THE REDUCTION OF THE CHROMOSOMES 243 



VIII) not only that the telophasic alveolation is too irregular to be 

 regarded as a splitting, bu1 also thai the reticulate condition of the pro- 

 phase, instead of developing directly into the definitive split, gives rise 



to simple thin threads in which a new split is developed. From this it 

 cannot be concluded that in no form docs the split develop directly from 

 the early reticulate condition, or thai the telophasic alveolation, though 

 irregular, may not later become so equalized as to constitute the firsl 

 stages of the split; but it does follow thai it is quite unsafe to use the 

 principle of telophasic splitting as a premise from which to draw tin- 

 conclusion that the approximation of thin threads in the early heterotypic 

 prophase represents the reassociation of the halves of a single split 

 chromosome. It is well to emphasize the possible importance of the 

 premeiotic telophase, but any ultimate solution of this perplexing prob- 

 lem must be reached mainly through a more refined analysis of th< 



Fig. 94. — Chromosome pair "B" in Phrynotcttix magnus, showing condensation of 

 bivalent pair during the heterotypic prophases to form the compact chromosomes appearing 

 on the spindle at metaphase. >! 1734. (After Wenrich, 1!»16.) 



prophasic changes which have led a long list of investigators to the con- 

 clusion that the early heterotypic association of slender threads represents 

 a conjugation of entire chromosomes which separate in the firsl matura- 

 tion mitosis. 



One of the most convincing pieces of direct evidence favoring Scheme 

 A is found in Wenrich's recent work on Phrynotettix (1916). Wenrich is 

 able to trace a single pair of chromosomes, distinguishable by their 

 peculiar form and the arrangement of their chromatic accumulations or 

 chromomeres, through every stage from the spermatogonia to the sperma- 

 tids: During the heterotypic prophase the two members of the pair 

 conjugate parasynaptically while in the form of slender filaments. Simi- 

 larly strong arguments are advanced by Robertson L916) as the result 

 of his detailed analysis of the chromosome groups in other Tettigidffi 

 and Acrididae, in which the homologous members can be followed with 

 much certainty because of their frequenl inequality in size. 



Reduction With Chromosome Tetrads. As already pointed out, the 

 marking out of the lines of separation for both maturation divisions 

 during the heterotypic prophase, with the resulting formation of chromo- 

 some tetrads, increases in no inconsiderable manner the difficulty of 

 interpreting the essential changes at these stag* The four chromatids 

 composing the tetrad represent two conjugate. 1 chromosomes each of 

 which is longitudinally split. Because of the variety of ways in which 



