l!H)7.J NATIKAL SCIIONCES OF I'lII LADKLI'HIA. 333 



their enclosed nuclei antl spheres are not ()\er<iro\vn l)y the ectonieres, 

 but they move out over the yolk at the margin of tiie ectodermal cap, 

 and this position they retain until the blastoderm has overgrown much 

 of the yolk (pi. XXVII, fig. 26). 



At a stage of about fifty cells, the macromcres .1, B, and C divide 

 in a la'otropic direction, giving rise to the remaining members of the 

 fourth quartet, 4.1.4/? and 4C(fig. 9). These cells are very large and 

 full of yolk, and for these reasons I have called them "secondary 

 macromeres." The}^ come tQ lie in the furrows 1)etwcen the macro- 

 meres, and in this position they are partly overgrown by the advancing 

 ectoderm and partly shoved before it. They serve as an excellent 

 means of orienting the egg and embryo, since they are found at the 

 right, left and anterior poles only, the posterior pole being occupied 

 by the derivatives of the much smaller mesentoblast cell, id. 



The secondary macromeres are purely endodermal in character. 

 Although their nuclei subdivide (pis. XXV, figs. 17, 18, et seq.), their 

 cell bodies usually remain undivided until the closure of the blastopore 

 (pi. XX^TII, fig. 31). Throughout all of this period the entomeres 

 consist of the four macromeres and of these three second ar,y macromeres. 

 Even as late as the gastrula stage, shown in fig. 31, these are the only 

 entoderm cells present except those derived from the cell 4d, which give 

 rise to the terminal portion of the intestine {In., figs. 29-31). 



In Crepidula a fifth cjuartet of cells, which are purely entodermal 

 in character, is formed from tlie macromeres about the time of the 

 closure of the blastopore; and at the same time the cells 4A, 4B and 

 4C divide. These divisions have not been observed in Fidgur, but, 

 with these exceptions, the origin, history and destiny of the ento- 

 meres of Fidgur are almost precisely like those of Crcpidida. The 

 four primary macromeres are similar in all regards, save only in bulk ; 

 the secondary macromeres arise at the same cell stage and l^ehave 

 in the same manner in these two genera. In Crepidula, as in Fidgur, 

 these seven cells, together with the entoblastic derivatives of tlie cell 

 4d, constitute for a long period the whole of the entoblast. 



With the stage represented by fig. 17, in which there are about 121 

 cells, the ectomeres become so numerous that their lineage cannot 

 readily be traced. At a slightly later stage, regions of the blastoderm 

 begin to differentiate into characteristic structures. We shall there- 

 fore consider this stage as the close of the cleavage and the beginning 

 of the period of organ formation. 



In conclusion we find that the cleavage of Fidgur resembles that of 

 Crepidula in almost every detail. Such slight differences as do exist, 



