NATURAL SCIENCES OF PHILADELPHIA. 151 



obtained in this way. All proved to be females. Similar investigations, which 

 he undertook with several thousands of specimens of different other species, 

 led to the same result. This total absence of one sex in the genera Cynips, 

 Neurosterus, Apophylhis Hartig, appeared the more strange and unaccountable, 

 as in the closely allied genera Teras, Andricus, Trigonaspis, Spathegaster, males 

 and females occurred in almost equal numbers. The supposition that the 

 males had escaped the investigations of entomologists seemed inadmissible in 

 the case of insects obtained mostly from rearing and not by means of catching. 

 This, combined with the circumstance that the presence or absence of the 

 males occurred constantly through whole genera and not merely in single 

 species, seemed to favor the belief that in some genera the males did not exist 

 at all. 



More than one hypothesis has been proposed for the solution of this anoma- 

 ly. Hartig, after having studied the anatomy of some Cynipidse, imagined 

 that he had discovered their being androgynous. Soon, however, he gave up 

 this conclusion and confessed his error. Erichson proposed another solution. 

 Those who have reared gall flies know that very frequently galls of the same 

 kind produce Cynipidce of two different forms, one of them being the true gall- 

 producing Cynips, (subfamily Psenides Hartig,) the other having been hitherto 

 considered as a parasite, (subfamily Inquilinee.) The latter occur in male 

 and female specimens. Erichson suggested, not as a result of actual observa- 

 tion, but as a mere idea for future investigation, that the two forms, reared 

 from the same gall, however different in their characters, might be the same 

 species ; that the male inquilinee were in reality the males of Cynips, and that 

 their females had to be considered as a second form of the gall-producing 

 female Cynps. He reminded that this would not be the only instance, among 

 insects, of the occurrence of two forms of females, the same being the case with 

 some butterflies and some Dytisci. 



The hypothesis of a parthenogenesis among the Cynipid e has also been re- 

 sorted to, and Siebold discusses it in the latter part of his well-known pam- 

 phlet. 



If I have dwelt at some length on all these hypotheses, it was in order to 

 show the importance which is attached to this question and the degree of its 

 apparent inextricability, which has compelled men of science to have recourse 

 to such improbable and far-fetched explanations. 



Engaged since last autumn in the study of galls and gall-flies, I have hap- 

 pened quite recently to make an observation which would lead to a more 

 simple solution of the vexed question. 



I discovered on the leaf of the red oak a very curious, elongated-fusiform, 

 petiolate, pale green gall, about an inch long, and, in its stoutest part, but a 

 little more than one-tenth of an inch broad. Having been lucky enough to 

 rear the insect from it, I was struck by its resemblance with Cynips confluens 

 Harris, the originator of the large, globular gall of the- red oak, commonly 

 called oak apple. The specimen being a male, and the specimens of C. con- 

 fluenta which I have seen being all females, it struck me immediately that 

 this was the solution of the mystery. Male and female occur in galls of two 

 altogether different forms ! 



It it is so, one might object, how does it happen that this observation has 

 never been made before ? Why did the German entomologists, having reared 

 so many, (I believe more than 150 gall-flies), never come across it ? The ex- 

 planation may be this : Generally the distinguishing characters of the species 

 of this family are so slight, that it would have been difficult to assign a certain 

 male to a certain female without having reared them from the same gall. 

 Eutomologists may have reared males from one gall and their females from 

 another, without suspecting their identity. If I succeeded better, it was merely 

 because I happened to make my observation on a species with very striking 

 characters, which were therefore easy to identify although the specimens were 

 1861.] 



