132 T. H. MORGAN. 



about the equator of the egg. It is evident, when this stage is 

 compared with the last one, that an extensive shifting of the cells 

 must have taken place. This is strikingly shown by super- 

 posing the outline of Fig. 22 upon that of Fig. 26. The next 

 stage, Fig. 27, shows important changes to have taken place in 

 the small cells around the sides of the segmentation cavity. 

 These have now pushed below the equator of the egg, and the 

 wall in this region has correspondingly diminished in thickness. 

 This last section is through an embryo that has just reached the 

 gastrulation stage, and although the section passes vertically 

 through the greater diameter of the egg, the blastopore lip lies 

 slightly to one side, and opens in neighboring sections that are 

 not quite so large. The mesoderm cells, containing more pig- 

 ment than the endoderm cells, form a distinct tongue extending 

 upwards from the level of the blastopore to the floor of the 

 archenteron. The general appearance of the section suggests 

 strongly that the innermost cells have remained stationary, or 

 have even pushed upwards a little, accompanying the general 

 upward movement of the floor of the archenteron. 



The later stages of the gastrulation-process in the toad, I have 

 not studied. King has recently given a series of figures showing 

 the later development. They show that the yolk-mass is thrown 

 high up into the blastocoel, especially at the anterior end, so that 

 the anterior part of the archenteron comes to lie higher up than 

 the original floor of the archenteron. It is not improbable that 

 the embryo of the toad lies higher up on the anterior side of the 

 egg than does the embryo of the frog, although the difference is 

 only one of degree. 



During the period of closure of the blastopore, the large 

 blastocoel becomes filled by the yolk and the drawing inwards of 

 the yolk brings the dorsal and ventral lips together without, in a 

 sense, their actually growing over the lower hemisphere. 



CONCLUSIONS. 



The large number of cells in the frog's egg at the time when 

 the embryo appears makes it impossible to trace the cell-lineage 

 after five or six divisions, so that we must be content with less 

 refined methods in locating the embryo-forming materials. 



