500 
GENETICS: W. E. CASTLE 
being essentially a synaptic system. It would appear that that primi- 
tive type of nervous organization predominantly present in such coelen- 
terates as the sea-anemones, but preserved in vertebrates only among 
certain autonomous internal organs,^ still forms in molluscs a highly 
important feature of the animal's action system.^ 
^ Contributions from the Bermuda Biological Station for Research, No. 113. 
2 Crozier, W. J., and Arey, L. B., ''Sensory reactions of Chromodoris zebra," /. Exper. 
Zo'dL, Phila., (in press). 
3 Crozier, W. J., 1919, /. Gen. Physiol, Baltimore, 1, No. 6. 
* For more detailed treatment of the complex conditions here entering, consult papers cited 
in footnotes two and three. 
^ cf. Parker, G. H., 1919. The Elementary Nervous System. (Philadelphia.) 
^ Frolich (Zs. Allgem. Physiol., Jena, 11, 1910, p. 269) had already demonstrated that in 
Aplysia the site of action of strychnine is the 'cerebral' ganglion. 
Cushny {Quart. J. Exp. Physiol., London, 12, 1919, p. 153) points out that the depression 
of the reflex thresholds, rather than, as often held, the conversion of 'inhibition' to 'activation' 
(where reciprocal innervation is involved), is the essential feature of the strychnine effect. 
* This conception of the nervous organization of Chromodoris agrees with the opinion held 
by Bethe respecting Aplysia, but the evidence here relied on is much more complete (cf . Bethe, 
Allgem. Anat. u. Physiol, d. Nervensy stems, Leipzig, 1903). 
ARE GENES LINEAR OR NON-LINEAR IN ARRANGEMENT? 
By W. E. Castle 
BussEY Institution, Harvard University 
Communicated, August 13, 1919 
As to the question whether the genes in a linkage system are linear 
or non-linear in arrangement, Morgan and his associates^ still main- 
tain their former view that the arrangement is strictly linear. I 
have questioned the validity of this view on the following grounds. ^ 
(1) The forces which link the genes together are possibly molecular 
rather than mechanical. If so, it is doubtful whether the entire 
system consists of a simple thread-like chain. (2) Construction of 
a model in three dimensions of the relations of the genes in the sex 
chromosome of Drosphila ampelophila as shown by the data of Morgan 
and Bridges,^ and on their own assumption that distances are pro- 
portional to cross-over values, proves that the arrangement can not 
be linear. A similar reconstruction for the sex-linked genes of D. 
virilis shows the same thing for that species even more emphatically. 
(3) The linear hypothesis makes necessary the further assumption 
that cross-overs greater than 50% occur within the linkage system. 
