ARTIFICIAL PARTHENOGENESIS IN CUMINGIA 29 



number throughout its development. The essential point in all 

 these cases is of course the same; namely, that the females 

 develop from eggs in which reduction has not occurred. 



Finally, there are two cases of natural parthenogenes^'s de- 

 scribed in which there is apparently a fertilization of the egg 

 by a polar nucleus — a condition similar to that induced in the 

 eggs of Cumingia by the treatment described above. The first 

 of these instances is that of Astropecten. Natural partheno- 

 genesis had been observed in the eggs of this starfish by Greeff 

 in 1876 and the cytological phenomena were studied by 0. 

 Hertwig in 1890. In this egg the first polar body is given off, 

 and the second polar spindle formed normally. But instead of 

 being extruded, the second polar nucleus is formed inside the egg 

 and fuses with the female pronucleus. The segmentation of such 

 eggs is irregular, seldom leading to the formation of normal 

 blastulae. Some eggs were found which had formed two polar 

 bodies, but these seemed incapable of further development. 



Brauer's work ('94) on Ai'temia salina shows a similar con- 

 dition. In the parthenogenetic eggs of this crustacean, the 

 first polar body is given off, but the second is not. Develop- 

 ment may proceed, according to Brauer, in one of two ways. A 

 cleavage spindle may be formed at once, or the second polar 

 spindle may form and give rise to two nuclei which are both re- 

 tained in the egg, and fuse to form a cleavage nucleus. In the 

 first case, the egg develops with 84 chromosomes, in the second 

 with 168. Brauer assumes that the 84 chromosomes were 

 bivalents and concludes that '^fiir die Einleitung einer normalen 

 Entwickelung eines Thieres die Zahl der Chromosomen gleich- 

 gtiltig ist wenn niir die Masse dieselbe ist." Here, also, when 

 both polar bodies are formed, there is no normal development. 

 It may be mentioned that although this work of Brauer's has 

 been widely accepted and quoted, Hertwig's on Atropecten is 

 really a more convincing instance of self-fertilization. Petrunke- 

 vitch ('02) has pointed out the lack of intermediate stages in 

 Brauer's series between the formation of the second polar 

 spindle and the fusion of the two nuclei, and has stated his be- 

 lief that the binucleate condition is a pathological one, not lead- 



