PLANT-EATING TNSECTS 205 



gave the insects which emerge from each kind of gall a 

 different name, calling the spangle gall-wasp Neurotcrus 

 lenticular is, and its currant gall form Spathegaster bac- 

 carum. For the sake of convenience these names are 

 still retained ; we must bear in mind, however, that they 

 do not designate two different kinds of insect, but merely 

 two alternating forms of the same species. 



A phenomenon so remarkable as alternation of genera- 

 tion has naturally given rise to much discussion and 

 theorising ; but so far no adequate explanation of the fact 

 has been advanced. Moreover, the case of the marble 

 gall-wasp (Cynips kollari) is even more astonishing and 

 mysterious. This species has only one generation in the 

 year, and every individual is a female. No male has ever 

 been discovered, although from time to time entomologists 

 have bred many thousands of specimens in captivity, 

 Among British gall- wasps, therefore, we find (1) species 

 (such as those of the wild rose mentioned above) with 

 only one generation, but having males and females ; (2) 

 species such as Cynips kollari with one generation and no 

 males ; (3) species with two generations, the one agamic 

 and the other sexual. Observation of the first group has 

 shown that at least in certain species parthenogenesis is 

 undoubtedly prevalent. In the case of the bedeguar gall- 

 wasp, for example, it is probably the rule, seeing that the 

 males of this insect are excessively rare. We may 

 suppose, therefore, that the males tend to disappear as the 

 faculty of virgin reproduction increases ; and thus we find 

 a possible clue to the mystery of the marble gall-wasp, 

 which, as we have seen, has no males at all. But in the 

 present state of our knowledge it seems impossible to 

 explain adequately the phenomenon of an agamic genera- 

 tion alternating with a sexual one. The entomologist 

 Adler, who devoted much time and thought to this 



