234 THE entomologist's eecoed. 



dragonflies, butterflies and moths. Winged ants, as we have seen, 

 collect in vast numbers when they first emerge, and it seems to be a 

 general habit among many species to disperse as quickly as possible, 

 and to get as far away as may be from the spot where they themselves 

 have been reared. The necessity of this is self-evident, if the reason 

 we have previously suggested be the correct one, viz., that crossing 

 between the inhabitants of different nests may occur, for the males are 

 exceedingly short-lived and the females lose their wings as soon as 

 fertilisation has taken place. There seems, however, to be a possi- 

 bility of confusion between the different phenomena here presented. 

 The Driver ants {Anow)i)a) seem to have no fixed homes, but to live 

 in a constant state of migration, en masse, and simply moving in 

 search of food. Colonies of Fonnicidae or Mj/rmicidae only occasion- 

 ally shift their quarters en )))asse (sometimes more than once in a 

 season, at others not for years), and then merely because the old 

 quarters have become for some reason inconvenient, and does not 

 result in the establishment of fresh communities like the " swarm- 

 ings " of Apis nielli nca, or the autumnal dispersal of fecundated J 

 ants after the nuptial swarming. In the latter case the original com- 

 munity remains, the old queens having long ago lost their wings, 

 whilst each fresh queen that is successful starts a new community 

 elsewhere. The comparatively few species of social wasps and bees 

 {B(i)itbiis) do not exactly " swarm " like ants, i.('., there is no collective 

 exodus of all the males and females on one day, still it seems that, 

 after fecundation, their females disperse to various quarters and start 

 new colonies, each separately, next spring, the old colony perishing 

 entirely at the approach of winter. The solitary bees and wasps do 

 not properly swarm at all. They appear, it is true, in multitudes, at 

 certain times and places, possibly not from any instinct to congregate, 

 but simply from the simultaneous birth of many individuals from 

 eggs laid together under favourable conditions, the individuals com- 

 prising such pseudo-swarms proceeding to copulate and nidificate at 

 no distance from their own birth-places, forming as it were a sort of 

 colony, which is, however, in no sense a true community, but merely 

 a number of individuals living side by side, so long as conditions are 

 favourable, but breaking up as soon as they cease to be so, choosing a 

 new place, where, by chance, some of its members may find them- 

 selves together again, the movement being individual and not collec- 

 tive, a totally different phenomenon from the collective migration of 

 a society, or from the movements resulting in the production of new 

 societies by the dispersal of founder-queens. The fossorial hymen- 

 optera, Chrysids, &c., never really migrate collectively ; they simply 

 abound, apparently, but not really, swarming, wherever the environ- 

 ment is favourable. They appear to be found in aggregations not 

 congregations. The migrations of Tenthredinidae possibly mean little 

 more than exceptional abundance locally, the swarm being transported 

 mechanically, but, here, as with the movements of aphides en masse, 

 more evidence is necessary. The swarming of Ajiis melli/iea in con- 

 finement appears to be a real instinctive phenomenon, due to the stock 

 outgrowing its hive ; but there seems no reason why, if a colony have 

 the' power to extend its quarters ad intinitidii, these should swarm at all. 

 ]\Iorice, to whom we are indebted for the above suggestions, says 

 {in litt.) : "On the whole, I believe that migration proper plays 



