672 • C. E. McCLUNG 



While therefore the unmodified rod lying in the equatorial 

 plate is rare, the form produced by bending in the plane of the 

 longitudinal cleft to form V's and U's is very common. In 

 Mecostethus, as may be seen by inspection of figures 32 to 34, 

 37, 41 to 44 and 46, practically all of the long, slender chromo- 

 somes are V- or U-shaped except such as extend across the plate, 

 in which case they are more or less straight, simple rods. It 

 is rare to find so much uniformity of shape as appears in this 

 genus, but its significance with regard to the movements of the 

 chromatids is great. Lateral views of the Mecostethus chromo- 

 somes at this time show that while the small ones are extended 

 in a plane of the spindle axis (fig. 39) as in other species, the larger 

 ones suffer an earlier separation of the chromatids and become 

 diamond shaped with an open center (fig. 35) imstead of linear 

 and solid as in most cases. This is merely the precocious appear- 

 ance of a chromatid separation which occurs in the early anaphase 

 of all species. 



Should the outer, divergent ends of the V-shaped chromosomes 

 become approximated (or should they remain united, if the re- 

 verse is the character of the movement), annular chromosomes 

 are produced. Much confusion regarding these structures exists 

 needlessly in the literature. The method of their formation is 

 clear and unmistakable and transitional forms are so common as to 

 point definitely to the relations existing between them and the 

 V- and U-shaped forms. In figure 2, for instance, the upper 

 chromosome is a ring but the outer ends are only slightly joined. 

 With just a little wider separation the form would be similar to 

 the next chromosome to the left, a little more space and it would 

 be shaped like the second one to the left, and with wider diver- 

 gence of the ends would approach the type of the one below and 

 to the right. In figure 2 there are at most two ringed-shaped 

 chromosomes, while in figure 3, from the same animal, there 

 appear six without divergent ends. It is evident that some of 

 the chromosomes which in figure 2 are V's are, in figure 3, rings. 

 Similar evidence is afforded by the chromosome groups of Tropi- 

 dolophus shown in figures 21 to 30, in which the annular chromo- 

 somes vary in number from three to seven. As a rule the larger 



