138 George Lefevre 



sion and produce ciliated structures faintly simulating the 

 appearance of trochophore larvae. My observations on Thalas- 

 sema, however, bring the annelids into closer accord with the re- 

 sults which have been obtained with echinoderms, especially with 

 the starfish, where parthenogenetic development has been shown 

 to be far more normal in character than in other groups of animals 

 experimented with. 



The differences between the parthenogenetic embryos of Thal- 

 assema and the structures which have been obtained from the 

 unfertilized eggs of other annelids, notably Chaetopterus, are most 

 marked and involve important considerations. Differentiation 

 without cell division has never been observed in Thalassema, but 

 on the contrary progressive differentiation of the embryo in this case 

 depends upon cell division at every stage. 



The formation of the differentiated masses of ciliated proto- 

 plasm which he observed in Chaetopterus, Lillie insists, must be 

 interpreted as just "as truly a process of development as the forma- 

 tion of the trochophore" ('02, p. 493). The eggs, without dividing 

 into cells "pass through well defined phases of differentiation, the 

 yolk accumulating in a dense mass in the interior, and the periph- 

 eral cytoplasm becoming vacuolated and. ciliated. The ciliated 

 ectoplasm and the yolk laden endoplasm are analogous to the 

 ectoderm and endoderm of the trochophore, and the phases of 

 differentiation resemble some of the normal processes, though 

 the resulting object can by no stretch of the term be properly 

 called a trochophore" (i.e., p. 477). Furthermore, "in some cases 

 it is even possible to homologize the regions of these unsegmented 

 ciliated eggs with the regions of the trochophore" (p. 495). The 

 process of cell division, as such, Lillie concludes, "is necessary 

 neither to growth, differentiation, nor the earliest correlations; 

 but it is accessory in Metazoa, to all three as a localizing factor, 

 often from the earliest stages" (p. 494). 



These results are in sharp contrast with my observations on 

 Thalassema, as it has been shown that the processes of differentia- 

 tion, if they do not depend upon cell division, nevertheless, do not 

 occur in its absence. In this form, if the egg remains unsegmented, 

 no differentiation of the cytoplasm takes place and a ciliated em- 



