ALTEENATION OF GENERATION. 7 



ceeding give forth, flies which are all female, and 

 which produce the woolly galls of spring" (Johnson's 

 Cyclopsedia, p. 422). From this and other observa- 

 tions Riley deduces the fact that there can be little 

 doubt that " all species known only in the female sex 

 exist also in the bisexual form, though the gall pro- 

 ducing this last may present an entirely different 

 appearance from that producing the former" (I.e.). 

 That, however, is too trenchant a surmise, for in 

 Europe at least we know species only existing in the 

 female sex. 



But it was not until the classic researches of Adler 

 appeared that the natural history of the Cynipidse 

 was thoroughly elucidated. By a series of careful 

 and long-continued elaborate studies he proved that, 

 at least so far as the North European species were 

 concerned, there is no one species with two sexes but 

 has an autumnal agamic form — that there was an alter- 

 nation of a bisexual with a unisexual form — that here 

 we had a case of alternation of generations, and not 

 of dimorphism, as was surmised by Walsh. 



How are we to explain the occurrence of this 

 remarkable phenomenon ? We find (1) species (but 

 not on the oak) with only one generation, and having 

 males and females ; (2) species on oak with one gene- 

 ration and no male ; and (3) species with two genera- 

 tions, the one sexual, the other (autumnal and winter) 

 agamic. In the first division there is undeniable evi- 

 dence of parthenogenesis ; and, further, as the case of 

 Rhodites rosce testifies, it is clear that the males have 

 disappeared as the faculty of virgin reproduction in- 

 creased. From this fact, and from the considerations 

 given in vol. i, p. 29, and ii, p. 218, we may fairly 

 assume that those of the second section had at one 

 time males. So far, therefore, as these two groups are 

 concerned, there is nothing unusual, since we find the 

 same phenomenon in the Tenthredinidge and in other 

 insects ; but in the third division we have a state of 

 matters much more unusual and complex, and conse- 



