2o6 Edmund B. JFilson. 



close against the outer end of 3D (Figs. 9, 13) ; and while I 

 have not actually seen the division, it is nearly certain from the 

 position of the spindles that the division is unequal. The study 

 of a good many preparations of this stage leads me to believe that 

 this is a constant relation. 



The last cells to divide in the sixth cleavage are the macro- 

 meres, and of these 3D is the first. At the time of its division it 

 is only connected with the surface by a very narrow neck, as- 

 suming the extraordinary appearance shown in Fig. 14. The 

 result of this division is to form a large rounded cell, that lies 

 quite in the upper hemisphere (shown in Figs. 15, 16) and a 

 more superficial cell. From the conditions observed at a slightly 

 later stage I believe the former to be 4D, the latter the primary 

 somatoblast 4d or M ; but I am not entirely certain of this identi- 

 fication. Slightly later the remaining macromeres divide some- 

 what unequally, the cells in the meantime undergoing consider- 

 able shiftings and extending further up into the egg, so that it is 

 exceedingly difficult to identify them individually. The ectoblast- 

 cap has now extended far down towards the lower pole, so that 

 the macromeres are connected with the surface by narrow necks. 

 The cell I believe to be 4d now divides symmetrically into two 

 to form two large symmetrical cells lying between the entomeres 

 and the ectoblast (Figs. 15, 16), which correspond with the 

 mesoblast pole-cells as figured by Patten (<?. g., in his Figs. 27, 

 36). I have not positively traced these cells into the coelome- 

 soblast, but believe there can hardly be a doubt as to their na- 

 ture.^ At this period the large inner cell (identified as 4D) is 

 still undivided (Fig. 15) the primary trochoblasts have devel- 

 oped cilia, and the apical tuft is present (10-12 hours). 



Beyond this point I shall not for the present attempt to trace 

 the general cleavage, but will pass on to some points in the later 

 development. Patten has given figures of the trochophore of 



^Sections of the trochophores of 24 hours clearly show two large meso- 

 blastic pole-cells (one of which appears in Fig. 17) near the posterior end, 

 from which two mesoblast-bands extend forward as figured by Patten, c. g., 

 in his Fig. 50. 



