E. A. Cockayne 11!) 



Von Siebold, tor instance, tliought that in .sucli cases (jne sperma- 

 tozoon enters the egg instead of many. This is incorrect for, though 

 polyspermy is very usual in Lopidoptera, only one spermatozoon- is 

 necessary and only one takes part in the actual process of union with 

 the female pronucleus. Morgan supposes that two spermatozoa enter 

 one egg, on(> uniting with its nucleus, the other developing alone. This 

 is open to tiie same obj(>ctions as von Diinhoff's theory, i.e. the existence 

 of halved gynandromorphous hybrids, showing the characters of both 

 parents on both sides equally. 



Menzel's supposition is based on the phenomenal numbers of gynan- 

 dromorph bees of the Eugster hive. He considered that the queen was 

 malformed and that the ova progi-essed so slowly past the receptaculum 

 seminis that cell division had already commenced before the sperma- 

 tozoon entered the egg. A parthenogenetically developing egg of the 

 honey-bee usually produces a drone, so those ova, which began to 

 develop parthenogenetically and continued in a more normal way, 

 produced bees, part drone, part worker. 



Lang also thought that premature division of the egg nucleus 

 occurred and that the resulting nuclei were fertilised by spermatozoa 

 which were different in their characters. 



In connexion with this subject the experiments of Herbst on sea 

 urchins are interesting. He obtained hybrids artificially between 

 Strong ylocentrotus and S ph a er echinus, and these were intermediate in 

 characters between the two parent species. But by treating the eggs 

 for 1^ hours with fatty acid before adding the spermatozoa, he obtained 

 a hybrid very like the female parent. On examination of eggs so 

 treated the nuclei were found to have undergone mitotic division and 

 modification before the spermatozoa reached them. Paternal chromo- 

 somes were eliminated in this way. A few eggs so treated produced 

 plutei with one half resembling the male and one half the female 

 parent. This he regarded as due to irregular distribution of chromatin 

 to the first two cleavage cells, one receiving too much paternal chromatin 

 and one too little. 



Boveri has produced sea urchins in one half of which all the 

 chromosomes are purely maternal, in the other a mixture of paternal 

 and maternal. From this observation he thinks gynandromorphs may 

 be similarly constituted, and supports a modification of von Donhoff's 

 theory. 



This phenomenon of hybridity in sea urchins obviously bears a 

 resemblance to halved gynandnMnorphism. But that it is identical 



