24 DUBLIN 



ja and jb, PL XXII) of Riickert's memoir, " Zur Eireifung 

 bei Copepoden." 



The sister tlireads of a bivalent chromosome, now fully ex- 

 tended, may secondarily unite at one of these free ends and in 

 this way give rise to a new long-armed V (Figs. 80-8 1 ). This 

 figure must not, however, be confused with those observed be- 

 fore the opening up of the angle of the synaptic structures. 

 The arms of the early V's are each longitudinally split, of which 

 condition there is not a trace in the later ones ; and secondly, 

 while the arms of the former are univalent chromosomes which 

 have united in synapsis, those of the latter are bivalent sister 

 chromosomes which have become secondarily apposed at one 

 end. The middle point of the long arms of the second V's, 

 therefore, corresponds to the apices of the V's in the first (cf. 

 Text-fig. I, p. 13). In the greater number of cases the sister 

 bivalent chromosomes do not unite at one point alone, but often 

 twine around each other, thus forming 8's and loops of many 

 types ; in all, however, the same homologies pointed out above 

 must hold (Figs. 82-84). 



The chromosomes, having lost somewhat in their staining 

 capacity during the last period, again increase in this regard 

 and lose at the same time their ragged appearance. The linin, 

 staining very lightly, forms a fine meshwork over which the 

 chromosomes appear distributed, very different from the deep 

 staining reticulum of the earlier stages. The chromosomes are 

 in all stages of concentration and vary greatly in size (Fig. 84). 

 In general, the ensuing process may be summarized as follows : 

 The two sister threads unite at one of their ends, if they have 

 not at some earlier stage already done so, and then the second 

 or the free ends come into contact, forming a ring. Or the arms 

 may first cross and then unite at the ends ; in this way are ob- 

 tained, I believe, the very common 8 -shaped figures (Figs. 84— 

 85). In other cases the second pair of free ends may fail to 

 unite, but come close together, thus forming two thick bars in 

 the form of a V. Finally (and this, as the figures show, is of 

 very frequent occurrence) the bivalents may fuse along their 

 whole length, forming long rods of considerable thickness. 



