976 J. B. GATENBY AND J. H. WOODGER 
the cytoplasm, and they offer no remarkable behaviour for 
study. Around each mitotic aster are grouped approximately 
one-half of the Golgi elements or dictyosomes; for con- 
firmation of this phenomenon in cells other than those of 
the euinea-pig testis see Deinecka (8), Golgi (14), Murray (28), 
Perroncito (25), Fauré-Fremiet (9), and Gatenby (11, 18). 
This behaviour of the Golgi element or dictyosome does not 
entail any sort of division of the element itself, but only 
a haphazard, though subequal, sorting out of the whole 
elements between the daughter cells.4 
At apa in fig. 6 of Pl. 12 are the proacrosomic granules, 
which become scattered in the cytoplasm during division. 
As with the Golgi elements, the individual granules in the 
spermatocyte archoplasm are not themselves divided, but 
sorted out whole between the daughter cells. 
At cup is the chromatoid body whose fate in the maturation 
divisions has not been followed out ; one fact, however, may 
be mentioned, it is that by far the majority of spermatids 
contain a chromatoid body (Pl. 12, fig. 7, cus). In many 
animals the spermatocyte and spermatid contain a chromatoid 
body of some kind, and in the case of Smerinthus strong 
evidence has been accumulated which indicates that this 
body has the power of binary fission (10). 
8. Pertop III. The Newly-formed Spermatid. 
In Pl. 12, fig. 7, is a drawing of the newly-formed spermatid ; 
it contains the same categories of cytoplasmic elements as the 
spermatocyte, only they are approximately one-quarter in 
amount. With reference to the fact that the spermatid cell 
is generally much more than one-quarter the size of the 
spermatocyte, it may be pointed out that between the stages 
drawn in Pl. 12, figs. 5 and 7, there must be a period during 
which the cells are rapidly growing. While it is certain that 
the spermatid Golgi apparatus and archoplasm is usually 
1 Dictyokinesis in the maturation of the germ-cells of Mus, Cavia, 
Stenobothrus, Limnaea and Helix is the subject of a forthcoming paper 
by Ludford and Gatenby. The process is even more haphazard than 
depicted in fig. 6. 
