22 2 Edmund B. JVilson. 



of various sizes that are obviously isolated cells of the general 

 pras-trochal ectoblast. 



The relatively large size of the apical cells, whether completely 

 isolated or forming part of the ]'i or ^o larvae, is a matter I 

 have not yet fully cleared up. The apical pole of the normal 

 larva of 24 hours (Fig. 18), seen in surface view, gives some- 

 what varying appearances, but clearly shows a central group of 

 four to six larger cells. This does not exactly agree with what 

 appears in the partial larvae, whiqh, as a rule, show two large 

 apical cells, apparently of nearly equal size; it is, however, dif- 

 ficult to determine the exact size of the cells in the normal larva, 

 owing to their crowding together. It seems probable that this ap- 

 parent discrepancy may be due to the fact that, as stated above, 

 the primary rosette cell is so frequently too large in the }i and 

 tV embryos, and that this results in a slightly abnormal later 

 development of its products. It is clear, however, that this does 

 not affect in any essential way the differentiation of the apical 

 cells. 



5. Development of the isolated entire first quartet. 



In the sea-urchin Driesch (1902) has shown that both the 

 upper and the lower quartets of the 8-cell stage may produce 

 complete dwarf larvae, though the two quartets show certain char- 

 acteristic differences in development, proving that they are not 

 identical. In Patella it is difficult to perform this operation, and 

 still more difficult to rear the larvae, since the cells always separ- 

 ate more or less after replacement in normal water. I neverthe- 

 less succeeded in obtaining a few cases. The cleavage of the iso- 

 lated first quartet is essentially the same as in a whole embryo. 

 The first division, leiotropic in all the cells, produces four upper 

 cells (i^) and the four trochoblasts (i") alternating with them, 

 the eight cells forming a nearly flat plate (Fig. 65). The sec- 

 ond division, dexiotropic in all the cells, produces a plate of 16 

 cells (Fig. 66) in the centre of which is the rosette, around which 

 lie the four i^-^ cells and, at the margin, four groups each of two 

 trochoblasts (i^S i'"). As shown in the figures, the form and 



