earwigs (Dermaptera), Costa is able to start with the master of eighteenth-century biological observation, the Frenchman Charles de Geer, writing in 1773 of finding: “a female earwig with several small insects, which were quite obviously her progeny. They did not leave her, and even placed themselves under her body as chickens under a hen. So insects of this kind take care, in a way, of their offspring”. Subsequent work has shown that, of the earwig species whose life histories are known, maternal care is practically universal, the female building a nest, grooming the eggs, feeding the young and protecting them. Grooming of eggs serves to remove microbial pathogens. The inevitable experiments have shown that eggs deprived of maternal care have a very low survival rate – most are killed by fungi or by predatory soil mites. Females of the common European earwig Forficula auricularia may produce numerous clutches, sometimes from a single initial mating, as they are able to store sperm for long periods. These earwigs appear to buck the trend of the semelparity hypothesis – and male earwigs appear to play no part in care, their role being that of a mere mating machine. Males of two genera are curiously well adapted to this latter role for, remarkably, they have two penises, one kept as a spare.
Maternal care is even suspected to occur among the small number of earwig species that live in close association with bats in South-East Asia and as ectoparasites on rats in Africa. The aptly named hairy earwig Arixenia esau lives on cave bats in Malaysia, grazing on skin flakes and what are euphemistically termed “glandular secretions”. Females give birth to living first instar young rather than eggs, and there is a moving observation of a mother earwig cleaning and helping on its way a sticky newborn on the back of a rat.
Females of the Alpine earwig Anechura bipunctata are semelparous, and it is reported that females can and will pool their eggs and jointly provision the brood. They provide the longest span of maternal care found in any earwig, and stay with the nymphs for several instars and even to maturity. In the related Japanese species Anechura harmandi, however, maternal care is elevated to a whole new level of sacrifice, and having assisted and fed her nymphs through the first few days of life, the mother allows herself to be eaten by her brood, a process noted by Costa to be “the ultimate Head Start program”. He notes that this is by no means as unconventional as the reproductive behaviour of the cecidomyiid fly Miastor:
whose young larvae hatch and consume their mother from within, before they are even “born”. What makes this even more bizarre, however, is the fact that the mother itself is a larva. These flies are paedogenetic, “young-reproducing”, where several generations of larvae are telescoped into one, the larvae of which reproduce without maturing and each of which consumes its larva – mother from within.
This “Russian doll ate my mother” scenario is perhaps more suitable for the National Enquirer than Hello!.
Costa concludes his masterly review of the earwigs with a look at the work that has been done on how reproduction in earwigs is controlled hormonally and how plasticity in reproduction, notably between high- and low-elevation populations of Forficula auricularia, correlates not only with environmental factors and hormonal regimes but also seems to correlate with local genetic variation. Low-elevation earwigs usually produce successive broods, whereas high-elevation females adopt a one-shot strategy, producing a single annual brood. Transplanted to the laboratory and a level playing field, this feature becomes a trend rather than an absolute among individuals from each population. Finally, Costa speculates that there may be much more to earwig social biology than we know, as the life histories of just twenty of 1,500 species have been studied. “In cases where F. auricularia nymphs are present when a second batch is laid, egg cannibalism appears to ensue. . . . but out there somewhere may be a species whose nymphs stick around and help groom the eggs, and perhaps even forage [for food]”.
This wonderful natural history proceeds in much the same vein and with the same authority through the insect orders. The eastern lubber grasshopper, Romalea microptera, has conspicuous warning coloration and the nymphs forage in well-defined groups. Certain grasshoppers and locusts may group together and bask in a single mass in order to absorb and conserve body heat from the sun. But the function of Romalea groups is defensive. These grasshoppers sport a fearsome arsenal of offensive chemical compounds that are launched at predators as a spray from the thoracic tracheal glands and exuded over the grasshopper’s body as an “unappetising foam”. More than fifty different defensive chemicals are employed, some sequestered from foodplants and others synthesized de novo. Thomas Eisner discovered, furthermore, that in Florida Romalea ingested 2–4 dichloro-phenoxyacetic acid, widely used locally as a pesticide and, apparently unmoved by the experience, converted it into the eye-watering compound 2,5-dichlorophenol and proceeded to use this on its adversaries who were suitably impressed by this delightful insouciance.
African Zonocerus grasshoppers utilize pyrrolizidine alkaloids (PAs) sequestered from plants such as Asclepias and Heliotropium as defensive chemicals and, like Romalea, the nymphs form up into groups for mutual protection. Danaid (milkweed) butterflies use the same compounds as sex pheromones and in defence, and feed opportunistically as adults on any sources that can be ingested through the proboscis. PAs act as virulent liver poisons in vertebrates. It was the frantic feeding of danaid butterflies on Zonocerus road kills that led Miriam Rothschild and co-workers to first suspect and investigate these grasshoppers as PA-users. Costa tells us that Zonocerus are “eaten as a relish by the Wa-Sambaa people of Tanzania (fried and tossed with plenty of salt)”, but they are apparently an acquired taste – a little like fugu.
Aphids are the new hot property in studies of social insects. Widespread caste differentiation among gall-making species from two families is a comparatively recent discovery, and there are numerous variations on the theme. A generalized scenario involves a foundress mother aphid inducing, by a combination of mechanical damage and possibly the injection of growth-inducing chemicals, a hollow gall in a leaf in spring. The foundress mother may have been the product of sexual reproduction on an alternative hostplant in autumn – hostplant alternation is common in many aphids. She then populates the gall by giving birth to first instar nymphs (rather than laying eggs); these are all daughters produced parthenogenetically – ie, each is a clone of the mother. While some daughters are conventional aphids like their mother and mature to resemble her precisely, others more resemble small scorpions with muscled raptorial forelegs and forward-directed dagger-like stylets – the stabbing mouthparts that in the familiar greenfly are plugged into a leaf or stem to suck sap. These agile, aggressive soldiers never moult and mature. Their sole function is to defend the colony. In one Taiwanese species, the soldiers mount suicide attacks against mammals, including man, and this appears to be primarily a defence against squirrels eating the Styrax gall that is home to the colony. If we consider the cloning aphid as a superorganism, the foundress might be considered to have split herself into an array of feeders and breeders (for the daughters repeat the cloning process) and mailed fists. The numerous parthenogenetic generations that may ensue through the summer, coupled with the remarkable fecundity of aphids, place some of these aphids potentially among the biggest superorganisms in existence. But it is often not as simple as this. There is evidence in many species of numerous foundresses being found in one gall; there is cannibalism, parasitism, competition and a host of other dirty tricks as well as co-operation between different clones. Teasing apart the genetic and ecological advantages, and determining the relatedness (or not, in view of the dirty tricks department) of aphid “families” has become a compelling and very active research area.
There are hundreds more riveting examples of insect sociality in this tour de force of natural history. Costa has meticulously researched this book and drawn on thousands of references. And still, he claims, it is by no means comprehensive. Thrips defend fortresses; beetles roll dung balls; carrion beetles nest communally under dead bodies; male giant water-bugs play at proud fathers with their egg family glued to their backs. Australian sawfly larvae – “spitfires” – spew eucalyptus oil in one’s eye; processionary caterpillars march, following a sophisticated pheromone trail; tent caterpillars encamp. I may have missed a reference to one of my favourite social behaviours among caterpillars, the co-operative tree-shaking ritual by head-banger larvae of Arcte coerula driving off vertebrate intruders. The absence of even a passing reference to under-observed and under-recorded communal roosting (as distinct from aestivation) in adults of certain moth species is a pity, but references and details are sparse and mostly inhabit the realm of “personal communication”.
James Costa’s eloquent coda to this work, titled “Sociality in an Appalachian Spring”, teaches that insect sociality is all around us, ever-changing with season and habitat. The Other Insect Societies provides an encyclopedic and data-rich overview of that sociality, beautifully written with a love for the subject and with humour. It is a remarkable and eye-opening collation, a ground-breaking and first-class reference work of science and natural history.
_________________________________________________________
Gaden S. Robinson researches small tropical moths at The Natural History Museum, London. His recent books include Hostplants of the Moth and Butterfly Caterpillars of America North of Mexico, 2002, and Hostplants of the Moth and Butterfly Caterpillars of the Oriental Region, 2001.
