402 KING. [VoL. XIX. 
which did not go into the spireme and they are always pro- 
duced by the resolution of compound-nucleoli. Oxychromatin 
filaments similar to those shown in Figs. 40-43 are figured in 
Bouin’s work on the oogenesis of Rana. Bouin considers that 
these filaments are a part of the general chromatin of the egg, 
and he does not distinguish them from the true chromosomes. 
By means of a series of camera drawings of all of the sections 
of nuclei in about the stage of development shown in Fig. 43, 
I have endeavored to ascertain the number of oxychromatin 
filaments and of chromosomes at this time. While the chromo-_ 
somes appear to be twenty-four in every case, the number of 
oxychromatin threads seems to vary from 20-50 in different 
nuclei. This difference in the number of oxychromatin threads 
in various cases can doubtless be attributed to the fact that 
the compound-nucleoli from which the filaments are derived 
vary in number and in size in different nuclei and that these 
bodies do not all resolve at the same time. 
Carnoy and Lebrun distinguish three distinct stages in the 
development of amphibian oocytes, and they state that there 
are many generations of nucleoli which alternate with various 
kinds of chromatin figures; the nucleus frequently containing 
one kind of structure exclusive of the other. In Bufo I have 
never found an oocyte in a stage of development between that 
shown in Fig. 38 and that of Fig. 50 in which the iucleus 
did not contain nucleoli, chromosomes, and oxychromatin fila- 
ments provided the egg had been satisfactorily preserved and 
stained. As the ova grow the number of nucleoli increases; 
but the number of chromosomes remains constant, and the 
maximum number of oxychromatin filaments is found at the 
stage of Figs. 40-43. After this time the oxychromatin 
threads stain more faintly; the granules of which they are 
composed gradually draw apart (Fig. 48), and finally become 
scattered throughout the nucleus. Many of these minute 
chromatin granules can still be found in the nucleus at the 
beginning of the maturation period. 
Although in Bufo there is no periodic resolution of nucle- 
oli into chromatin threads followed by the development of a 
