BIGELOW : NUCLEAR CYCLE OF GONIONEMUS MUEBACIIII. 335 



good reason to believe that such cells never attain functional maturity, 

 but that they sooner or later degenerate. It is interesting to note in 

 this connection that some individuals appear to produce them in com- 

 parative abundance, while others do so but seldom, if at all. 



In addition to the giant spermatozoa just described, there are to be 

 seen in all preparations of the adult spermary of Gonionemus a consid- 

 erable number of multiple spermatozoa, of which two specimens are 

 represented on Plate 5 (Figs. 107, 108). These exhibit all possible con- 

 ditions of partial separation, but agree in this that the nuclei have divided 

 and separated more or less completely, while the cell bodies have failed 

 to divide, thus differing from the giant spermatid, in which both nucleus 

 and cell body have failed to divide. Combinations of the two conditions 

 may, however, occur, as is shown by the specimen represented in Figure 

 108. 



A typical example is shown in Figure 107. Here there is a single 

 cell body enclosing two perfectly formed nuclei, one of which exhibits 

 at its apex a typical acrosome. Although the nuclei are apparently 

 normal in structure, they invariably exhibit a much more elongate form 

 than is the case with normal spermatozoa, a condition due in all proba- 

 bility to their being crowded together within a single cell body. Two 

 distal centrosomes are visible at the periphery of the cell, and extending 

 inward from each of them is a well-defined axial filament, but a proximal 

 centrosome is to be seen in the case only of the nucleus without an acro- 

 some. From each of the distal centrosomes there arises a tail filament 

 of entirely normal appearance and length. The example shown in Figure 

 108 is an interesting modification of this condition, because while the 

 smaller of the two nuclei presents the appearance just described, the 

 other, which is nearly twice as large, is double. That it is really double 

 is shown not only by its large size, but also by the existence of two 

 tails and distal centrosomes. In all the multiple spermatozoa the archo- 

 plasmic organ is moi'phologically a single structure, although it may 

 exhibit two or more staining centres, — a fact of some significance in 

 connection with the question of the origin of the tail filament. 



C. Oogenesis. 



In the oogenesis of Gonionemus I have been able to trace the history 

 of the germ nuclei from the prophase of the last oogonial division until 

 the germinative vesicle comes to lie at the surface of the nearly ripe 



