MATURATION PHENOMENA OF GERM CELLS. 143 



divide in the second mitosis. Finally the definitive shape is 

 reached, as shown in Figs. 13 and 14. One peculiarity has often 

 been described in heterotypic chromosomes : as they are placed 

 in the spindle (Fig. 13), very frequently at the middle of each is 

 one thickening, and this is more frequent than two thickenings. 

 By comparison of a chromosome, such as that of Fig. 14, with 

 earlier conditions, such as those of Fig. 12, it becomes evident 

 that such a thickening corresponds to the separated ends of the 

 U-shaped loop, which have finally came into close juxtaposition. 



When we consider these points, we find two important facts : 

 (i) that the twenty-four chromosomes are regularly paired in the 

 spermatogonia, and that there the two of a pair lie close together ; 

 and (2) that there is no evidence that the space enclosed by any 

 one of the twelve heterotypic chromosomes has been formed by a 

 longitudinal splitting. In the spermatocytes there are twelve 

 loops of the shape of a U or V. There is a longitudinal splitting, 

 but along the long axis of each loop. The simple explanation 

 of these facts is that in the spermatocytes, in the synapsis stage, 

 there takes place the close conjugation of every two such chro- 

 mosomes as were found in the spermatogonia ; that two together 

 constitute a U-shaped loop ; and that therefore the first matu- 

 ration division results in separating entire univalent chro- 

 mosomes. 



The difference of opinion between Janssens and Dumez and 

 myself is more one of interpretation than of observation, though 

 they did not notice the pairing of the chromosomes in the sper- 

 matogonia, nor yet the true longitudinal split. They frankly ad- 

 mit that by their interpretation the reduction in number of the 

 chromosomes remains a mystery. They give no explanation of 

 why there should be regularly disposed U-shaped loops. In as- 

 suming that a heterotypic chromosome has been formed by a 

 longitudinal splitting, instead of by a junction of two univalent 

 chromosomes, they contend for a kind of splitting very wide at 

 the middle of the chromosome, but narrow at its ends ; yet no 

 such longitudinal splitting is known in any other case, and its 

 difference is brought out sharply by comparison with the un- 

 doubted longitudinal splitting in the chromosomes of the later 

 prophases of the spermatogonia. These heterotypic chromosomes 



