EMBRYOLOGY OF THALASSEMA MELLITA 219 



dition of affairs in Aplysia, as recently described by Carazzi 

 ('oo), seems to stand as a connecting link. Here each of the J/'s 

 buds off four small cells, three of which are mesoblastic and one 

 entoblastic. Yet in Umbrella, according to Heymons ('93), 

 although the divisions are the same in number and quite similar 

 in character, all of the cells are stated to be mesoblastic. 

 Finally, in the aberrant cleavage of Capitella, corresponding 

 cells are believed by Eisig to give rise to the paedomesoblast. 



The following interesting possibility has recently been sug- 

 gested by Wilson ('98) in regard to these cells. If we interpret 

 them as having the same morphological significance in every 

 case, e. g., entoblastic, we may arrange a series of annelids and 

 molluscs in which at one extreme the entoblast budded off from 

 the posterior member of the fourth quartet is greater in amount 

 than the resulting pole-cells of the mesoblast bands, as is actually 

 the case in Crepidiila (Conklin), and at the other the entoblastic 

 element is contained in a mere rudimentary cell as in Aricia. 

 This graded series, according to Wilson, may represent a gradual 

 elimination of the entoblastic element from the posterior mem- 

 ber of the fourth quartet and the final complete conversion of 

 the latter into mesoblast. It also clearly indicates the possi- 

 bility that the mesoblast pole-cells are to be regarded phylo- 

 genetically as derivatives of the archenteron, as was long since 

 suggested by Kowalevsky ('/i). With this idea the fate of the 

 small cells in Podarke and TJialasseina also falls in line. 



In the light of subsequent research, however, it would seem 

 that we must recognize the probability that the small cells 

 budded off from the coelomesoblast pole-cells before formation 

 of the coelomesoblast bands may be either entoblastic, meso- 

 blastic, or both. Of especial interest in this connection is a com- 

 parison of the results of Carazzi and Heymons (see above). In 

 Aplysia (Carazzi) only one of the four divisions eliminates ento- 

 blast, while in Umbrella (Heymons) even this is lost and all of 

 the small cells are mesoblastic. This suggests the possibility 

 that the form of division has persisted while the resulting cells 

 have changed their prospective value, or, to adopt Wilson's ter- 

 minology, that the small cells still remain homoblastic but are 



