Papers from the Marine Biological Laboratory at Tortugas. 221 



The occurrence of variations such as these seems to indicate that the 

 processes concerned with the nucleus have great independence from those 

 concerned with the centrosome and vice versa greater than is known to 

 exist in both normal and abnormal mitosis. In abortive mitoses, such as 

 the so-called oligopyrene spermatocytic divisions of Paludina, there is 

 always maintained a constant time correlation between the phenomena 

 which occur in the nucleus and those which occur in the centrosome. An 

 interpretation of the processes described above as being an abortive mitosis 

 of the apyrene spermatoblast of S trombus seems to be unwarranted by the 

 facts in the case. 



GROWTH AND DIFFERENTIATION OF THE APYRENE SPERMATOSOME. 



At first there is very little, if any, growth of the spermatosome; whatever 

 growth there is, takes place slowly and for a long time but little difference 

 in size exists between the spermatosome and the spermatoblast previous to 

 the breaking down of the nucleus. The changes that occur are concerned 

 chiefly with the chromatic vesicles and the centrioles. 



By the time the centrioles have divided, the vesiculation of the karyo- 

 merites has been completed, or nearly so (figs, 33 and 34). The chromatic 

 vesicles now acquire a very definite position in the cell. They lie in the 

 outer portion of the cell, not far beneath the ceH-membrane, and are con- 

 fined almost entirely to the distal half of the cell. In a section passing 

 through the chief axis of the cell, the chromatic vesicles are seen to lie in a 

 semicircle, the open side of which points towards the base of the cell. This 

 arrangement is shown particularly well in figure 39. At about this time, 

 too, there is a very noticeable swelling of the chromatic vesicles due to the 

 gradual increase in the amount of nuclear sap within them (fig, 41), In 

 many instances an achromatic network is also visible in them (figs. 36, 39, 

 40, and 41). In a word, the chromatic vesicles now present the appearance 

 of so many small but active secondary nuclei whose reconstruction is not 

 yet completed; and such, indeed, they are considered to be by the writer. 



It will be remembered that the secondary rays of the centrioles, at the 

 time of the division of the latter, pass through the ring of mitochondria and 

 then, diverging, they reach to the furthest parts of the cell. There is very 

 little evidence in support of any belief that they attach themselves to the 

 different chromatic vesicles; on the contrary, the ends of many of them 

 can be clearly seen to pass on beyond the vesicles and lie free in the cyto- 

 plasm (figs. 33, 34, 35, 36, and 39). After the division of the centrioles, 

 the connection of the rays with the distal halves of the former, or the distal 

 centrioles as they will now be designated, becomes less distinct. In front 

 of the advancing distal centrioles there is formed a dense and homogeneous 

 substance which very frequently obscures those portions of the rays which 

 pass through it (figs. 34 and 35). Into this substance, too, from the distal 

 centrioles, there grow out slender fibrillar processes which become more 

 pronounced as the centrioles advance (figs. 34, 35, 36, and 39). These for a 

 time are probably continuous with the rays, but as the latter become less 



