SOCIAL LIFE AMONG THE INSECTS'

By Professor WILLIAM MORTON WHEELER
BUSSEY INSTITUTION, HARVARD UNIVERSITY

LECTURE IV-ANTS, THEIR DEVELOPMENT, CASTES, NESTING AND FEEDING HABITS

II

There is throughout the animal kingdom, as I believe Espinas was the first to remark, a clear correlation on the one hand between a solitary life and carnivorous habits and on the other hand between social habits and a vegetable diet. The beasts and birds of prey, the serpents, sharks, spiders and the legions of predacious insects all lead solitary lives, whereas the herbivores, rodents, granivorous and frugivorous birds and plant-eating snails and insects are more or less gregarious. Man himself is quite unable to develop populous societies without becoming increasingly vegetarian. Compare, for example, the sparse communities of the carnivorous Esquimaux with the teeming populations of the purely vegetarian Hindoos. The reasons for these correlations are obvious, for plants furnish the only abundant and easily obtainable foods and, at least in the form of seeds and wood, the only foods that are sufficiently stable to permit of long storage. In the previous lectures I have shown that the social beetles and bees are strictly vegetarian and that the social wasps, though descended from highly predatory ancestors, are nevertheless becoming increasingly vegetarian like the bees. The ants exhibit in the most striking manner the struggle between a very conservative tendency to retain the precarious insectivorous habits of their vespine ancestors and a progressive tendency to resort more and more to a purely vegetable regimen as the only means of developing and maintaining populous and efficient colonies. Anthropologists have distinguished in the historical development of human societies six successive stages, designated as the hunting, pastoral, agricultural, commercial, industrial and intellectual. Evidently the first three, the hunting, pastoral and agricultural, are determined by the nature of the food and represent an advance from a primitive, mainly flesh-consuming to a largely vegetarian regimen. Lubbock showed that the same three stages occur in the same sequence in the phylogenetic history of the ants. At the present time we are able to give even greater precision to his outlines of this evolution.

1 Lowell Lectures.

All the primitive ants are decidedly carnivorous, that is predatory hunters of other insects. That this must have been the character of the whole family during a very long period of its history is indicated by the retention of the insectivorous habit, in a more or less mitigated form, even in many of the higher ants. Always striving to rear as many young as possible, always hungry and exploring, the ants early adapted themselves to every part of their environment. They came, in fact, to acquire two environments, each peopled by a sufficient number of insects, arachnids, myriopods, etc., to furnish a precarious food-supply. Most of the ants learned to forage on the exposed surface of the soil and vegetation and became what we call epigæic, or surface forms, while a smaller number took to hunting their prey beneath the surface of the soil and thus became hypogæic, or subterranean. Many of the latter are very primitive but their number has been repeatedly recruited from higher genera, which by carrying on all their activities within the soil have found a refuge and surcease from a too strenuous competition with the epigaic species. We have here some very interesting cases of convergence, or parallel development, since the underground habit has caused the workers, which rarely or never leave their burrows, to lose their deep pigmentation and become yellow or light brown and to become nearly or quite blind. As will be evident in the course of my discussion, the tendency towards vegetarianism is apparent among both the epigæic and hypogaic forms.

The ants belonging to the oldest and most primitive subfamilies, the Ponerinæ, Doryline and Cerepachyinæ and also to many of the lower genera of Myrmicina, feed exclusively on insects and therefore represent the hunting stage of human society. Owing to the difficulty of securing large quantities of the kind of food to which they are addicted, many of the species form small, depauperate colonies, consisting of a limited number of monomorphic workers. Many of these species lead a timid, subterranean life. In the size of their colonies, which may comprise hundreds of thousands of individuals, the Doryline alone constitute a striking exception, but one which proves the rule. These insects, known as driver, army or legionary ants and very largely confined to Equatorial Africa and tropical America, are strictly carnivorous, but being nomadic and therefore foraging over an extensive territory, are able to obtain the amount of insect food necessary to the growth and maintenance of a huge and polymorphic population. They are the famous ants whose intrepid armies often overrun houses in the tropics, clear out all the vermine and compel the human inhabitants to leave the premises for a time. In Africa they have been known to kill even large domestic animals when they were tethered or penned up and thus prevented from escaping.

The pastoral stage is represented by a great number of Myrmicine and especially of Formicine and Dolichoderine ants which live very largely on "honey-dew." This sweet liquid, concerning the origin of which there was much speculation among the ancients, is now known to be the sap of plants and to become accessible. to the ants in two ways. First, it may be excreted by the plants from small glands or nectaries ("extrafloral nectaries") situated on their leaves or stems, where it is eagerly sought and imbibed by the ants. Second, a much more abundant supply is made accessible by a great group of insects, the Phytophthora, comprising the plant-lice, scale-insects, mealy-bugs, leaf-hoppers, psyllids, etc., which live gregariously on the surfaces of plants. These Phytophthora pierce the integument of the plants with their slender, pointed mouth-parts and imbibe their juices which consist of water containing in solution cane sugar, invert sugar, dextrin and a small amount of albuminous substance. In the alimentary canal of the insects much of the cane sugar is split up to form invert sugar and a relatively small amount of all the substances is assimilated, so that the excrement is not only abundant but contains more invert and less cane sugar. This excrement or honey-dew either falls upon the leaves and is licked up by the ants or is imbibed by them directly while it is leaving the bodies of the Phytophthora. Many species of ants have learned how to induce the Phytophthora to void the honey-dew by stroking them with the antennæ, protect and care for them and even to keep them in specially constructed shelters or barns. Some ants have acquired such vested interests in certain plant-lice that they actually collect their eggs in the fall, keep them in the nests over winter and in the spring distribute the hatching young over the surface of the plants. Linnæus was therefore justified in calling the plant-lice the dairycattle of the ants ("ha formicarum vacca"). This dairy business is, in fact, carried on in all parts of the world on such a scale and with so many species of Phytophthora that it constitutes one of the most harmful of the multifarious activities of ants. Their irrepressible habit of protecting and distributing plant-lice, scaleinsects, etc., is a source of considerable damage to many of our cultivated plants and especially to our fruit-trees, field and garden crops. Ants mostly attend Phytophthora on the leaves and shoots. of plants, but quite a number of species are hypogæic and devote themselves to pasturing their cattle on the roots. Thus our common garden ant (Lasius americanus) distributes plant-lice over the roots of Indian corn.

The habit of keeping Phytophthora was probably developed independently in many different genera, and it is easy to see how the habit of feeding by mutual regurgitation among the ants themselves might have led to the behavior I have been describing. Cer

Vol. XV.--34.

tainly the genera that have developed trophallaxis among the adult members of their colonies are the very ones which most assiduously attend the Phytophthora. And it is equally certain that the latter habit is very ancient, because it was already established among the ants of the Baltic Amber during Lower Oligocene times and that, as we have seen, was many million years ago.

The dairying habit has led to an interesting specialization in certain species known as "honey ants," which inhabit desert regions or those with long, dry summers. These ants have found it very advantageous to store the honey dew collected during periods of active plant growth, and as they are unable to make cells like those of wasps and bees, have hit upon the ingenious device of using the crops of certain workers or soldiers for the purpose. In all ants, as we have seen, the crop is a capacious sac, but in the typical honey ants it becomes capable of such extraordinary distention that the abdomen of the individuals that assume the rôle of animated demijohns or carboys, becomes enormously enlarged and perfectly spherical. Such "repletes" (Fig. 66) are quite unable to walk and therefore suspend themselves by their claws from the ceilings of the nest chambers. When hungry the ordinary workers stroke their heads and receive by regurgitation droplets of the honey dew with which they were filled during seasons of plenty. The condition here described, or one of less gastric distention, has been observed in desert or xerothermal ants in very widely separated regions and belonging to some nine different genera of Myrmicinæ, Formicinæ and Dolichoderinæ (Myrmecocystus and Prenolepis in the United States and Northern Mexico, Melophorus, Camponotus, Leptomyrmex and Oligomyrmex in Australia, Plagio

a, lateral

lepis and Aëromyrma in Africa and Pheidole in Australia and the southwestern United. States).

A more direct vegetarian adaptation is seen in many Formicidæ that inhabit the same desert or xerothermal regions as the honey ants. In such regions insect food is at no time abundant and is often so scarce that the ants are compelled to eat the seeds of the sparse herbaceous vegetation. At least a dozen genera, all Myrmicinæ, illustrate this adaptation: Pogonomyrmex, Veromessor, Novomessor and Solenopsis in America, Messor, Oxyopomyrmex, Goniomma, Tetramorium and Monomorium in the southern Palearctic region, Meranoplus in the Indoaustralian, Cratomyrmex and Ocymyrmex in the Ethiopian region and Pheidole (Fig. 57) in the warmer parts of both hemispheres. It was at one time believed that some of these ants actually sow around their nests the grasses and other herbaceous plants from which they gather the seeds, but this has been disproved. They are merely collected, husked and stored in special chambers or granaries in the more superficial and dryer parts of the formicary. Emery has shown that as food the proteids are preferred to the starchy portions of the seeds and are also fed to the larvæ. Messor barbarus, the ant to which Solomon refers, is one of these harvesters. Probably none of them disdains insect food when it can be had. Nevertheless the adaptation to crushing hard seeds is so pronounced in certain genera that the mandibles have become distinctly modified. Their blades have become broader and more convex and the head has been enlarged to accommodate the more powerful mandibular muscles. In certain forms (Pheidole, Messor, Novomessor, Holcomyrmex) the soldiers or major workers seem to function as the official seed-crushers of the colony.

The harvesting ants can hardly be regarded as true agriculturists because they neither sow nor cultivate the plants from which they obtain the seeds. Yet there is a group of ants which may properly be described as horticultural, namely the Attiini, a Myrmicine tribe comprising about 100 exclusively American species. and ranging from Long Island, N. Y., to Argentina, though well represented by species only within the tropics. The tribe includes several genera (Cyphomyrmex, Apterostigma, Sericomyrmex, Myrmicocrypta, etc.) the species of which are small and timid and form small colonies with monomorphic workers, while others (Atta and Acromyrmex) are large and aggressive and form very populous colonies with extremely polymorphic workers. The Attas or parasol ants inhabit the savannas and forests of South and Central America, Mexico, Cuba and Texas. Their extensive excavations. result in the formation of large mounds and often cover a considerable area (Fig. 67). According to Branner, a single mound of