ON PARTHENOGENESIS IN THE HYMENOPTEBA. \&J 



and hybrid. As a matter of fact, there is no proof 

 that three sorts of workers inhabited the hives ; 

 and farther, we have no evidence that workers lay 

 eggs so commonly as would be necessary to explain 

 so large a number of mixed drones, especially where, 

 as in Perez's experiments, the queen happened to 

 be laying for the first time. 



(Second) — It has been suggested that they are cases 

 of atavism — of reversion to the original form. But 

 which is the original form ? — mellifica, or ligurica, or 

 fasciata? The dubiousness of this theory becomes 

 apparent when we remember that the same results 

 happen whether an Italian queen be fertilised by a 

 common drone or vice versa. 



It has also been suggested by Lubbock that the 

 mixed drones were actually produced partheno- 

 genetically, but that the male of a different race 

 may have so influenced the ovary as to affect the 

 purity of the future progeny, just as when (say) a 

 pure mastiff bitch is fertilised by a dog of another 

 breed she ceases afterwards to have pure pups. 

 There may, of course, be some truth in this ; but I 

 must confess that 1 cannot see, with our present 

 knowledge of the modes of generation in insects, 

 why we should pin our faith so strongly to the 

 absolute uniformity of the parthenogenetic mode of 

 reproduction of the drones in the hive-bee. Consider 

 that the queen is not a cast-iron machine : she is a 

 living being, subject to various troubles, and exposed 

 more or less to the varying influences of her 

 surroundings, whether meteorological or connected 

 with the denizens of the hive ; and surely it may 

 be concluded that while the general rule is that the 

 male does not fertilise the drone-eggs, yet, from 

 various circumstances, more or less of them may be 

 fertilised from time to time. To my mind that 

 conclusion is more in harmony with the observed 

 facts than any of the reasons I have just mentioned. 

 That there is not invariable uniformity in the 

 process of fertilisation might be proved by the 



