i 4 4 THE VARIATION OF ANIMALS IN NATURE 



alike both as larvae and adults, and the food-plants are 

 identical. 



Differences of this type seem to be fairly common in 

 phytophagous insects, but there is usually some overlap between 

 the seasons. Where the female of a species is always impreg- 

 nated immediately after emergence and the male emerges 

 before the female, very small differences in the total period of 

 occurrence may have considerable effect. 



In other cases seasonal occurrence appears to play no 

 part in isolation. Thus Schubert (1929), in his account of 

 the dragon-flies of the neighbourhood of Neustadt, records 

 that all the 18 species (6 genera) have overlapping periods, 

 with the possible exception of the two species of Orthetrum. 

 Richards (1930, p. 321), in his account of the British flies of 

 the family Sphaeroceridae, shows that most of the species occur 

 throughout the year, and many of them seem to have no 

 restricted breeding season. 



Isolation by means of differences in seasonal occurrence 

 has a special interest because of its relation to the environment. 

 It is a general rule for insects to have more broods in the south 

 than in the north and, although partial broods, in which only 

 a few individuals of a given generation emerge, are often 

 found, there is a natural tendency for a species to fix on a 

 definite reproductive rhythm. The intermediate state, where 

 partial broods are formed, would appear to be one of unstable 

 equilibrium. A species which is single-brooded in the north 

 will be double-brooded in the south and, if the range is suffi- 

 ciently great, even more broods may develop still further south 

 — e.g. Agrotis segetum (Filipjev, 1929), Pyrausta nubilalis Hb. 

 (Babcock, 1927). 



Owing to climatic conditions there will be a tendency for 

 the single-brooded form to occur between the broods of the 

 bivoltine form in time. If we knew more as to how such 

 rhythms become fixed, we might see a way in which the two 

 forms could remain permanently isolated, even if their ranges 

 came to overlap. This subject has been ably reviewed by 

 Uvarov (1931, pp. 104 ff), who concludes that rhythms 

 originally induced by climatic conditions are eventually 

 hereditarily fixed. Pictet (1913), experimenting on Lasiocampa 

 quercus, obtained results suggestive of such a process. (See 

 also Chapter II.) 



