The Development of Ischnochiton. 583 



are probably do larger than either of the first two. All coutain yolk 

 in about the same relative quantity, and for this reason the free 

 surfaces of the cells always stand out round and full, in marked con- 

 trast to the ectodermal cells of this same stage in embryos in which 

 the polar differentiation is strongly marked. 



4. Division of Primary Tr ochobl asts; Formation of 



Accessory Primary Troclioblasts; First Cleavage of 



the Second Quartette, 36 cells (Figs. 13, 14, 15). 



The primary trochoblasts were formed at practically the same 

 time as the second quartette and in this cleavage both sets of cells 

 divide at almost the same time. The former segments dexiotropically, 

 the two resulting cells occupying the same position as did the parent 

 cell (Figs. 13, 15). The second quartette also divides dexiotropically 

 into a lower larger and upper smaller cell, which group also retains 

 the original position of the mother cell, though at times a slight 

 shifting to the right occurs in the upper product. 



About this time the parent cells of the trochoblasts and the 

 apical series (Figs. 14, 15) are seen to be in the process of division. 

 For some time I looked upon them as the cells corresponding to the 

 basal cells of the Molluscan cross {1 a ^ ^, etc. of Crepidula) and hence 

 considered that this division would result, as in Crepidula, in the 

 formation of the median and basal cross cells proper, but a more 

 careful study has shown this to be incorrect for the cells that 

 form at this division are true trochoblasts, the second 

 set formed from the first quartette. These cells have formed 

 by a leiotropic division and lie closely wedged in between the parent 

 cell and the two trochoblasts of each quadrant. The first quartette 

 of ectomeres therefore gives rise to the primary trochoblasts, and 

 this third division produces cells which I shall term the accessory 

 trochoblasts. Strictly speaking these accessory cells are just as much 

 to be considered primary as are those produced at the first division 

 of the first quartette , but I shall let this be tacitly understood and 

 therefore speak of them simply as accessory trochoblasts. Also I 

 have used tlie term accessory in view of the fact that these cells form 

 but half the number of cells produced by the primary trochoblasts, 

 that is to say, the primary trochoblasts give rise to four cells in each 

 quadrant while the accessory form two. These latter cells appear from 

 the start similar to the primary trochoblasts, and throughout their 

 development they are alike in all essential regards. 



ZooL Jahrb. XJI. Abth. t. Morph. 38 



