A Comparative Study of tlic Chromosomes etc. 
261 
iorni that they enter the first spermatocyte division. The a;-component 
is now of practically the same size as the y-component. 
B. Enehenopa ourvata. 
The history of the autosomes in the growth of the sperniatocytes of 
E. curvata presents nothing fundamentally different from that of E. - 
hinotata. There are, however, several points of interest, which are worthy 
of consideration: first, the disposition of the chromatin granules in the 
zygonema and strepsinema; secondly, the formation of the macrochromo- 
sonie tetrad; and, thirdly, the behavior of the a;-chromosome and its 
relation to the autosomes and chromatic nucleoli. 
The beginning of the growth period in E. curvata is niarked by the 
gradual condensation of the chroniatin along definite lines of the very 
fine network into which the spermatogonial chromosomes of the last 
generations have resolved themselves (Plate XX, Fig. 71). When the 
leptotene tkreads first appear, they preserve in their angular courses 
an indication of their origin from a chromatic network. They are loose 
in texture, and connected one with another by fine linin fibers (Plate XX, 
Fig. 72). In preparation for syndesis, the threads lose their angularity, 
and also become finer and more distinct (Plate XX, Fig. 73). Here, 
also, it is possible to make only approximate counts of the leptotene 
threads, on account of their tortuous condition. I am convinced, however, 
that theh’ number is nearer that of the spermatogonial chromosomes than 
of the spermatocyte or reduced number. 
The parallel union of the thi’eads is shown in E. curvata even more 
clearly than in E. linotata, and often in a nucleus containing many lep- 
totene threads, one or two zygotene threads are found which are fre- 
quently parallel in pairs either along their whole length or part of their 
length (Plate XX, Figs. 76, 77). A polar arrangement of the leptotene 
threads was not found in E. curvata. Figures 74, 75, and 77 (Plate XX) 
Show the first zygotene threads forming in nuclei in which most of the 
chromosomes are in the leptotene stage. It is not unusual to find a pair 
of long threads united at both ends, but not in dose contact along their 
entire lengths (Plate XX, Figs. 74, 75, 78). 
When the double chromosome threads have been formed, they assume 
a definite polar arrangement. The ends of all the loops usuaUy point 
toward one side of the nucleus; but occasionally exceptions were found 
in which a loop was inverted. A comparison of the chromatin granules 
on either side of the interchromosomal cleft of each double loop shows 
that very often the granules do not coincide either in size or position. 
