62 THE MATURATION OF THE EGG OF THE MOUSE. 



probably relates to those of the second spindle. Sobotta (1895, p. 46; 

 1907, p. 522) and Gerlach (1906, p. 14) state that the division is trans- 

 verse, but for theoretical reasons they believe that the division of the 

 chromosome of either the first or second spindle must be longitudinal. 

 Sobotta (1908, fig. 7) alone figures a dividing second spindle. His 

 ''biscuit" shaped chromosomes remind one very much of some of the 

 dyads we have described (p. 37) as constituted of 4 parts, inasmuch as 

 the "biscuit" forms are in some instances in groups of 4. What he calls 

 a whole chromosome looks more like half of a dyad. 



The same criticisms which have been made regarding the number 

 of chromosomes of the first spindle apply also to those of the second. It 

 was in polar views (the most favorable for counting) that Tafani found 20. 



ACHROMATIN. 



In his paper of 1895 (p. 45) Sobotta stated that the spindle fibers 

 of the single spindle (which occurred in nine-tenths of the eggs) were 

 derived in part from the achromatic portion of the germinative vesicle. 

 As already pointed out, he no longer holds this view. 



The second spindle as drawn by Sobotta (1895, 1907) is barrel-shaped, 

 the ends being somewhat truncate, the fibers only slightly curved, and the 

 poles open. As illustrated by Gerlach, Kirkham, and the present writers, 

 this spindle is elliptical, with fibers incurving at the poles. 



The flattening of some of the second spindles described on page 38 

 is apparently a result of their lying close to the surface of the egg. There 

 is a possibility that the flattening is caused by shrinkage due to fixing 

 and dehydrating. Shrinkage to produce this result would have to be 

 greater in a radial than in other directions, and could be explained only 

 on the supposition that the substance in which the spindle lies, being 

 probably more fluid than the surrounding cytoplasm, is extracted more 

 rapidly on the side nearest to the surface of the egg. However, were the 

 flattening due to shrinkage the chromosomes should be crowded in a 

 radial direction ; but that this crowding does not exist is clear from plate 

 4, fig. 20, in which the spaces between the chromosomes are as uniform 

 as in fig. 21. 



All investigators agree that the second spindle is smaller than the 

 first. Sobotta (1907, pp. 508, 520) insists that the second is but half the 

 size of the first, although he does not state whether he used averages for 

 his conclusion. It seems unlikely that he did, since he says that his fig. 3 

 is the broadest first spindle. It must be admitted that a first spindle may 

 be about twice the size of a second spindle, for we have found that the 

 largest two first spindles measure 29.5X11 micra, and 22.6X14 micra, 

 respectively, and the smallest second spindles 14X6.5 micra and 18X5.5 

 micra, respectively. 



All who have published papers on the mouse, except Kirkham, 

 figure the polar ends of the fibers as thickened. In regard to the fibers 

 which are not attached to chromosomes, there is no conflict between the 



