260 
Sidnev I. Kornhauser 
it is seen that the ends are iii contact at tbeir thin edges by thick f ihres, 
but that elsewhere they are separate ; thus an oval opening appears between 
the two ends, as seen in the ring occupying the upper part of Figure 66 
(Plate XIX). The opposite end, where the union is only temporary, shows 
no thickening, and the ends are united only by fine linin fibres. IVith 
the shortening of the cliromosome bands, these narrower ends separate 
froni each other, and the linin fibers are thus soon stretched apart and 
become invisible before the tetrad has beconie completely straightened 
out (Plate XIX, Figs. 67, 68, 70). The longitudinal division of each chromo- 
some band becomes more noticeable as the tetrads become less ragged 
and more deeply stained (Figs. 68, 70). The shorter chromosomes usually 
separate at one end in the strepsinema, and in condensing form ITs and 
V’s. These straighten out, thus giving rise to short tetrads, usually in 
the form of crosses (Plate XIX, Figs. 67—69). There are eight of these, 
which con’espond to the sixteen small autosomes of the sperniatogonia, 
and they form the eight small dumb-beU shaped tetrads of the first 
sperniatocyte (Plate XIX, Figs. 67—69)®). 
In addition to the macrochroniosome tetrad and the eight small 
tetrads, we have stiU another double cliromosome, the xy pah, which 
we have already followed to the strepsinema (Plate XIX, Fig. 65). The 
K-component was there represented as short and deeply stamed, the 
?/-component as longer and granulär. Between the stages represented bj' 
Figures 65 and 66, this pair could not be followed. AVhen, however, the 
tetrads became less scattered through the nncleus, it could be distinguished 
in three ways: 1. by its staining reaction, 2. by the dissiniilarity of its 
two component parts, and 3. by the method of union between the com- 
ponents. This pair stains more deeply than the ordmary tetrads before 
complete condensation, the a:-component being noticeably darker than 
the y (Plate XIX, Fig. 66). The ^-component is larger and less regulär 
in outline, and it often shows one or more light regions quite devoid of 
chromatin. Xeither shows a longitudinal division. The two are connected 
by a single, short, deeply staining Strand, which is probably referable 
to the intimate end-to-end union of this pair (Plate XIX, Figs. 54,58—61). 
A little later, the y-component also condenses and stains quite as deeply 
as the Ä-component (Plate XIX, Figs. 68, 70). The rod of chromatm con- 
necting the two components appears with great clearness; it is in this 
Figure 67 (Plate XIX) is diawn from a smear preparation luade without cliying, 
aiul fixed in fumes of Flemmixg’s fluid while still moist. There is in this case httle 
distortion of the chromosomes or cell body, as may be observed by a comparison with 
sections. 
