130 
Thos. H. ilontgomery, jr. 
much smaller than any of the others and is tlie smallcr diplosome {d, 
figs. 8, 16, 33, PI. IX; 38, PL X). The larger diplosome is either the chro- 
raosome next in size, as I held, or one of the still larger ones, as Wilsox 
maintained, but it is difficult to be certain of this larger diplosome at 
this stage. There are then 6 pairs of autosomes or ordinary chromosomes, 
the two of each pair of equal size, and one pam of diplosomes or modified 
chromosomes of very imequal size. All these di^'ide in the spermatogonial 
mitoses. ln the growth period, which was not studied by Wilson, I found 
(1906) : „In the synapsis the 12 autosomes con jugate to form 6 bivalent ones. 
The diplosomes also alwaysunite then endtoend”, andthelatter arereadily 
recogiiizable in the growth period by maintainiug tlieh- compact structure 
and d“nsity. The diplosomes separate from each other just before the 
first maturation division, so that in the spindle (figs. 2, 10. 11, 18, 30, 
40, 41, 57, 58) there are 6 bivalent autosomes and 2 univalent diplo- 
somes, a total of 8 clements (1 more than half the number in the spermato- 
gonia). ln the first maturation mitosis the 6 autosomes dmde reduction- 
ally, and the 2 diplosomes equationally. Each second spermatocyte recei- 
ves, accordingly, 8 elements (fig. 21, PI. IX), and in each before its division 
the unequal diplosomes conjugate to compose a bivalent {d, e, figs. 4, 22, 
PI. IX; fig. 52, PI. X) — this important phenomenondiscovered by Wilson. 
ln the second maturation mitosis each autosome divides equationally, but 
the bivalent diplosome reductionally (<Z, e, figs. 5, 12, 23, PL IX; fig. 61, 
PL X), consequently each spermatid receives 6 autosomes and 1 of the 2 
diplosomes (a total of 7 elements, just half the spermatogonial number). 
Such is in brief the normal course of the spermatogenesis, and my 
reuewed study of the past summer has confmned it. We will find that 
this occurs in the greater jiercentage of the cells. But there are depar- 
tures from it that need explanation. For this purpose I have counted the 
chromosomes in the equatorial plates of the spermatogonia and fü'st and 
second spermatocytes in six testes from different individuals. The cluomo- 
somes of the spermatids are often so densely crowded that accurate counts 
of them are much less easy to make, hence these are included in only a 
couple of instances. The phenomeua will be described for each testis sepa- 
rately, so as to determine the Variation in cells of one and the sanie indi- 
vidual. So far as 1 recall such large statistics have never been presented 
before on corresponding cells of the same spermatogenesis; for AVilson’s 
study of Metapodiiis^) considered differences not in the same individual 
but in different individuals of the same species. 
I) Studies on Chromosomes III. Joum. Exper. Zool. 3. 1909. 
