352 J. A DETLEFSEN AND E. ROBERTS 
present in a general population and are the basis and cause of 
this variable chiasmatype relationship. If this explanation is 
correct (and we are inclined to believe it the most plausible one 
of those we have suggested here), then we cannot escape a 
marked change in our view-point on crossing over and related 
phenomena. If, for example, all of the difference between prac- 
tically no crossing over in our series A and A' and normal cross- 
ing over (33 per cent) is due to numerous modifying factors, then 
we naturally begin to wonder just what part 'distance between 
two genes' on a chromosome may play in determining linkage 
values. Our current view is that ''the percentage of cases in 
which two linked genes separate (amount of crossing over between 
them) is necessarily proportional, other things being equal, to 
the distance between the genes," (quoted directly from Weinstein 
('18)). But evidently the percentage of crossing over is a vari- 
a.ble which is the expression of different possible combinations 
of multiple modifying factors; hence the percentage of crossing 
over cannot be proportional to the distance if the distance 
remains uniform. For example, in series B we find 6 per cent 
crossing over, and so we should conclude that the distance in 
this stock is 2/11 or 18 per cent of what it was when we began 
selection! Thus, to maintain our original position, we must 
conclude that the percentage of crossing over and distance are 
correlated variables, if the proportion between the two is to 
remain reasonably constant. We then naturally begin to wonder 
what has happened to all of the distance (and the genes) between 
and 33 in series A and A' where crossing over has been prac- 
tically eliminated. In view of these considerations, it would 
perhaps be simpler to conclude that the percentage of crossing 
over is not necessarily proportional to the distance. The dis- 
tance may remain fairly constant, but the amount of crossing 
over (i.e., twisting of the chromosomes) will depend upon numer- 
ous hereditary factors. 
One recalls in this connection Goldschmidt's ('17) suggestive 
paper in which he postulated variable forces that hold genes to 
their customary loci on the chromosome and which aUow an 
exchange between allelomorphs in a certain average percentage 
