1904.] NATURAL SCIENCES OF PHILADELPHIA, 707 



bilateral divisions begin, in the arrangement of the cells." In other 

 words, the adult bilaterality is expressed in the arrangement of the 

 four cells la^-^-, Ib^-*-^ Ic*-*-' and Id*-*-^ brought about through spiral 

 cleavages, and in fact foreshadowed in the 8-cell stage. These facts 

 have all been shown to be true for all the forms in which a cross occurs, 

 and need not be dwelt upon here. 



While the anterior stem cells are dividing to form the anterior arms 

 of the cross, new spindles appear in the intermediate girdle cells la*-^-^ 

 and lc'-2-^ (figs. 25, 26 and 27). These cells then give rise at their 

 peripheral margins to tiny cells with a deeply staining nucleus (figs. 

 30, 31 and 32). The direction of these divisions is somewhat uncer- 

 tain, but judging by the position of the smaller products it is clexio- 

 tropic in la*-^-* and leiotropic in Ic^--*- Later Ib^-'-^ gives rise to a 

 similar cell (figs. 37 and 41). This division is unmistakably dexio- 

 tropie. It is of interest to note that in Amphitrite (]\Iead, 1897) and 

 Are7iicola (Child, 1900) these same cells should also be small and pro- 

 vided with a densely staining nucleus. In Thalassema (Torrcy, 1902, 

 1903) these cells are truly rudimentary, and it seems quite probable 

 that such is the case in Dinophilus also, judging from their minute- 

 ness and staining reactions. 



Closely following the division of Ib^-^- to form the right anterior 

 cross arm, indeed almost simultaneous with this division, is that of 

 the posterior basal cells of the cross. The plane of this division is 

 meridional, as is also the case in the annelid cross, but differs from the 

 latter in that the division is an asymmetrical one, since the left product 

 of the left posterior basal cell is very much smaller than the right 

 product, Avhile the products of the right posterior basal cell are equal 

 (figs. 30 and 37). This curious divergence from the annelid type I 

 am at present unable to explain. Possibly a further study of the 

 cleavage and of the history of the individual cells of the cross might 

 offer a solution of the problem. Closely following the divisions of the 

 posterior basal cells comes the division of the posterior terminal cells. 

 The spindles for this division are meridional, and the upper and central 

 products of this division are minute cells, which lie above their parent 

 cells as shown in figs. 37, 38 and 39. These correspond in origin to the 

 " nephroblasts," which in Nereis (Wilson, 1892) form the head kidney 

 (or possibly slime glands), and which in Amphitrite (Mead, 1897) give 

 rise to the large mucous glands of the umbrella of the trochophore. In 

 Podarke (Tread well, 1901) these cells are also minute and occupy 

 relatively the same position as in Dinophilus. Treadwell describes 

 small cells of the first quartette which sink through the ectoblast and, 



