44 THOS. H. MONTGOMERY jr., 



stages alone one might conclude that a ring had the value of more 

 than one chromosome; but a study of the subsequent stages shows 

 that it is typically a single chromosome, and that one or more of 

 the other chromosomes is destined to segment to furnish the necessary 

 number (7) of chromosomes. 



Thus there are in the dense spirem from 3 to 6 chromosomes; 

 these become thicker and shorter , and at the commencement of the 

 loose spirem one or more of them segment transversely until the 

 definitive number, 7, is produced, this being the number always found 

 in the following monaster stage. The 7 chromosomes of the monaster 

 are bivalent in value, and the reduction in their number must be 

 referred to the preceding synapsis. But, as we have seen, the number 

 of chromosomes at the latter stage and in the dense spirem is always 

 less than 7 (from 3—6). Hence the chromosomes appear in less than 

 half the normal number (14) in the postsynapsis and dense spirem, 

 so that it is necessary that one or more of those of the dense spirem 

 segment in order to produce the 7 definitive chromosomes. Until 

 this definitive number is formed, some of the long chromatin loops 

 necessarily are quadrivalent. But I nevertheless would apply the term 

 chromosome to these loops, for reasons which will be brought out in 

 the chapter "General considerations". 



It is remarkable how early the transverse constrictions of the 

 chromosomes appear. In the loose spirem stage it is already apparent 

 on the chromosomes, and becomes still more marked as the latter 

 shorten and thicken, that is, long before they are arranged in the 

 equatorial plate (Figs. 127 — 159). Frequently also in the loose spirem 

 are found quadrivalent chromosomes with a constriction about the 

 middle destined to divide it into definitive chromosomes, while at the 

 same time a constriction is seen around either half of the chromo- 

 some, the latter being the constrictions of the definitive chromosomes 

 (Figs. 126 — 128); since this constriction appears so early on the 

 chromosomes, while the nuclear membrane is still intact and before 

 the mantle fibres appear in the nucleus, we must regard it as the 

 result of automatic movements on the part of the chromosomes 

 themselves, i. e. as the result of automatic wandering and segregation 

 of the microsomes. Sometimes the constriction cuts in so deeply 

 before the nuclear membrane disappears, that the two chromatin 

 halves of the chromosome are united by only a bridge of linin ; and 

 occasionally a chromosome may divide into two before it reaches the 

 equator of the spindle {x, Fig. 162, PI, 4), which clearly proves that 



