BEHAVIOR OF CHROMATIN IN HYBRIDS 503 



mosomes. With Ehrlich's hematoxylin, however, no mistake 

 can be made, as it is only the chromatic material that takes the 

 deep purple stain, all the rest of the egg being stained red. 



No records were made of the percentage of eggs dividing in 

 the hybrid cultures, so that only the most general statements can 

 be made on this point. The Fundulus eggs are fertilized readily 

 with the Ctenolabrus sperm and it is estimated that about half of 

 those in a hybrid cultm'e go through cleavage. Division was much 

 rarer in hybrid Ctenolabrus eggs, not more than 10 per cent of 

 them being fertilized. The head of the Fundulus sperm is con- 

 siderably larger than that of the Ctenolabrus sperm (figs. 35 and 

 36) and this may account for the greater ease with which the 

 F 9 X C cf cross is made. 



Polyspermy is apjmrently rare. Only one tetraster was found 

 in the preser\"ed Fundulus material, and one case is recorded of 

 a 4-cell stage in a culture in which the other eggs were in the 

 2-cell stage. This was probably a dispermic egg, dividing at 

 once into four cells. There was no e\'idence of polyspermy in 

 the Ctenolabrus eggs, but all the evidence from these eggs is poor, 

 and pol3'spermy may possibly occur in them. 



In neither cross did I succeed in raising the hybrids to a late 

 stage of development. Loeb has raised hybrids from the same 

 two species of fish, but it is necessary to put only three or four 

 eggs into a single dish, and take particular care to change the 

 water frequently. Even then the proportion of late stages is 

 small, so that to have satisfactory results one must fertilize large 

 nunil^ers of eggs daily. With the limited space at my command, 

 T was unable to use this method. The conditions under which 

 I kept the eggs were perfectly satisfactory for the normally fer- 

 tilized ones, but not for the less hardy hybrids. The Fundulus 

 eggs died about twenty-four hours after fertilization, at the time 

 when the embryo has begun to form and the germ-ring has cov- 

 ered about half of the egg. The Ctenolabrus eggs also died 

 within twenty-four hours; but they had developed further in 

 that time than the Fundulus h^^brids, the natural rate of devel- 

 opment being more rapid for Ctenolabrus than for Fundulus. 



