THE REDUCTION OF THE CHROMOSOMES 251 



An interpretation put forward by Bonnevie (1906, 1908) is shown in 

 Fig. 99, C. Here the chromosomes conjugate parasynaptically and 

 come into very intimate union: although they appear to undergo a real 

 fusion their identity is maintained for a time. Owing to the fact that 

 these bivalent chromosomes are inserted upon the spindle with their 

 halves in juxtaposition (side-by-side with respect to the poles) rather than 

 in superposition (one toward each pole), the members of a conjugated 

 pair separate neither at the first division nor at the second. As a result 

 each of the four cells receives the haploid number of chromosomes, all 

 of which are bivalent, and no qualitative reduction occurs. Bonnevie 

 believes that the conjugating members of each pair finally fuse completely 

 in the subsequent stages. In this case, therefore, as in the preceding one, 

 numerical reduction is supposed to result from a complete fusion of the 

 chromosomes in pairs. 



Whether any confidence is to be placed in such interpretations or not 

 and according to most cytologists none should be they at least serve 

 to show how it is possible that numerical reduction may occur without 

 effecting any qualitative reduction, and that the essential feature of the 

 reduction of the chromosomes is something other than the mere change 

 in their number, as pointed out at the beginning of the chapter. 



SYNAPSIS, OR CHROMOSOME CONJUGATION 



The phenomenon of chromosome conjugation, or synapsis, which we 

 have seen above is such an important feature of the reduction process, 

 must now be somewhat more closely examined. Attention will be 

 directed to three points: the relationship of the conjugating members 

 (the "synaptic mates"); the stage at which the synaptic union takes 

 place, and the exact nature of this union. 



Relationship of the Synaptic Mates. We may first inquire into the 

 relationship which may exist between the two chromosomes pairing to 

 form a given bivalent chromosome: is any chromosome of the duplex 

 group (the two intermingled parental chromosome sets in the individual's 

 nuclei) present in the gonotokont free to pair with any other chromosome, 

 or does the pairing take place according to more restricting rules? 



It was suggested by Henking (1891) that the two synaptic mates are 

 ultimately derived from the two parents at the previous fertilization, 

 one from the father and the other from the mother: the chromosomes of 

 one parental set pair with those of the other parental set to form the 

 haploid number of bivalent chromosomes appearing on the first matura- 

 tion spindle. This idea was later emphasized and developed by Mont- 

 gomery (1900-4), Sutton (1902), Boveri (1901), and others, who found 

 for it much supporting evidence in organisms with chromosomes differ- 

 ing in size and shape. An observation made by Rosenberg (1909) on 

 Drosera hybrids is significant in this connection. When Drosera rotundi- 



