506 
GENETICS: J. ZINN 
With reference to point (4), it is admitted by Morgan and his 
associates that ''double crossing-over has no meaning, if three genes 
are imagined as not lying in a straight line/' It is accordingly a 
secondary hypothesis needed to help out the hypothesis of linear 
arrangement, but it can not be cited as proof of that hypothesis, 
which must stand or fall on its own merits. I can see no reason a 
priori why two or more breaks should not occur simultaneously in 
different parts of a linkage system, whether linear or non-linear, but 
this is no evidence that, every time a particular gene separates from 
two others, it has done so by two independent breaks, the view necessi- 
tated by the linear hypothesis. The relation of certain observational 
facts to the idea of double crossing-over is correctly stated by Plough^ 
in relation to the alternative hypotheses. On a non-linear hypoth- 
esis, temperature affects ''long chromosomal distances'' (high cross- 
over values) less than small ones; on the linear hypothesis, temperature 
acts by changing the frequency of double crossing-over. This is no 
proof of either hypothesis, but a statement of fact in terms of each. 
1 Sturtevant, A. H., Bridges, C. B., and Morgan, T. H., these Proceedings, 5, 1919, (168). 
2 Castle, W. E., these Proceedings, 5, 1919, (25). 
^ Morgan, T. H,, and Bridges, C. B., "Sex-linked inheritance in Drosophila," Carnegie 
Inst., Washington FubL, No. 237, May 8, 1916. 
4 Castle, W. E., these Proceedings, 5, 1919, (32). 
6 Plough, H. H., these Proceedings, 5, 1919, (167). 
ON VARIATION IN TARTARY BUCKWHEAT, FAGOFYRUM 
TATARICUM (L.) GAERTN. 
By Jacob Zinn 
Maine Agricultural Experiment Station, Orono 
Communicated by R. Pearl, September 5, 1919 
Introductory. — From the morpho-genetic point of view the manifesta- 
tion of dimorphism in certain races of plants — the so called ever-sporting 
varieties — presents a very interesting problem. The remarkable feature 
of these races is the constancy with which the two diverging forms of the 
same organ are transmitted in ever-sporting fashion: no breeding 
method has, as yet, been conceived by which, for instance, certain varie- 
gated types of plants or certain strains of Matthiola, could be induced to 
breed true. These races appear as compound forms ever-transmitting 
the potentialities of the two component types. 
