560 Gates .— The Trisomic Mutations of Oenothera. 
Mende’lian characters whose inheritance can be used as a check, a further 
analysis of the situation will be possible. There is apparently, however, no 
evidence at present that any of the trisomic mutants in Datura are inter¬ 
changeable, so the conditions in the two genera differ in some respects. The 
further study of the offspring of triploid Oenothera mutants and hybrids will 
also throw further light on these questions. 
One question which remains to be discussed is the possible visible 
differentiation of the seven pairs of chromosomes in Oenothera . Hance ( 1918 ) 
claims that the chromosome pairs of Oenothera scintillans form a graded 
series in length, but it must be said that his treatment of the subject is not 
entirely convincing. The difficulties of obtaining all the chromosomes of 
a large number of equatorial plates exactly in the plane of section are very 
great, and the frequent bendings of the chromosomes render accurate 
measurements still more precarious where small differences in length are 
involved. Van Overeem ( 1922 ), in an important paper, has, however, 
reached a similar conclusion. He identifies the extra chromosome in lata 
(which we have called A) as belonging to a trio which are long and strongly 
bent. He calls this chromosome No. 1 (see his PI. II, Figs. 9-12). In cana 
he identifies the extra chromosome as belonging to No. 5 pair, a pair 
which he believes is probably of medium length and is bent near the end 
and often somewhat constricted (PI. II, Figs. 1-8). If these differences can 
be proved to be constant this will help to substantiate the views expressed in 
the present paper. But more critical and extensive measurements of these 
chromosomes must be made before one can regard it as proved that the 
chromosomes of Oenothera form a series which are constantly differentiated 
in size and shape. The facts, as far as they go, are entirely in accord with 
the views here expressed, and it is to be hoped that with further evidence it 
may be possible to identify the chromosome composition of each form. 
The possibility should be kept in view that some forms with an extra 
chromosome may at the same time be lacking both members of a different 
pair. 
It may be pointed out here that van Overeem ( 1922 ) concludes from the 
genetic data that the extra chromosome in lata belongs to the gaudens 
complex (since in crosses it gives, like Lamarckiana , the twin types laeta 
and velutina) and that the same is true of scintillans and oblonga , although 
the extra chromosome is a different one in all three. In a paper now in 
the press I have shown reasons for believing that the difference between 
gaudens and velans resides in a single pair of Lamarckiana chromosomes. 
How these two views are to be harmonized can only be determined by 
further breeding experiments with various trisomic forms. There seems no 
reason for assuming that in any of these mutants the gaudens-velans pair of 
chromosomes is the extra one. There should however, on my view, be one 
trisomic mutant in which this was the case, and this should distort the 
