272 EVOLUTION AND ANIMAL LIFE 



B represents an early spindle stage in the division of the 

 primary spermatocyte, in which not four bandlike chromo- 

 somes, but two tetrads, or chromatin groups or four rounded 

 bodies are found. C to F show clearly the further steps in 

 the spermatogenesis. In C the tetrads are grouped in the 

 equatorial plate, and in Z), in the closing stages of the first di- 

 vision into tw r o spermatocytes, each tetrad has divided into 

 two "dyads/' which are drawn to the poles, and the division 

 of the cell body follows. Without an intervening rest stage 

 each spermatocyte now divides again, as in E and F, each 

 dyad being separated into halves, so that in the spermatids of 

 F but two chromatin masses are present. Thus the tetrads 

 of the primary spermatocyte are divided up among the four 

 spermatids, so that each of the latter receives one fourth of 

 each tetrad. Since later stages show that the two chromatin 

 masses in each spermatid of F represents two chromosomes, we 

 see that the number of chromosomes has been reduced from 

 the four in A to the two in F. 



Manifestly the key to the explanation lies in the relations 

 which exist between the four chromosomes of A and the tetrads 

 of B. The two divisions consist merely in the distribution of 

 the already separated parts of the tetrads ; in the rearrange- 

 ment of the four chromosomes into the two tetrads lies the 

 possibility of the reduction which is carried out by the fol- 

 lowing divisions. The problem thus resolves itself into the 

 question, What is the nature of each tetrad? Is it made up 

 of a single chromosome? of two? of four? or have the constituent 

 parts of the original four chromosomes become so completely 

 rearranged and redistributed that their identity as such is 

 completely lost? 



Turning for a moment to the lower Crustacea, we find among 

 the Copepods forms admirably suited for the careful following 

 out of the changes taking place in the rearrangement of the 

 chromosomes into the tetrads. To Ruckert we owe the clearest 

 account of the process as exhibited in the egg maturation of 

 Cyclops. Here the normal number is twenty-tw r o, or perhaps 

 twenty-four, the minute size rendering counting difficult. Fig. 

 148, A to F, taken from Ruckert, gives the essential points of 

 the formation of the tetrads and their following divisions, not 

 all the chromosomes being represented. In A the chromatin 

 filament has broken up into one half the usual number of 



