THE SPERM ATOGENES1S OF THE SPIDER. 1/5 



SEQUENCE OF REDUCTION AND EOUATIONAL DIVISIONS. 

 Most authors now grant that one transverse and one longi- 

 tudinal division obtain in the maturation divisions of the germ 

 cells, but there is still a lack of argreement in regard to the se- 

 quence of these two divisions. There has been a decided major- 

 ity in favor of the view that the longitudinal division occurs first. 

 Montgomery has emphasized the importance of determining the 

 origin of the chromosomes in order to get at the truth of this 

 matter and he claims to have established the fact that " the 

 heterotypic mitosis, the first maturation mitosis, is not an equa- 

 tion division but separates entire univalent chromosomes, while 

 the second maturation mitosis is equational." This conclusion 

 is supported by the work of Korschelt on the ovogenesis of an 

 annelid, by the work of Henking and Paulmier on Hemiptera, fo 

 Miss Nichols on isopods and by several others. There is the 



e. f g k 



^^ 



3 



fYXx 



FIG. i 



possibility that uniformity in regard to the sequence of the divi- 

 sions does not exist since in either case the ultimate result is the 

 same, but if there is uniformity the spider spermatogenesis seems 

 to me an especially good field for determining this point and it 

 shows strong grounds for believing that the first division is a re- 

 duction division. The process is as follows : When pseudo-re- 

 duction occurs in synapsis, giving rise to nineteen loops of dif- 

 ferent lengths, there is at first no split visible in the thread of the 

 loop. When the split does appear it becomes steadily more 

 noticeable during the growth period and is then obscured in the 

 condensation of the chromatin to form the thickened V-shaped 

 chromosomes (text figure, a and &). In the prophase of the 

 primary spermatocyte, the V's split from apex to base, parting 

 along the line, I maintain, of the original, longitudinal split of 



