192 



Dr. L. Doncaster. Gametogenesis and 



and sinks for a short distance into the egg. The two groups so formed may 

 be temporarily connected by strands of chromatin (Figs. X, 1 and 2). A 

 system of spindle-fibres can then be seen running inwards from the inner 

 ends of the strands, and also, but less clearly, outwards from these strands 

 towards the outer group ; in at least one case (Fig. II, 4) the inner spinflle- 

 fibres can be seen to be 10 in number. At this stage the strands of the inner 

 group lie as a rule fairly accurately parallel with one another, but. show a 

 tendency to become clumped, so that they commonly have the appearance of 

 less than ten moderately thick rods (Figs. I, 3 ; II, 4 ; III, 5, 6 ; VI, 7). 

 The outer strands or rods are often much less regularly arranged, and may 

 appear to cross one another diagonally, or still to be connected by cross- 

 threads (I, 3 ; VI, 5, 7). I have very few examples of this stage in eggs 

 which belong at all certainly to the second (" spherical nucleus ") class, and 

 in them the arrangement of the inner group of strands is so much less 

 regular that it suggests a rather different process (VIII, 2 ; VII, 7), which 

 will be discussed after the eggs with " top-shaped " nuclei have been 

 considered. 



In the eggs of the first class (" top-shajied " nuclei), when the inner group 

 of strands has become arranged as a sheaf of rod-shaped chromosomes on a 

 spindle, the outer group may either follow suit and form a similar group of 

 rods, usually, however, less regular (II, 4 ; III, 5, 6), or, probably more 

 commonly, a division of the inner group takes place before the outer group 

 has formed any very regular arrangement (V, 1, 2), but in these eggs the 

 division of the outer group takes place quite definitely, though it may lag 

 somewhat behind that of the inner group (IV, 5 ; I, 4). Although it is 

 quite clear that a double division takes place, I am quite unable to describe 

 in detail how it is effected, for I have failed to find any eggs (except, 

 perhaps, that shown in Fig. X, 2) in stages between that represented in 

 Figs. II, 4, and III, 5, and that shown in I, 4, IV, 5, and VI, 1, 2. 



In the earlier stages the chromosomes consist of fairly thick rods 

 (presumably ten, though usually aggregated so as to appear fewer), lying 

 lengthwise on a spindle. In the next stage represented these chromosomes 

 have divided into a group of about ten rods, which sink in to form the egg- 

 nucleus, and ten others, which remain as the inner group of polar chromosomes,, 

 but the figures give no indication of whether the division is longitudinal 

 or transverse. The arrangement of the chromosomes on the spindle would 

 suggest a transverse division : the existence of ten rather long, narrow 

 rods in both the egg-nucleus and inner polar group suggests that the 

 division is longitudinal, and this is confirmed by the longitudinal split which 

 is sometimes visible (Fig. X, 1), corresponding with the apparently double 



