406 EDMUND B. WILSON 



andi of synapsis, the cumulative force of the evidence in favor of the 

 fundamental fact is irresistible. This question is not to be judged 

 alone by the study of any one of its single phases. The whole 

 extensive series of facts must be reckoned with ; and despite their 

 variations in detail the data are too consistent in their fundamental 

 aspects, to be explained away. On the other hand, it is obvious 

 that the problem as to how the parental chromatin-homologues 

 become definitely associated in pairs is still far from a definitive 

 solution. We should certainly expect a phenomenon so funda- 

 mental to follow everywhere the same type; but I am in agree- 

 ment with the opinion that has been expressed by some other 

 writers (cf. Gates, '11) that the particular mode of union may be of 

 subordinate significance. In accepting the main conclusions of 

 the Schreiners and of Janssens in regard to Tomopteris and Batra- 

 coseps I do not mean to imply that end to end conjugation, or 

 telosynapsis, may not also take place in other forms. I hold no 

 brief for parasynapsis ; and I fully recognize the weight of the evi- 

 dence in favor of telosynapsis recently brought forward especially 

 by the botanical cytologists that have been cited. The studies of 

 King ('07, '08) on Bufo should also be emphasized in this connec- 

 tion. I repeat that the phenomena seen in the insects by no 

 means exclude the possibility of synapsis of this type, at least 

 in some forms. Nearly all observers are agreed that the side by 

 side union begins at or near the free ends of the leptotene-threads 

 and advances step by step along their course. We can here see 

 how readily the one mode might pass into the other; and the sug- 

 gestion of Gates that the mode of synapsis may be correlated 

 with the shape of the conjugating elements at the time of their 

 union seems well worthy of consideration. 



It is not to be denied that the acceptance of parasynapsis in- 

 volves some very puzzling difficulties. It is, for instance, hard to 

 comprehend how long loop-shaped chromosomes can become 

 so disposed as to undergo progressive side by side union from their 

 free ends towards the central points, as both the Schreiners and 

 Janssens have concluded. Janssens appears to recognize this 

 when he says: ''L'eloignement des filaments (i.e., the wide diver- 

 gence of the branches of the F-figures) avant leur soudure est un 



