ON THE DEVELOPMENT OF LINGULA ANATINA. 17 



upon the cleavage-plane, I re-examined the preserved eggs and 

 found that the things above mentioned held true in them. 



As the cells continue to increase in number there results a 

 typical cœloblastula (PI. I., Fig. 10. ; PI. IL, Figs. 31, 32.). 

 In this a spacious segmentation cavity can be seen upon applying 

 a slight présure to the cover glass (PI. I., Fig. 11.). All the 

 cells are not of equal height ; those at one pole are a little higher 

 than the remaining ones (PI. I., Fig. 11.; PI. IL, Figs. 31, 32.). 

 The cells are prism-shaped, pentagonal or hexagonal in section 

 (PI. IL, Fig. 31.) and are somewhat vaulted on the outer as well 

 as on the iuner surfaces. Histological differentiation has not as 

 yet taken place ; all the cells show the same structure. As alread} 7 

 stated the dissolution of yolk-granules proceeds from outside 

 inward in every cell. This can best be seen on comparing the 

 thickness of the yolk layer in PL IL, Figs. 29-32. During early 

 stages the nuclei are found lying in the yolk layer (PI. IL, Figs. 

 29, 30.), but later they are found outside the yolk layer, as the 

 latter is reduced as in PI. IL, Fig. 32. 



Of abnormal development I found many interesting cases. 

 One or two blastomeres at the 2-, 3-, or 4-cell stages were 

 observed to continue normal cleavage processes after other blasto- 

 meres had degenerated and gave rise to dwarf blastula?, gastrulse 

 or even embryos. This must have been caused by some external 

 stimuli such, for instance, as an increase of salinity of the 

 water by evaporation, or want of oxygen. Experiments similar 

 to those recently made on the eggs of the sea-urchin, Amphioxus, 

 hydromedusee, etc. were thus accomplished in our case without 

 carrying them out on purpose. At degeneration of the eggs or 

 blastomeres they, in a majority of cases, segment off into many 



