The Bionomics of South African Insects. 468 



dangers attend the phenomenon, and I think it is very 

 probable that they are more than compensated b}'- tlie 

 analogous benefits. The instinct to emigrate probably 

 exists in a dormant state in all species liable from their 

 powers of rapid multiplication suddenly to outrun the 

 food-supply in any part of their range. The stimulus 

 which evokes the instinct is, in such insects as the locust, 

 or such mammals as the lemming, probably merely the 

 direct and obvious incentive of hunger (A. R, Wallace, 

 " Geographical Distribution," London, 1876, vol. i, p. 18). 

 In the majority of perfect insects, however, we cannot 

 accept this interpretation, and we are compelled to look 

 for a stimulus in some other result of undue increase — the 

 crowds of individuals everywhere, the food-plant covered 

 with eggs and young larva?, and females laying still more 

 eggs. Then probably arises the imperative instinct to 

 move, perhaps in both sexes, perhaps only in the female, 

 the males accompanying them (in many species in far 

 larger numbers). And the instinct further compels the 

 individuals to move together in vast masses in the 

 same direction, rather than to scatter and fly in all 

 directions. The increased danger from enemies is of 

 course lessened, as compared with the hybernating com- 

 panies, by the enormous number of emigrating individ vials; 

 but there is, I believe, the solid advantage that fresh food- 

 plant may be found in another uncrowded area; that the 

 limits of the noi'mal range of the species may be overpassed; 

 that areas from which the species has been driven may 

 be regained : — not by single individuals or by a very few 

 pairs, but by immense numbers of both sexes without any 

 of the dangers of in-and-in breeding when once they have 

 established themselves as a fresh colony. In this way 

 the range of many species has probably been extended 

 in the past, and, although the emigrating crowds so often 

 described may again and again be landed in a foodless 

 desert or the sea, the instinct iy advantageous in that it 

 utilizes individuals which are at the moment useless and 

 even injurious to their kind, in a manner which may be 

 in a high degree beneficial (see also Trimen, " South 

 African Butterflies," vol i, 1887, p. 31). The suggestion 

 is made that the crowded masses, resulting from over- 

 production and inability of enemies to cope with the 

 increase, are injurious to the species, because it is likely 

 that food-plants would be checked for years or even killed 



