120 



infrequently one of the daughter nuclei starts to divide again before 

 the cytoplasmic constriction appears (Fig. 11). In this cell (Fig. 11) 

 it is impossible to determine whether the two nuclei have been produced 

 by amitosis or mitosis, but a study of mitosis has led the writer to 

 conclude that such cases are the result of the former method of 

 division. Figs. 13 — 16 represent a series of mitotically dividing cells. 

 Here it will be noted that by the time the mid-anaphase is reached 

 the cytoplasm has begun to constrict (Fig. 14), and when membranes 

 begin to form about the daughter nuclei the cytoplasmic division is 

 completed (Fig. 16). It is very probable, therefore, that such cases 

 as those shown in Figs. 9, 10 and 11 have been produced amitotically^ 

 and are simply in a stage more advanced than that of Fig. 7. 



Very often one portion of the nucleus stains more deeply than 

 the other, and when the division appears it separates these two 

 regions (Figs. 6, 19, 23). Dr. Child has described a similar condition 

 in Moniezia ^). He believes this ditference in staining probably indi- 

 cates that the two regions are functioning more or less independently. 

 There is, however^ one important ditference between the two forms. 

 In the pigeon I have never found a case in which the more deeply 

 staining portion contains a nucleolus. Since the nucleolus is a 

 karyosome, it seems very probable that the chromatin which is ordinarily 

 concentrated in that body is here scattered over a portion of the 

 nucleus. 



The Frequency of Mitosis and Amitosis. 



In order to determine the ratio of mitoses to amitoses during the 

 various stages of development the following method was adoped : In 

 one hundred sections taken through the central part of the blastoderm 

 every tenth section was selected, all its nuclei were studied and the 

 number of amitoses and mitoses noted. By studying a large number of 

 sections in a given egg I found that ten sections thus taken sufficed 

 to give approximately the correct ratio which exists between the two 

 methods of division. 



The data given in the following table are taken from series 

 representing distinct steps in the differentiation of the blastoderm. 

 The series, therefore, were not selected at regular intervals of time. 

 It might be well to add that in all probability the percentage of 

 amitoses, as given in the tables, is low as compared with that of 

 mitoses, because in counting, only the clearest cases of amitosis have 



1) C. M. Child, "Studies on the Relation between Amitosis and 

 Mitosis". Biol. Bull, Vol. 12, No. 2. 



