Gates .— The Trisomic Mutations of Oenothera. 
behaviour of such 14-chromosome mutants as nanella and rubrinervis 
through double non-disjunction, although their origin and the fact that they 
breed true may be accounted for in this way. There are a few other forms 
whose [origin could be explained by crossing-over, but this hypothesis 
equally fails to explain their later hereditary behaviour. Perhaps a critical 
repetition of some of the breeding experiments might throw light on these 
difficulties. 
Let us now consider the situation as regards the trisomic forms. We 
have already seen that lata can apparently give rise to all the other trisomic 
forms,andthat they usually arise with higher frequency from lata x Lamarck- 
iana than from Lamarckiana itself. This must mean that when they arise 
from lata x Lamar ckiana it is, usually at any rate, by the union of an 8-egg 
with a 7-male cell. The mutation giving a different trisomic form might 
then arise from a male gamete in which double non-disjunction has taken 
place. 
Thus if we suppose that lata is 
AABCDEFG 
ABCDEFG ’ 
its female gametes 
will be AABCDEFG and ABCDEFG in equal numbers unless some irregu¬ 
larity occurs. Double non-disjunction in a pollen mother-cell of Lamarck - 
iana might produce a male gamete ACCDEFG. Hence a trisomic 
individual could arise which was 
AABCDEFG 
ACCDEFG 
Let us suppose this is 
the formula for scintillans. Then the germ cells of this mutant could form 
various recombinations of chromosomes, such as (considering only the un¬ 
balanced pairs) 
binations could occur as 
A ABC A ACC, ABCC 
~acc' abc ’ Mac 5 
AABC 
&c. In fertilization such recom- 
(, lata ), (scintillans), —(which 
might represent a different trisomic form), &c. Hence we can understand 
how lata could give rise to scintillans and vice versa, and also how lata 
might be the starting-point for a whole series of other trisomic forms 
arising through further irregular distributions of the chromosomes. 
Whether such a form as, e. g., --— is viable would depend on how far 
A C C 
the chromosomes are differentiated from each other. 
It has recently been found that a haploid mutant can arise in Datura 
(Blakeslee, Belling, Farnham, and Bergner, 1922 ). This number of chromo¬ 
somes is then sufficient for the production of a complete sporophyte. If 
duplication and redistribution of chromosomes could go on indefinitely it 
might be possible to produce an organism with all A or all C chromosomes. 
Weismann’s theory of the germ-plasm contemplated the view that each id 
contained all of the representative germinal material, each chromosome 
being made up of a number of ids. Speculation has since swung to the 
opposite extreme, and it has been rather tacitly assumed that the chromo- 
