234 WALTER S. SUTTON. 



in the equatorial plate of the reducing division is purely a matter 

 of chance that is, that any chromosome pair may lie with ma- 

 ternal or paternal chromatid indifferently toward either pole irre- 

 spective of the positions of other pairs and hence that a large 

 number of different combinations of maternal and paternal chro- 

 mosomes are possible in the mature germ-produces of an individ- 

 ual. To illustrate this, we may consider a form having eight 

 chromosomes in the somatic and presynaptic germ-cells and con- 

 sequently four in the ripe germ-products. The germ-cell series 

 of the species in general may be designated by the letters A, B, 

 C, D, and any cleavage nucleus may be considered as contain- 

 ing chromosomes A, B, C, D from the father and <7, b, c, d, from 

 the mother. Synapsis being the union of homologues would 

 result in the formation of the bivalent chromosomes Aa, Bb, Cc, 

 Dd, which would again be resolved into their components by 

 the reducing division. Each of the ripe germ-cells arising from 

 the reduction divisions must receive one member from each of 

 the synaptic pairs, but there are sixteen possible combinations 

 of maternal and paternal chromosomes that will form a complete 

 series, to wit : a, B, C, D ; A, b, C, D ; A, B, c, D ; A, B, C, d ; 



a, b, C, D ; a, B, c, D ; a, B, C, d ; a, b, c, d ; and their conju- 

 gates A, b, c, d ; a, B, r, d ; a, b, C, d ; a, b, c, D ; A, B, c, d ; A, 



b, C, d ; A, b, c, D ; A, />', C, D. Hence instead of two kinds of 

 gametes an organism with four chromosomes in its reduced series 

 may give rise to 16 different kinds ; and the offspring of two un- 

 related individuals may present 16 x 16 or 256 combinations, 

 instead of the four to which it would be limited by a hypothesis 

 of parental purity of gametes. Few organisms, moreover, have 

 so few as 8 chromosomes, and since each additional pair doubles 

 the number of possible combinations in the germ-products 1 and 

 quadruples that of the zygotes it is plain that in the ordinary 

 form having from 24 to 36 chromosomes, the possibilities are 

 immense. The table below shows the number of possible com- 



of certain fish hybrids to distinguish the maternal from the paternal chromosomes by- 

 differences in form, and if the same can be done in the maturation-divisions the 

 question of the distribution of chromosomes in reduction becomes a very simple 

 matter of observation. 



1 The number of possible combinations in the germ-products of a single individual 

 of any species is represented by the simple formula 2" in which n represents the 

 number of chromosomes in the reduced series. 



