HUMAN SPERMATOGENESIS: A STUDY OF INHERITANCE. 11 



of man to contain twenty-two, and of woman, twenty-four chromosomes," in 

 parallel to the conditions found by Wilson (1909) in the hemipteron Syromastes. 



It is hazardous to attempt to solve this problem at present for we understand 

 one side of the equation, and know nothing of the chromosome relations Kb 

 the human female. But it may be a long period before we secure for study 

 maturation and fertilization stages of the human egg, so that a preliminary 

 working hypothesis may be permitted. This hypothesis must account not only 

 for the two classes of spermatozoa recognized by Guyer, but also for the two 

 additional classes found by me. There is reason to believe all four classes of 

 spermatozoa are capable of fertilization, because all have similar structure and 

 no evidence was found of any kind of them degenerating. 



The primary spermatocytes contain both D and d, and so far as we know this 

 combination only. The primary oocytes might be supposed to contain any of 

 the following combinations : d, d, or D, d y or D f D, or D, D, d, d. But the first two 

 of these combinations cannot occur because we know that in other animals the 

 allosome mass of primary oocytes is always greater than that of primary sperma- 

 tocytes. Also the third possible combination cannot occur, for then the mature 

 egg would contain simply Z), a smaller mass than certain spermatids contain. 



Therefore the best explanation of the probable relations in the oogenesis is 

 that given by Guyer, that the primary oocytes contain D, D, d, d> and each mature 

 egg, D, d. Then fertilization by the four classes of sperm would result as follows: 



(a) Egg with D, dX Sperm without D, d = D, d ( d"). 



(6) Egg with D, d X Sperm with D, d = D, D, d, d ( 9 ). 



(c) Egg with D,dX Sperm with D « D f D, d ( 9 ) . 



(d) Egg with D,dX Sperm with d = D, d, d ( 9). 



The primary spermatocytes of both male individuals so far described have 

 the allosomes D and d, and no other combination; therefore fertilization of the 

 type (a) should result in males and the other fertilization types in females. 



In the female individual resulting from fertilization of type (6), it is to be 

 presumed that D would conjugate with D and d with d } the bivalents then under- 

 go a reduction division, and the mature eggs contain D and d. 



In the female individuals resulting from fertilization of types (c) and (d) 

 there might be two possibilities: (1) Such individuals might not develop, but 

 perish. But there would seem to be no good reason for such an assumption. Or 

 (2) (and this result would seem more probable), these individuals also would con- 

 tain D and d in their mature eggs, provided that in them the unmated (unpaired) 

 allosome, whether D or d, remains within the egg and is not discharged entirely 

 into a polar body. There is some basis by analogy for this assumption, for 

 Morgan (1909) found that in the male-producing eggs of phylloxerans two entire 

 allosomes always pass into the polar body; there might then be in the human egg 

 some mechanism by which an unpaired allosome is retained within the mature 

 egg. 



