626 HAROLD HEATH, 



Fig. 32 the spindles are not perfectly formed and serve to show in 

 what direction the centrosomes moved. 



At a stage represented in Fig. 39 the next division of the parent 

 cells occurs. In the direction of the spindles and the size of the 

 resulting cells this cleavage is a repetition of the preceding. I have 

 been unable to determine from surface views the position of the 

 daughter cells but one wax reconstruction showed them to be located 

 between the parent and the first formed cells. Thus two raesoblastic 

 bands exist but they are in close contact with one another and con- 

 sequently do not diverge as in some Annelids and Molluscs. 



Beyond this point I have been unable to follow the development 

 of this germ layer. Invagination has proceeded to such an extent 

 that the cells are no longer visible from the exterior and the lack of 

 clearly defined cell boundaries and nuclei renders it impossible to trace 

 the development from surface preparations. 



For a comparatively long period the large pole cells of the meso- 

 derm may be seen in section occupying a position in the wall of the 

 archenteron, but it is impossible to determine whether they contribute 

 to the formation of the alimentary canal. In other words it is not 

 known whether 4^ is "entomesoblast" as described by Patten ('86), 

 Stauffacher ('93) or Conklin ('97). 



Later stages show the mesoblast to form a continuous sheet of 

 cells extending from the posterior border of the blastopore along the 

 dorsal surface of the embryo. This gradually extends laterally and 

 anteriorly. Several isolated cells occur at various points but whether 

 they have been derived from the primary mesoblast or as larval 

 mesoblast 1 cannot state, though I hope a more extended study of 

 sections may yield a definite answer. 



X. History of the Entomeres. 



1. The Fourth Quartette. 



The history of 4d (mesoblast) has already been considered and 

 it now remains to trace briefly the history of the remaining cells of 

 this generation, all of which enter into the formation of the mesenteron. 

 The origin of these cells and their behavior in the process of in- 

 vagination has already been considered. 



The only division of this quartette which I have observed occurs 

 simultaneously in all the cells (Fig. 40). The spindle in 4 & is per- 

 fectly meridional with its inner end almost in line with the animal 



