Mr Buxton, Animal Oecology in Deserts 391 



a small element in the fauna which is not protectively coloured, 

 and these animals are all black. 



Concentration. 



One is justified in saying that concentration in point of time 

 and of place is a characteristic of desert animals. The rain falls, the 

 plants blossom, the animals appear and breed and are no more 

 seen. All that is most apparent in the life of the desert is con- 

 centrated in a couple of months. Concentration in place is just as 

 noticeable. The great leaf-bases of the date palms in Lower 

 Mesopotamia give cover, especially after rain, to a host of animals, 

 grass-hoppers, beetles, bugs, ants, termites, centipedes, milhpedes, 

 woodlice, scorpions, spiders, snails, oligochaets, lizards and others. 

 In deserts generally plants do not cover the ground and every bush 

 or patch of scrub is an oasis in itself, full of specialized forms of 

 animal life. In some places in the Algerian deserts for instance, 

 nearly every large stone shelters a collection of insects, myriapods, 

 arachnids, isopods, and often also lizards, snakes, and even small 

 mammals and birds. The hordes of migratory birds which suddenly 

 appear and pack every bush in an oasis are yet another example of 

 the concentration of life in the desert. 



Winter and surmner. 



In the palaearctic desert region the butterflies for the most 

 part are on the wing in spring, and a few appear also in the autumn. 

 In cold weather they are not seen and presumably they pass the 

 winter as they do in more northerly climates: that is to say some 

 of them as eggs, some as hibernating larvae, some as pupae. In 

 the hot weather also the majority of species are not in the imago 

 stage, and we do not yet know what they are doing. One might 

 suppose that the rise of temperature would cause the stages to be 

 passed through more and more rapidly, but this is not the case. 

 In Lower Mesopotamia (at Amara) Colias croceus {edusa) is on the 

 wing from March to May, and in November and December ; Pieris 

 rapae February to May and October to December, and as a great 

 rarity in summer. Both these species certainly produce a succession 

 of broods in spring, and probably in autumn also. They are in fact 

 continuous-brooded while the temperature is neither very hot nor 

 very cold. The three lycaenids, Zizera lysimmon, Tarucus mediter- 

 raneae and T. balcanicus, have, I think, similar periods of emergence, 

 but I have not a sufficient number of records for each species to 

 enable me to state definitely that this is the case. We do not know 

 in what stage these five species pass the summer, nor what is the 

 factor which retards their development during the hot season. It 

 is not lack of food for the pabula of C. croceus and P. rapae are 



