The Behavior of the Chromosomes as Studied through Linkage. 257 



Cytoplasmic Inheritance. 



Man.v authors have insisted that the cytoplasm, as well as the 

 nucleus, is directly involved in heredity. There are several experiments 

 which are cited in favor of this view. When sea-urchin eggs are 

 fertilized with sperm of starfish, crinoids, annelids, or moUusks there 

 result purely maternal larvae (Loeb '08, Godlewski '11, Kupel wieser 

 '09, &c.). In some of these cases the paternal chromatin can he seen to 

 behave abnormally, and it seems likely that it does not function normally 

 in any of these crosses between widely different forms, so the result 

 may be due simply to the action of the maternal chromatin. However, 

 Godlewski ('11) has succeeded in fertilizing an enucleated egg fragment 

 of a sea-iu'chin with crinoid sperm; and he obtained a maternal lan^a. 

 Here there can be no question of maternal chromatin, and the cytoplasm 

 must be the determining factor. 



The difficulty with these experiments is that they deal only with 

 the early stages of the Fi individuals from crosses. They show that 

 the characters of these may be determined by the maternal cytoplasm, 

 but the critical data regarding the next generation are lacking. I am 

 acquainted ^\^th only two cases where these data are supplied, but these 

 cases furnish what seems to me to be the key to the whole matter. 

 TOYAMA ('12) finds in the silkworm moth certain races which differ in 

 the color of the serosa of the embryos. When these breeds are crossed 

 the Fl embryos always have the color of the maternal race, no matter 

 in which direction the cross is made. This then must be a case of 

 "cytoplasmic inheritance". But Tüya_m.\ demonstrates that this character 

 is strictly Mendelian, and shows complete segregation. The two kinds 

 of Fl moths behave exactly alike when bred. The color of the embryonic 

 serosa is then determined solely by the mother, but is influenced by 

 the maternal grandfather as much as by the maternal grandmother. 

 Obviously we are dealing here with a character determined by the 

 chrnmatin before fertilization; therefore the somatic »'haracter of any 

 individual is determined l)y its mother, but the germinal constitution is 

 derived equally from both parents*). A similar explanation will cover 

 the cases of cytoplasmic transmission described above. 



There is another grouj) of cases often cited as showing the occurrence 

 of cytoplasmic inheritance, and with more justice than those just dis- 



') Shull ('13) has described a similar case in Hydatina, and has drawn concloBions 

 similar to those here expressed. 



Induktive Alistammangs- und Vercrbungalchrc. XUI. Ig 



