HABITS AND LIFE HISTORY OF THE TOADFISH. 1 099 



contrary, even during the early stages of segmentation — 8, 16, and 32 cells — 

 the blastoderms flatten out, the cells on the periphery presenting an appear- 

 ance which can only be likened to that taken by a pellet of soft mud when 

 thrown against a smooth wall. To say that the blastoderms present an extraor- 

 dinarily irregular appearance, but feebly expresses the fact. In some cases it 

 looks as if the segmentation had begun at the center and worked outward, as 

 Watase (1891) has shown in the squid. The explanation of the whole matter 

 is to be found, I think, in the fact that segmentation begins before the flow 

 of the protoplasm into the germinal region has ceased. 



To the best of the writer's knowledge, the only work other than experi- 

 mental done heretofore on the segmentation of the egg of the toadfish is to be 

 found in Miss Clapp's paper published in 1891. Here she figures stages from 

 two to sixteen cells, inclusive, the later blastoderms being very irregular. 



Figure 3, plate cviii, shows the right valve of a Pinna shell containing eggs 

 in segmentation one-half natural size. This reduction was made in order that 

 this half of the nest might be shown in its entirety. The photograph portrays 

 a nest typical in all respects, the irregular manner in which the eggs are affixed, 

 the crowding together — some eggs being on top of others as cannon balls are 

 piled — and last, but by no means least, the fact that the eggs are in two stages. 

 For the most part they are in late segmentation, but on the upper right-hand 

 part of the shell some two dozen are still in early division, as the small size and 

 somewhat rounded outline of the germinal disk show. Figure 4, plate cviii, 

 shows another nest with eggs in segmentation intermediate between the two 

 stages shown in the previous figure. Here the eggs are shown twice their natural 

 size in order to bring out details more clearly. Recalling to the reader's atten- 

 tion the fact that the micropyle is practically always found opposite the point 

 of attachment of the egg, it is interesting to note that some of the blastoderms 

 are not "opposite the point where the ovum is attached" as Ryder (1887) 

 states, although this is generally true. Of especial interest is the fact that there 

 are in this nest some half dozen eggs with divided blastoderms. Should these 

 have come to maturity it is my opinion that they would have formed twins. 

 (For a much magnified figure of such a blastoderm see the writer's paper on the 

 pipefish, 1905, plate vi, figure 26, and for the histology of the same, plate viii, 

 figure 54, and plate ix, figure 55.) Miss Clapp, in the paper referred to in the 

 preceding paragraph, figures such a divided blastoderm in the 8-celled stage. 



Invagination. — The next stage in the history of the eggs whose development 

 we are studying is that in which, segmentation having ended, the blastoderm 

 flattens out into a thin cap of cells covering about one-fourth the upper half 

 of the egg. Shortly thereafter it is the rule in teleost eggs that the edges of this 

 protoplasmic cap begin to turn in under the cap and between it and the yolk. 

 Technically this is called invagination, the in-pushing cells making a distinct 



