the Sexual Generation. 153 



galls 'lay their eggs. The few males that are still 

 produced are thus superfluous, and we can predict 

 that they will probably become extinct in the course of 

 time. In the other species, Rhodites eglantc7'iae, some 

 males have been observed, but after repeated experi- 

 ments, I have only succeeded in obtaining females. 



These facts show that parthenogenesis occurs very 

 frequently among the Hymenoptera, and that it then 

 springs directly from the sexual generation. The 

 result as regards the sex of the offspring is varied, and 

 appears to be subject to no definite law. Among some 

 Hymenoptera, especiallyamong bees, malespreponderate 

 when the eggs are unfertilized ; among gall-flies females 

 usually preponderate \ It appears that where partheno- 

 genesis has been long continued, males ultimately 

 become extinct : they are unknown in Nematits Vallis- 

 nierii, and in several species of Aphilotrix, but it is not 

 utterly impossible that a male might occur. In the 

 alternating generations of gall-flies the case is somewhat 

 different ; the agamous generation occurs only in the 

 female sex, while in the sexual generation males and 

 females emerge in equal numbers. Since, however, it is 

 the agamous generation which produces both sexes, we 

 may assume, a pr'iori, that in the ovary there are plasms 



[^ Rolph classifies the offspring of parthenogenesis in groups (i) $ 

 only ; ^ froni fertilized ova (Apidae). (2": One mixed brood 

 (Cynipidae). (3) Mostly (^ , occasional p (Nematus^. 4^ Mcstp^ 

 periodic or exceptional $ (Apus). (5) p only (Rotifera). 



Ova fertilized by the queen-bee yield workers and queens. If the 

 nuptial flight be prevented, or the queen be old, and her sperm store 

 exhausted, the ova yield drones only ; but Blochmann saj'S these ova 

 show two polar bodies. May these be the first polar body divided ? 

 If not, is the next generation fertile ?] 



