Studies on Synapsis. 



63 



chromosomes well separated in the equatorial plate, were available ; in every 

 case there were seen to be twenty-three stout, rod-like and slightly curved 

 elements of approximately the same size. In spite of the fact that Smith gives 

 the number twenty-five for the diploid complex of L. basalis, it is without the 

 slightest hesitation that the former figure is stated as the correct count in 

 the case of L. depressa. In many preparations (Plate 4, fig. 1) there are 

 seen two (or sometimes one) small granules among the metaphase chromo- 

 somes ; these occur also in the metaphase figures of follicle cells and 

 oogonia (figs. 26-7) ; and it was at first an attractive supposition that these 

 may be the vestigial representatives of the additional pair of chromosomes 

 in L. hasalis, especially in view of the fact that Lefevre and McGill describe 

 a pair of m-chromosomes in Anax Junius. This possibility was fully 

 explored ; and it was found that three such bodies were present in many 

 cases in prophase nuclei of spermatogonia, while one, two, or even three 

 were present, separated well apart in the first spermatocyte metaphase 

 (fig. 22). It will be remembered that McGill originally could only find one 

 " m-chromosome " in Anax ; and the inconstant number of these granules, 

 taken in conjunction with their failure to segregate in the reduction division, 

 makes it very improbable that they can be justifiably interpreted as elements 

 of the diploid complex in L. depressa at least. 



An alternative proposition, that they constitute a group of Y-elements, is 

 disposed of by comparison with the oogonial nuclei. In the prophase, these 

 granules are first seen at the time when the plasmosome disappears; and 

 there are considerable indications that they represent persistent remnants of 

 the nucleolus passed over from one cell generation to another (fig. 10). One 

 of the most interesting discoveries made with the aid of the new technique is 

 that ot Carleton, 1920(7), who has shown that one or more intranucleolar 

 elements capable of fission in mitosis pass over from one cell generation to 

 another, and constitute the centre around which the new plasmosome is 

 formed in the late telophase. In the light of this discovery, as well as in 

 view of the facts mentioned, the last supposition appears to be the most 

 acceptable. 



As the chromosomes pass to the poles of the spindle, they are seen to 

 become more curved; they also give clearer evidence of individuality in 

 size ; in the late anaphase (fig. 2) they become more sharply curved, and one 

 or two pairs are seen to become attenuated into looped filaments. There is 

 no evidence of cleavage either in this stage or in that which follows. When 

 the nuclear membrane is defined, the chromosomes are seen to spin out into 

 finely beaded loops, the extremities of which are polarised on the side of the 

 nucleus remote from the centrosome (figs. 3, 4). This polarisation is 



