416 EDMUND B. WILSON 



found to divide but once (equationally) in the course of the two 

 spermatocyte-divisions, as in case of the X-chromosomes, or the 

 m-chromosomes . 



When these facts are taken together the conclusion seems to 

 me unavoidable that one of the divisions of the bivalent chromo- 

 some (and hence, one of the division-planes seen in the bivalent 

 prophase-figures) is a consequence of its hivalence — ^i.e., of its origin 

 from two chromosomes instead of one, of the original diploid 

 groups. An almost conclusive demonstration of this is given by 

 the fact that when the X- and F-chromosomes are united to form a 

 bivalent in the prophases, this body, like the others, often shows a 

 tetrad structure (as in Brochymena, Wilson, '05 b, or Nezara, 

 Wilson, '11 a); and in the case of Ascaris fehs, recently described 

 by Edwards ('11) this bivalent has a double cross-form, closely 

 similar to that of the other bivalents save for the inequality of 

 two of the components (op. cit., fig. 2). I do not mean to imply 

 that either division-plane of the tetrad represents the actual plane 

 of separation of the same two chromosomes that have united in 

 synapsis; on the contrary, I think it probable, as already indicated, 

 that the original chromosomes may have undergone reconstruc- 

 tion. What may be said is that one division is independent of 

 bivalence, the other a consequence of it ; and it is further clear that 

 the former effects no reduction of valence, while the latter does. 

 Whether we regard the autosome-bivalent as to its origin or its 

 fate, it has, irrespective of its relative size, double thechromo- 

 somic value of a univalent in the maturation-process; and in 

 this respect it is exactly comparable to an XF-bivalent or an m- 

 bivalent, in which one of the divisions is demonstrably a reduc- 

 tion-division in the original sense. This value, or 'valence' is 

 reduced to one-half in one of the maturation-divisions. May we 

 not here find a definition of the reduction-division that may be 

 accepted even by those who deny the individuality of the chromo- 

 somes, or who believe that synapsis is followed by actual fusion? 

 We may define an equation-division as one that effects no reduction 

 of valence, a reduction-division as one that reduces the valence to 

 one-half. This conforms exactly to the observed facts; and such a 

 definition is, I think, equally consistent with complete disjunc- 



