942 Davis,—A Comparison of the Reduction Divisions of 
a number of writers have already suggested, many of the so-called mutants are 
likely to be interpreted as segregates splitting off according to Mendelian ex¬ 
pectations. Nevertheless,whatever shall be the final judgement concerning the 
significance of some or perhaps most of the forms derived from Lamar ckiana, 
we have in gigas at least a type in which profound changes have taken place 
in the germinal constitution through the doubling of the chromosome 
number characteristic of the genus Oenothera. 
Oenothera gigas appears to be a true mutant, or sport, since it is reason¬ 
able to believe that its strongly marked peculiarities are directly associated 
with that change in nuclear structure which increased the number of 
sporophytic (somatic) chromosomes from fourteen to twenty-eight. This 
peculiarity of nuclear structure involves a totally different form of origin 
from any that is likely to be established for the other derivatives from 
L amarckiana, and gigas therefore among these derivatives probably stands in 
a class by itself, and this is also indicated by the extreme rarity of its 
appearance. Whereas many of the other of de Vries’ ‘mutants’ have 
appeared in cultures in sufficient numbers and proportions as to suggest 
a Mendelian segregation of forms from a hybrid, gigas on the contrary has, 
as far as the writer is aware, been reported as arising only seven times 
among the thousands of cultures of Lamarckiana and its derivatives made 
by de Vries and the later investigators who have worked with these forms.* 
As described in ‘ Die Mutationstheorie gigas first appeared in 1895 
among a group of thirty-two rosettes of Lamar ckiana selected by de Vries 
(’ 09 , p. 227) from a very large culture of this type in the fourth generation. 
It is probable that all of the gigas material now in cultivation is descended 
from this plant. Cn two other occasions gigas is believed by de Vries (’ 09 , 
p. 327) to have arisen in his cultures, once in 1898 from seed of sublinearis 
(a ‘mutant ’ of L amarckiana) , and once in 1899 from a cross of lata x hirtella, 
but neither of these plants matured. MacDougal (’ 07 , p. 3) reports the 
appearance of one plant of gigas in his cultures of Lamar ckiana, from seed 
of de Vries, at the New York Botanical Garden. Lastly, Schouten (’ 08 ; 
Ref., Bot. Centbl., vol. cviii, 1908, p. 246), a student of de Vries, records 
three plants of gigas in cultures from commercial seed of Lamar ckiana. 
Thus gigas at the most has been noted only seven times, and, since appa¬ 
rently the cytology of de Vries’ strain alone has been studied, it is by no 
means certain that all of the forms reported later are the same as the first 
example from de Vries’ cultures of 1895. 
The extreme rarity of gigas as a derivative of Lamarckiana seems to 
the writer exactly what should be expected on the supposition that this type 
is a true sport, or mutant. The fact of its rarity, probably based on the 
marked peculiarities of its nuclear organization, indicates that gigas is an 
exception, and that other derivatives from Lamarckiana may finally be 
regarded as variants of quite a different sort from the ‘ mutants ’ of de Vries’ 
