226 Edmund B. fnison. 



It is exceedingly difficult to rear these embryos, many of them 

 dying, while most of those that live break up into smaller masses. 

 A few larvae were nevertheless obtained, the best of which is 

 shown in Fig. 74. Like the others, this larva has evidently gas- 

 trulated (though the entoblast-mass is relatively small, perhaps 

 owing to the loss of some of the cells). Its most interesting fea- 

 tures are the presen-ce, at one end, of two cells bearing powerful 

 and active cilia by the activity of which the larva rotates irregu- 

 larly, while near the opposite end are two cells bearing much 

 smaller and feebler ones. It is evident that the two anterior cells 

 are secondary trochoblasts, undoubtedly those derived from the 

 second quartet; and the fact that there are two of these may be 

 taken as evidence that two trochoblasts are contributed to the 

 prototroch, at least in some of the quadrants, by the 2^-^ cells, as 

 is indicated by a study of the normal embryos. The two weakly 

 ciliated cells were to me at first a puzzle, since I failed to observe 

 anything corresponding to them in the normal embryos. But it 

 may be recalled that Patten describes and figures a ventral reg- 

 ion, covered with fine short cilia, just anterior to the pre-anal 

 sense-organ ('85, Figs. 47, 48, 57, etc.), and while I became 

 aware of this when it was too late to re-examine the normal larvae, 

 it seems very probable that it is these cells that appear in the 

 posterior ciliated tract of the >^-macromere larva. This is sus- 

 tained by the development of the yV basal cells about to be de- 

 scribed. 



2. Development of the i/i6-macromere. 



The tV basal cell, obtained by successive isolations, divides 

 unequally to form the third quartet cell (Fig. 75), which after- 

 wards divides into two, while still later the fourth quartet-cell is 

 produced from the basal (Fig. 76). Only two or three such 

 cases were obtained, one of which developed into the larva shown 

 in Fig. 77. While this larva could not be very clearly analyzed, 

 it evidently consisted of an internal mass of cells (entomesoblast 

 or entoblast) surrounded by a superficial layer of ectoblast-cells. 

 At one end the ectoblast-cells were larger and at least one of 



