380 



and start development. Second, How far will development proceed in the 

 various combinations, and in what respects are the processes normal and 

 abnormal ? 



In every combination effected the earlier phases of cleavage are passed 

 through in a i>erfectly normal manner. The same is true of the later stages 

 of cleavage excepting the rate of development. This will be further consid- 

 ered below. From the late cleavage on. the history of the different hybrids 

 becomes much more varied. In those hybrids resulting from species nearly 

 related — belonging to the same genus or to closely allied genera — most of 

 the embryos may complete their development to the point of hatching, or 

 beyond. Even among these, however, a number variable but much greater 

 than in normal embryos, may show abnormalities along the course of their 

 development, such as occur more abundantly in the hybrids between more 

 distantly related forms. Hybrids between species more distantly related 

 than above indicated, so far as my experiments go, never complete their de- 

 velopment to the point of hatching. The stage to which they will go depends 

 again upon the nearness of their relationship. In the more successful of 

 such distant crosses Fuudulus-Menidia hybrids, many of the embryos may 

 go far enough to form fairly well developed eyes, ear vesicles, tail, muscles, 

 central nervous system, heart, color i^atern, fins, etc.. but many of these 

 structures in the later stages are variously abnormal. A large proportion of 

 all the embryos, however, fail to reach such advanced stage. From these 

 hybrids we have almost every condition to such as obtains in the hybrids 

 between Fuudulus heteroclitus x Tautogolabrus adspersus, where none of 

 the embryos go much beyond the closure of the blastopore, and where it is 

 not possible to speak of the formation of organs. The more characteristic 

 and striking abnormalities appearing beyond the cleavage stage in these 

 various hybrids may be briefly considered. 



In the last stages of cleavage and during the earlier phases of germ 

 ring formation it is usually not possible to distinguish the hybrids from the 

 normals excepting in the stage of advancement. In some combinations, such 

 as Fuudulus heteroclitus x Tautogolabrus adspersus, etc., one can very com- 

 monly see the formation of a rather large clear area under the blastodisc 

 which is tilled vrith a clear fluid. I have followed such eggs and they do 

 not bring their development to as advanced a stage as those eggs of the 

 same lots that do not show this abnormality. They may form a very good 

 embryonic ring and shield and may overlap the yolk for a third of the way 



