CYTOPLASMIC INCLUSIONS OF THE GERM-CELLS 275 
(apG), as it is they which ultimately form the acrosome, or 
head-cap of the sperm. 
In Pl. 12, fig. 5, the Golgi apparatus is seen to consist of a large 
number of semilunar platelets, rodlets, or dictyosomes (GE), 
which lie upon the outer surface of the archoplasm. By 
Mann-Kopsch technique the Golgi apparatus is not a reticulum, 
but is as drawn in Pl. 11, fig. 1 (ar), and Pl. 12, fig. 5. Examined 
after Cajal’s method, or by Da Fano’s modification of Cajal’s 
formalin-silver nitrate method, the Golgi apparatus is seen to 
be in the form of a reticulum, or of flat plates joined here and 
there, as shown in figs. 2 and 8 of Pl. 11. 
7. Pertiop Il. Maturation Divisions. 
The periods of division of the spermatocyte are difficult 
properly to study. In very little of our material were mitoses 
to be found, and this part of our work is the section about 
which we feel the most diffident to write. Meves, Niessing, 
and Moore all failed to follow the proacrosomic granules 
through the phases of the maturation divisions, and we have 
been unable to establish Papanicolaou and Stockard’s claim 
that these granules retain their individuality and become 
sorted out to the daughter cells during cell-division. Meves, 
Niessing, and Moore all agree that the proacrosomic granules 
soon become visible after the archoplasm is re-formed subse- 
quent to division—that is in the late telophase. We have 
adopted Papanicolaou and Stockard’s description for two 
reasons: firstly, it is extremely unlikely that the proacrosomic 
sranules would gradually accumulate and grow, especially 
before the first maturation division—only to become disin- 
tegrated at the mitotic prophase; and secondly, we are aware 
that the Golgi elements or dictyosomes hitherto had not been 
followed through division, but we now know that in mammals 
as well as invertebrates the Golgi elements may become 
sorted out during division and do not lose their individuality. 
In Pl. 12, fig. 6, we give a diagram illustrating the inter- 
pretations we at present consider to be the most likely to be 
correct : the mitochondria are spread haphazardly throughout 
NO. 258 U 
