W8 
L. DONCASTER AND J. GRAY. 
v2) 111 early anaphase it is often possible to see which 
chromosomes in the daughter-groups correspond Avith each 
other, and it is usually the swollen and irregular ones which 
appear to be Avithout mates; unfortunately, however, the 
narrow and crowded spindles make the determination of this 
point uncertain in many cases. (3) The third piece of evi- 
dence is from counts of the chromosome number. If acutus 
has 38 and rniliaris 34 chromosomes, the hybrid should have 
36. We have found two eggs in which the spindle is replaced 
by a monaster (fig. 26), and in these the chromosomes, though 
some of them are constricted, have not divided, and in each 
36 can be counted, in one case Avith confidence, in the other 
Avith great probability.^ Prophases after the nuclear mem- 
brane is dissolved are in this cross usually too crowded for 
counting, but Ave have found one in Avhich about 36 are also 
present. In almost all other counts of both batches of eggs 
,E IV. — Chromosome Counts 
rniliaris c? . 
in Acutus 
Propliases and Metaphases. 
Anaphases. 
36 (39 ?) (monaster) 
36 1 
36-f-l small vesicle (monaster) 
34 J 
32 (35 ?) -f 1 vesicle 
35 
36 + 1 ,, )' 
35 
31* + 1 „ ) 
35 (33-36) 
29* +1 „(?)) 
35 1 
34) 
32 (33 ?) J 
3/ f 
34 1 
30-34 -b 1 A'esicle ) 
35 ) 
34-35 1 
35-37 I 
33-35 i 
31 1 
37 J 
* These numbers are probably too low. 
' That the number when no vesicles are present should be 36 is 
further indicated by the similar cross escAilentus $ X rniliaris 
(see below), in which the number is almost certainly 36. 
