196 



KINGSLEY AND CONN 



At no time after eight blastomeres were reached did the segmentation proceed regularly, 

 and with each succeeding segmentation the irregularity became more and more marked 

 until at last, at about that stage when the blastoderm should theoretically consist of sixty- 

 four cells, every trace of regularity is lost and each cell divides entirely independently of 

 its neighbors, the nuclei appearing in one just as they are disappearing in another, while 

 a third is at the same insjtant dividing. In some eggs this irregularity is noticed at an 

 earlier stage and is much more marked than in others, but in all it soon reaches such an 

 extent that in any case it is difficult to ascertain when the theoretical segmentation is 

 completed. 



We have now to describe the division of the blastoderm from eight to sixteen cells. 

 Nine minutes after the last mentioned reappearance of the nuclei, they again disap- 

 peared almost simultaneously, but one or two seconds intervening between the times of the 

 first and of the last. After this disappearance and before the division of the cells amoe- 

 boid movements, similar to those which have been described were witnessed. The cells 

 lost their regular outlines and their smooth contours and became lobulated and fur- 

 rowed. Fig. 17 before referred to represents two cells from the blastoderm. While this 

 amoeboid motion was in progress the segmentation furrows appeared. Fig. 18 will illustrate 

 this division better than it can be described, the two interior cells divided first, then the 

 other two, and lastly the four corner ones. The fissures in the second began before those 

 in the first were completed and those in three before the segmentation of the second 

 was accomplished. The complete segmentation occupied about two minutes. The planes of 

 division were in general times at right angles with those of the proceeding segmentation 

 and the result was a parallelogram with four cells on a side. The nuclei were again seen 

 four minutes after the segmentation was complete and remained in sight for ten minutes. 

 The amoeboid movements after this segmentation were very strongly marked and lasted 

 for considerable time, and the cells did not attain their smooth contours until about the 

 time when the nuclei vanished. The regularity of the parallelogram was far from being con- 

 stant as frequently one' or more cells would segment obliquely and the result would be 

 more like that shown in fig. 19. 



From this point onward the segmentation in every egg studied by us was very irregular, 

 and by various stages of 20, 21, 28, 29, and 30 cells the theoretical 32-celled blastoderm is 

 obtained and a short period of rest (which was not timed but which could not have 

 exceeded three minutes) intervened after which the segmentation proceeded but so irreg- 

 ularly as to be beyond description. Figs. 20 and 21 will illustrate some of the later stages. 



In the following table are given the results of timed observations of the segmentation 

 of several eggs. Several others were studied but without noting the intervals between 

 the stages. The dash ( — ) indicates that stage of development when the eggs were first 

 noted and the figures the number of minutes since the last timed stage. 



