234 
Allen . — Nuclear Division in the 
suggest that it is in some way connected with the formation of only half 
the usual number of chromosomes ; and Moore (’94, ’95 a) first definitely 
maintained that it is during synapsis that the numerical reduction occurs. 
Those observers who have found what they consider reduction divisions 
in Weismann’s sense have for the most part interpreted their results in 
essential accordance with those of Riickert (’92 a, ’92 b, ’94), Hacker (’90, 
’92, ’93, ’95, ’99), and vom Rath (’92, ’93) ; that is, they find that in the 
prophases of the heterotypic division the chromosomes are bivalent, 
consisting of two somatic chromosomes attached end to end ; and that 
in one or the other of the two succeeding divisions these two somatic 
chromosomes are distributed to different nuclei. This method of reduction 
has been described especially among the Arthropods. Montgomery (’ 01 ) 
suggested that the two chromosomes so joined end to end are always 
derived from different parents ; synapsis is, then, a process of conjugation 
of the maternal and paternal elements, its object being a ‘rejuvenescence ’ 
of the chromosomes. He thought it probable that there is an exchange of 
substances during this temporary contact. 
Cannon (’02, ’03 a) suggested, as a cytological basis for Mendel’s law 
of the purity of the germ-cells, that c the chromosomes derived from the 
father and the mother unite in synapsis and separate in the metaphase of 
one of the maturation divisions, and also a single longitudinal division 
occurs, so that the end is attained that the chromatin is distributed in such 
a way that two of the cells receive pure paternal and two cells pure 
maternal chromosomes, and no cells receive chromosomes from both the 
father and the mother.’ Cannon’s notion of the purity of the germ-cells 
at this time was plainly quite different from that of Mendel, for the sort of 
division postulated by Cannon could not possibly result in new combinations 
of parental qualities such as Mendel actually found. This difficulty was 
recognized by Sutton ( 03), and later by Cannon himself (’03 b). Sutton 
proposed to modify Cannon’s conception by assuming that it is a matter 
of chance whether one or the other chromosome of a pair goes to a particular 
daughter nucleus ; and this notion has been accepted by Boveri (’04). The 
result of the operation of chance in this respect would be that in a sufficiently 
large number of germ-cells from the same individual every possible com- 
bination of the parental chromosomes will occur. The number of possible 
varieties to be found among the offspring would depend, accordingly, upon 
the number of chromosomes characteristic of the species. Sutton showed 
that in cases of the occurrence of the commoner chromosome numbers the 
amount of variation thus made possible is relatively great ; for instance, if, 
as in the lily, the reduced number of chromosomes is twelve, the number of 
possible combinations that may occur in the germ-cells of one parent 
is 4 , 096 ; and, as the result of the fusion of these germ-cells with those from 
a similar parent, the number of possible combinations of parental characters 
