OBSERVATIONS ON THE MATURATION PROCESSES. 31 



"equatorial plate," however, they become crowded, and the crowding 

 increases as division approaches. Figs. 15 and 16 illustrate exception- 

 ally favorable cases, in which the number can be determined satisfac- 

 torily, at least at one end of the spindle. It rarely happens that a spindle 

 lies wholly in one section; it is usually cut into two or three parts. This 

 is frequently of advantage. (See figs. 7 and 70, 8a and Sb, loa and lob, 

 etc.) When the axis of the spindle is parallel to the plane of cutting, 

 the chromosomes, which are hardened by the process of preservation, 

 are seldom cut by the knife, but are pushed to one side. Sometimes 

 they are dragged out of place (figs. i2,x, 136, x and x') , or even completely 

 out of the spindle into the cytoplasm (fig. 12), where they lie at the 

 surface of the section on the side of the spindle toward which the 

 knife moved. In the spindle shown in fig. 12 the chromosomes (not all 

 of which are drawn) number 20, including the one lying to the left of 

 the spindle. This fact, the displacement of chromosomes, doubtless 

 accounts for some of the cases where there seem to be fewer than 20. 

 In the spindle shown in figs. 130, and 136, for example, where there are 

 only 1 8, displacement is clearly shown in two chromosomes (x and #') 

 lying at the upper surface of the lower section (136); and it is quite 

 possible that others have been completely removed. 



3. ACHROMATIN PARTS OF FIRST MATURATION SPINDLE. 



The origin of the spindle has been described under Stage II. At 

 first broadly elliptical, it changes its form, becoming slightly sharper 

 at the poles and, on the average, longer and narrower, especially in the 

 later stages, as division approaches. The fibers are not limited to the 

 surface of the spindle, nor to any part of it, but are uniformly distrib- 

 uted, as can be seen in cross-sections of the spindle. They do not con- 

 verge as straight lines to a point, but curve inward toward the poles, 

 without, however, meeting (figs. 8, 9, n left end, 12, i$a, 136, 140). 

 Consequently they are never parallel, and the spindle poles are more 

 or less open. However, in two otherwise apparently normal spindles 

 (figs. 10, n) the fibers at one pole do meet at a point, from which there 

 are a few radiations extending into the surrounding cytoplasm. 



Besides the change in proportions, there is also, on the average, a 

 small increase in volume. At Stages III, IVa, and IV6 the average 

 dimensions are, respectively, in micra, 18.7 X 10.4, 19.2 X 10.8, and 

 22.4 X 9.9. The variations in size in each stage are considerable (see 

 table 2, p. 14). With metakinesis the spindles elongate considerably 

 and become narrower. Three such spindles, parallel or nearly parallel 

 to the surface of the egg (fig. 15), give as an average a length of 26 micra 

 and a diameter of 8 micra; another, almost exactly radial in position, 

 gives the corresponding measurements of 23 X n micra. 



As the spindle develops, the fibers, at first in the young spindle 

 evident only as feeble fibrillations, become more distinct. They are 



