CASTLE. — CELL LINEAGE OP THE ASCIDIAN EGG. 201 



is accordingly ivrong. Their terms ectodermal and endodermal, ven- 

 tral and dorsaly as employed up to this stage, must he interchanged. 



I shall iu the present paper reproduce the figures of the 32-cell 

 and 44-cell stages given by Van Beueden et Juliu, with the proper 

 orientation and tlie probable cell lineage as inferred from that actu- 

 ally determined in Ciona. 



Chabry, iu a paper concerned chiefly with teratology, traces the 

 cell lineage of Ascidiella aspersa to a stage with thirty-two cells, and 

 finds it identical, cell for cell, with that of Clavelina as given by Van 

 Beneden et Julin. He however adopts, apparently without question, 

 the orientation given by them, and makes accordingly, in his early 

 stages the same error, calling dorsal (endodermal) that side of the 

 egg which is really ventral (ectodermal). 



He gives in pai't the cell lineage of the ectodermal side of a single 

 older stage, in which five additional divisions are represented as hav- 

 ing occurred in each of the equivalent halves of the egg. (See his 

 Fig. 20.) These agree exactly with the divisions I have found 

 occurring in Ciona subsequent to the 32-cell stage. 



Seeliger studied an undetermined species of Clavelina, the same 

 genus on which Van Beneden et Julin worked. It has been more 

 than once observed by writers on tunicate embryology, that, though 

 engaged with the study of forms so nearly related, these authors diflfer 

 widely in their conclusions. 



In discussing the 4-cell stage, for example, Van Beueden et Julin 

 state that the two larger cells give rise to the anterior portion of the 

 embryo, and in this I fully agree, whereas Seeliger contends that 

 they produce the posterior portion. Seeliger however gives little 

 evidence in support of his view, for his determination of the cell lin- 

 eage IS in most cases so manifestly a matter of mere conjecture that it 

 hardly merits a serious consideration. But it may be worth while to 

 point out where he has made the mistake that led to his false conclu- 

 sion. It is in passing from the 16-cell stage to the next succeeding 

 stage, the number of cells in which he did not take the pains to 

 determine. In doing this he has reversed the poles anterior and pos- 

 terior.. For his 16-cell stage (his Figs. 14 and 15), if its anterior 

 and posterior poles fe reversec?, corresponds unmistakably, both in the 

 characteristic arrangement of its cells and in their relative size, with 

 the 16-cell stage figured by Van Beneden et Julin (their Figs. 8 and 

 8a), and that figured by Chabry (his Figs. 17 and 24), and I can 

 add from my own observations that it also corresponds with that of 

 Ciona. 



