112 George Lefevre 



body, which is often abnormally large, is occasionally fertilized 

 by a spermatozoon and develops into a small gastrula. 



Although in the great majority of experiments normal polar 

 bodies were extruded by the unfertilized eggs, this was not always 

 true, for in a number of cases the eggs divided and eventually 

 gave rise to trochophores without any external indication of a 

 previous maturation, or after the formation of but a single polar 

 body. It occasionally happened that eggs in the same dish would 

 show all of these conditions, i. e., some would mature normally, 

 some would extrude only one polar body, while others would form 

 none; yet all of these classes of eggs might undergo development 

 and produce larvae indistinguishable from each other. It was 

 more often the case, however, that in any one dish, in which the 

 eggs had been exposed to the same solution, all the developing 

 eggs would either maturate normally, or else throw off only one 

 polar body, or again none at all. As the egg of Thalassema is 

 quite opaque, the internal phenomena involved in these changes 

 can only be examined in sections, and their description must, 

 therefore, be reserved for the portion of the paper dealing with 

 observations on the preserved material. 



3 Cleavage 



After the formation of the polar bodies, there are no further 

 signs of change visible in the living egg for some time. The first 

 cleavage does not take place at a definite period after the eggs have 

 been subjected to the action of the solutions, but it may appear 

 at any time, varying from 2 to 3^ hours, the shorter interval, how- 

 ever, being the more frequent one. The appearance of the first 

 cleavage is not correlated with the time of extrusion of the polar 

 bodies, and a delayed maturation does not necessarily mean a 

 corresponding postponement of segmentation. As the first cleav- 

 age occurs after normal fertilization in about 50 or 60 minutes 

 from the time the spermatozoon enters, it is seen that the activi- 

 ties which lead to segmentation are called forth much more slowly 

 in the parthenogenetic eggs. 



The early cleavages are closely similar to the normal in a great 

 many cases and in favorable experiments where the optimum 



