1914] Snook-Long : Parasynapsis in Aneides lugubris 521 



similar arms of Y's on the other side, giving rise to the loop by 

 a process of telosynapsis in addition to parasynapsjs^ jjfig. A). 

 That there is no evidence for such a conclusion is indicated by 

 the observation that threads may frequently be followed for long 

 distances around the nucleus and that free ends have never been 

 found except in the vicinity of the sphere. When the threads 

 are definitely formed, there are no morphological indications that 

 an end-to-end union has taken place. 



These considerations indicate at once the validity of the 

 second alternative, for if the threads exhibit no free ends except 

 at the proximal pole of the nucleus where they unite to form the 

 Y-figures, it seems reasonable to conclude that they represent 

 the ends of twenty-eight threads which are evolved from their 

 ends towards their middle at the expense of the nuclear netw r ork. 

 In other words, each of the fifty-six threads becomes directly 

 continuous with one only of its own kind by the time that the 

 loops are complete. If this conception of their origin and nature 

 be sound, the way in which they become associated in the loops 

 might be interpreted in one of two ways: (1) The loops may 

 be formed by the association of two potential threads by a 

 process which begins at approximately the same time at both 

 ends (fig. J3). (2) One leptotene thread might be joined at the 

 ends to two other threads (fig. C). Both of these interpretations 

 have one feature in common, that of a side-to-side union. There 

 seems, however, to be little reason to believe that the second 

 condition actually exists in the male sex cells of Aneides. If 

 each leptotene thread were joined at the ends to other threads 

 at the beginning of polarization (fig. C), as the bifid figures 

 closed up, most of the threads would be drawn together in one 

 or more points in the region of the distal pole of the nucleus 

 (fig. D), which is, in fact, the region of widest separation. The 

 fact that there is no evidence of such behavior upon the part of 

 the threads would seem to dispose of this interpretation. 



Therefore, it seems probable that the number of pairing 

 threads is twenty-eight and that, as rapidly as they are evolved, 

 they unite two by two, side by side, and at both ends, the middle 

 part of the potential threads being lost in the distal half of the 

 nucleus. 



