liK) Studies in the rinjdohMpi of FertiUzatioii 



of a given race ; second, what factors influence the degree of self- 

 sterility ; third, the causes of the self-sterility; fourth, the inherit- 

 ance of self-sterility ; and fifth, the effect of self-fertilization on the 

 offspring. The present investigation was commenced in the belief 

 that IJiamt intestitiuliH is almost, if not coniplet(>ly, self-sterile, in which 

 case the first, seci.md, and fifth of the problems just eiuinierated would 

 be non-existent. It was proposed to investigate the inheritance of 

 the self-sterility on the lines of the extremely interesting work b}^ 

 Correns (4) on Cardamine pratensis. That is to say, separate cross- 

 fertilized families would be raised to maturity, and then the membei-s of 

 a family be crossed inter se, with individuals of the family derived from 

 the reciprocal cross and with unrelated individuals. 



Castle (1) originally demonstrated the self-sterility of Ciona in- 

 testinalis when making fertilizations to obtain stages in the development 

 of the embryo. The animals used were from Newport. A number of 

 specimens were isolated in separate dishes, where they discharged eggs 

 and sperm simultaneously nearly every day. In the majority of cases 

 none of these eggs were found to have been fertilized, although in some 

 as many as 90 "/„ segmented. In contrast with this, when several 

 animals were placed together in an aquarium, usually 100°/^ of eggs 

 which were discharged segmented. Repeating the observations with 

 eggs and sperm artificially removed from the genital ducts, he found 

 that self-fertilization usually caused few or no eggs to segment, although 

 in one instance 50 " j ^ were self-fertilized. Similar observations were 

 made by Gutherz in 1903 (7a). 



During the course of an extensive investigation on the phenomenon 

 of self-sterility in Ciona, Morgan never found such high percentages of 

 eggs self-fertilized as Castle had done. The first part of Morgan's work 

 was done at Woods Hole in 1903, when he states (9, p. 138) with 

 regard to artificial fertilizations, that " I have rarely seen more than 

 from one to ten per cent, of self- fertilized eggs segment, and in the greater 

 number of cases not a single egg segmented.'' The second part of the 

 work was done at Coronado Beach, California, in 1905. Here, in makingr 

 fertilizations w"ith eggs and sperui removed artificially from the bodies 

 of the animals, " in only one case out of many hundreds of eggs mixed 

 with their own sperm did fertilization occur." The final part of the 

 investigation was carried out at Woods Hole in 1909 (11), and this 

 time cases of self-fertilization were exti-emely rare. 



Castle himself suggested that when high self-fertilization per- 

 centages were obtained, tho result may really have bc^en due to the 



