No. 2.] THE OOGENESIS OF BUFO LENTIGINOSUS. 377 
cytoplasm of the germ-cells, usually close to the nucleus, a 
small centrosome which is surrounded by a rounded, granular 
attraction-sphere (Fig. 8, C). This centrosome divides very 
early in preparation for the cell mitosis, and, as shown in Fig. 
7, it is sometimes possible to find a section of a cell which con- 
tains two centrosomes as well as two vitelline bodies. Such 
a section shows conclusively that the vitelline body is not de- 
rived from the centrosome and that there is no relation be- 
tween these bodies. I have, as yet, no clue to the origin of 
the vitelline body which, as will be shown later, is undoubt- 
edly concerned in the formation of yolk nuclei. A structure 
similar to the vitelline body is found in the cytoplasm of the 
spermatogonia of Bufo, and it can be traced directly to the 
spermatids where it gives rise to the acrosome of the mature 
spermatozoon. Since Meves (69), McGregor (63), and Bro- 
man (13) have found that the acrosome of the amphibian sper- 
matozoon is derived from the idiozome, I suggested in a 
previous paper (King, 52) that the body in Bufo which forms 
the acrosome might possibly be derived “from a condensation 
of a portion of the attraction-sphere at an early period in the 
history of the primary spermatogonia.” My study of the 
primordial germ-cells has not given any support to this hy- 
pothesis since, although this body is usually found near the 
attraction-sphere (Fig. 7), the two structures are clearly dis- 
tinct at all times and there is not the slightest evidence that 
the former is derived from the latter. In its size and general 
appearance the vitelline body closely resembles the small 
nucleoli in the nuclei of the primordial germ-cells, but I have 
seen nothing that would indicate that it is of nucleolar origin. 
The later history of this structure in the ova strongly suggests 
that it is a secretion product of the cytoplasm formed, pos- 
sibly under the influence of the nucleus, but not from nuclear 
material. 
According to the investigations of Bouin, the increase in the 
number of germ-cells in Rana is brought about through a con- 
tinuous process of transformation of peritoneal and mesen- 
chyme cells into sex-cells, not by mitosis nor by direct division 
