298 
R. Ruggles Gates. 
first colonized, and that any such crosses would therefore have been 
repeatedly made there long before the interference of man. Another 
weakness of this view is that there is no particular reason for 
supposing that O. Lamarckiana characters can be duplicated by 
single crosses between biennis and grandiflora , any more than biennis 
could be produced by crosses between muricata and grandiflora. 
It has been suggested by Davis (10) that O. Lamarckiana 
Seringe is only a form of O. grandiflora Solander. The suggestion 
is based chiefly on a comparison of photographs of the type 
specimens in the Museum d’ Histoire Naturelle in Paris. 
The new introduction of O. grandiflora from Alabama in 1778 
was, no doubt, at least slightly different from the “Virginian 
grandiflora ” of Ray and Barrelier, and it appears probable that 
Seringe meant to express this difference in recognizing Lamarckiana 
as well as grandiflora. In any case it is obvious that the question 
cannot be settled merely by determining the source of De Vries’s 
race of Lamarckiana in 1860, for the origin of the much older race 
of Lamarckiana in England must also be considered, and this may 
easily go back to the time of Ray. 
Since it is now clear that numerous races both of grandiflora 
and Lamarckiana exist, it seems possible that the original seeds of 
the large-flowered form brought from “Virginia” contained a 
mixture of interbreeding races, some of which would now be classed 
with either species. Certain of the CEnotheras now grown in 
English gardens might very well represent such intermediate races. 
Considering the great number of local geographic races of 0. biennis 
now being described, it is probably safe to assume that the grandi¬ 
flora of Virginia was different from that of Alabama. 
Finally, three of the recent papers have dealt with various 
general aspects of the mutation problem. These papers are by 
De Vries (12), Heribert-Nilsson (24) and Gates (21). De Vries 
reviews the progress which has been made in the study of mutations, 
particularly in OEnothera , and reaffirms his earlier views, such as the 
premutation theory. He points out that natural selection, mutation 
and orthogenesis are not mutually exclusive as evolutionary factors, 
but that all have probably played their part. Heribert-Nilsson 
contributes a mass of breeding data on a Swedish race of 0. 
Lamarckiana , and attempts to explain the mutation phenomena in 
terms merely of Mendelian splitting. In doing so he disregards the 
cytological facts and frequently runs counter to them. He elaborates 
a purely hypothetical theory involving the gradual accumulation of 
unit-factors or genes in particular germ-cells, but his theory falls to 
pieces in the light of the cytological facts. As the writer has 
pointed out, the Mendelian theory of mutation has been disproved 
and the premutation theory of De Vries rendered unnecessary by 
the study of the nuclei. The time has come for a new theory of 
mutation, based on our present cytological and experimental data, 
and the main achievement of the last three years has been to show 
that mutation is an independent process requiring a special 
explanation. 
