The Biology of the Social Wasps (Hymenoptera, Vespidae)

Summary 1. After defining social wasps, an outline is given of their classification and possible evolution. The Vespidae probably arose in the Malayan region at about the beginning of the Tertiary period and spread from there all over the world. They are now highly developed in South America and it is suggested that they may have reached that country over the Behring Straits, probably in two waves, the first of an ancestral Polybiine, the second perhaps of Polistes only. 2. A biief account is given of the architecture of wasps' nests which is often specific to the genus. Nevertheless, there are examples of very similar wasps making very different nests and there does not seem any case for putting an overriding value on nest‐architecture in classification. While we know sufficiently well what we have to explain, only a very small beginning has been made in describing and analysing the behaviour which produces the elaborate constructions we find. 3. In the familiar Vespinae the concepts of queen and worker are well defined and the two castes are discontinuous. In genera such as Polistes the position is much more fluid, particularly in the tropical species in which it is not necessary to have a queen specialized for hibernation. Some time after nest‐production, however, a queen becomes recognizable, though often more by her behaviour than by her structure. In the Polybiini, queens and workers are sometimes well differentiated and sometimes not; but in most genera and species there is more than one queen in a nest. Other types of females not so well defined and of uncertain significance also occur. In temperate climates new colonies are founded by one or a few queens; in the tropics most colonies are founded by swarms of queens and workers. 4. The main differences between queens and workers seem to be determined in the larval stage, perhaps by some secretion administered by the adults, but clearly influenced by other factors as well. The full behavioural differences between the castes are often finally established by social interactions between the adults. 5. Males tend to be produced towards the end of the life of short‐lived colonies or towards the end of the reproductive cycles of longer‐lived colonies. There is need for more information about in‐ or out‐breeding patterns, particularly in relation to specific differences in the degree of sexual dimorphism. 6. Very few pheromones are recognized with certainty and only one, the queen‐substance of Vespa orientalis , has been identified chemically. However, some species produce a substance from the female sixth gastral sternite (Van der Vecht's organ) which at least in one genus is an ant‐repellent. Some genera of Polybiines have a somewhat similar gland on the fifth sternite which may be connected with caste‐differentiation. Some wasps ( Vespula ) produce a footstep pheromone concerned with the recognition of the nest entrance and there is less good evidence for the existence of significant substances in the saliva of workers and in their poison glands (alarm substance). 7. Trophallaxis, or the supply of salivary secretion by the larvae to adults more or less in exchange for the food provided from outside by the foragers, seems to be essential to the colonies of Vespa since the adults have no proteases. Trophallaxis also occurs in some Polistinae but its significance there is not known. 8. Social hierarchies of ‘peck orders’ are always established amongst the adults of Polistes and play a part in determining which of several potential queens becomes the acting one. They are also important in establishing the relations of the queen to the workers and of the latter amongst themselves. A similar hierarchy occurs in Belono‐gaster . Some sort of hierarchy also exists amongst adult Vespula and is probably important in relation to trophallaxis, but the hierarchies must become less definite in large colonies with hundreds of individuals, especially if there is more than one queen. 9. The principal hurdle in the evolution of social behaviour is the establishment of genes which determine that some females lay most or effectively all of the eggs while others nurse, build and forage. This evolution may only be possible amongst females which are so closely related that their genomes are almost identical. In this connexion we need much more direct field evidence on the dispersal of queens, on mating systems and the incidence of multiple insemination. It appears that, once true social life is established, considerable diversification in nest architecture and social organization may happen relatively rapidly, probably much influenced by various ecological pressures, especially predation. 10. A number of species of Vespula, Vespa and Polistes have become social parasites, in the first and the last cases without a worker caste. This seems like a wrong turning taken in the course of the normal evolution of the queen‐worker relationship. 11. A brief account is given of the attempts to provide a theoretical framework for the population dynamics of wasp‐colonies. 12. An account is given of the relations of wasps, chiefly as foragers, with plants and with other animals.