CRITICISMS AND CONCLUSIONS. 6l 



The chromosomes of the second spindle are not described by 

 Tafani, except as the description which he gives of those of the supposed 

 first spindle really applies to those of the second. Sobotta (1907, p. 521) 

 holds that they are short rounded rods, similar in form to the daughter 

 chromosomes of the first spindle, though generally somewhat smaller, or 

 at least slimmer. Gerlach (1906, p. 14) is unable to distinguish between 

 the chromosomes of the first and second spindles, except that the latter 

 are the smaller; he figures the same shapes as Sobotta, and also a 

 spindle (fig. 16) having elongated granular chromosomes. We have 

 found in many spindles in which the chromosomes are closely packed 

 that the appearance especially of those chromosomes which are seen 

 in end view, without careful, critical study and comparison with more 

 favorable examples seems to be about like that figured by Sobotta 

 and Gerlach. Lams et Doorme (1907, p. 283) think that the presence 

 of the first polar cell is the only reliable criterion for identifying the second 

 spindle. Kirkham (19076, p. 78), Sobotta (1895, P- 48), and Gerlach 

 (1906, p. 19) state that the daughter chromosomes elongate, but they 

 describe no other structure. We have shown this lengthening to be 

 characteristic of old spindles. 



So far we have made no definite statement concerning the homol- 

 ogies of the chromosomes of the second spindle with those of the first. 

 Whether the mother chromosomes of the second spindle are identical 

 with the daughter chromosomes of the first it is impossible to say with 

 certainty, for the reason that there is no way of determining directly 

 whether or not the chromosomes which become fused into a single mass 

 in the egg after the first polar cell is cut off keep their individuality and 

 reappear when the mass breaks up preparatory to the formation of the 

 second spindle. The striking similarity between the daughter chromo- 

 somes of the first spindle and the mother chromosomes of the second in 

 certain cases, and also analogy with those invertebrates in which the 

 daughter chromosomes of the first spindle are known to pass directly 

 to the second spindle without undergoing an intervening nuclear or 

 resting stage, make it seem highly probable that in the mouse the daugh- 

 ter chromosomes of the first spindle are identical with the mother chro- 

 mosomes of the second. If this is true, then the division between the 

 parts of the chromosome of the second spindle is the same as the longi- 

 tudinal division in the daughter chromosome of the first spindle and is 

 therefore apparent in the fundaments. On this ground it is proper to 

 call the chromosomes of the first spindle "tetrads," because they pos- 

 sess the two divisions which mark the planes of separation of the daughter 

 chromosomes of two quickly ensuing mitoses, and to designate those of 

 the second spindle "dyads." The division of the dyad, then, is a longi- 

 tudinal splitting, and the reduction is a so-called prereduction. 



Tafani (1889) makes the statement that the chromosomes of the 

 first spindle divide longitudinally; but, as we have seen, this statement 

 5 



