132 GERM-CELL CYCLE IN ANIMALS 
rod, the other half do not. Of the forty-nine penul- 
timate spermatogonia examined, twenty-four ex- 
hibited a rod and twenty-five did not. This result 
has been confirmed by von Winiwarter. When the 
rod-containing penultimate spermatogonia divide, 
there is a similar segregation of the rod in one of 
the daughter cells, hence only one-fourth of the cells 
resulting from the divisions of the antepenultimate 
spermatogonia possess a rod. Of one hundred 
and forty-two cells of this generation studied by 
Montgomery, twenty-five were found with a rod and 
one hundred and seventeen without. That this 
ratio is less than one to three (1:3) is explained by 
the fact that some of the spermatogonia with rods 
may already have become Sertoli cells. The further 
history of the rod in the Sertoli cell is as follows: A 
primary rodlet is produced by a splitting of the rod 
(Fig. 41, C) after which the rod either disappears 
at once or else persists for a time, in which case it 
may split longitudinally as shown in Fig. 41, D. 
However, in four-fifths of the cells examined (one 
hundred in number) the large rod disappeared 
before the growth of the Sertoli cell had begun. 
Each primary rodlet splits longitudinally into two 
approximately equal parts, called secondary rodlets 
(Fig. 41, D, 72), which persist until the end of the 
cycle of the Sertoli cell. 
Neither Montgomery nor von Winiwarter were 
able to determine the origin of the rod. They do 
not consider it mitochondrial in nature, although 
it may arise from granules lying in the cytoplasm. 
