288 
W. E. A(iA]{. 
bridge niiiy be :is wide as the rest of the climmosome. or :i, 
mere thread. 
(2) The frequent inequality of the segments into which the 
chromosome is divided. 
(8) Tlie fact tliat the metaphase split does not extend 
through the point of segmentation till after it is complete in 
the other parts of the chromosome. This leads to the assump- 
tion by the chromosomes of various shapes. In the case of 
long chromosomes it often leads to an appearance of two 
V’s connected by their apices by a thin strand (fig. 8,c, e, and 
fig. G, a,h,c). Later, of course, the longitudinal .split will 
extend through the connecting bridge also, and then an effect 
mav be produced as in n, fig. 8. More often, however, in 
the case of long chromosomes the extension of the metaphase 
split through the connecting bridge is accompanied by a partial 
flattening out of the transverse constriction. 
Another effect of the tardy splitting of the connecting 
bridge is the frequent occurrence of X-shaped chromo- 
somes, some of which are shown in fig. 8 {g, h, j, k) . 
Their mode of origin is obvious if compared with those chromo- 
somes in which the daughter halves of the two end-to end 
segments have not diverged so widely (fig. 8, e, and fig. 6, 
a, c). 
Spermatogonial MpI'OSES. 
In the newly formed spermatogonial equatoi'ial plates the 
chromosomes attain a great length, and correlated with this 
there is little, if any, indication of transverse segmentation. 
Such an equatorial plate is shown in fig. 9. The same absence 
f)f ti’ansverse segmentation, or even of sharp bends at the 
apices of the V’s, is shown in fig. 6 of my former paper. As 
the chromosoTiies shorten, however, transver.se segmentation 
becomes evident in many of them, and at the time that the 
metaphase splitting takes place it is often very pronounced, 
especially in the smaller plates. These are very difficult to 
analyse, as the confusion caused by the crowding together of 
the chromosomes fa.r more than counter-balances the advantage 
