124 
REVIEWS. 
offspring there occurrred fresh mutations, but always in smaller 
numbers than in the case of the parent, 0. Lamarckiana. 
The characters of the new forms are given at length by de Vries. 
Tables showing the date and frequency of their appearance are provided, 
and analytical tables are given, by the aid of which the new forms can be 
distinguished, either as seedlings or from the habit or by examining 
the leaves or the floral organs. The new species are also compared 
with several previously known species of the genus (Enothera allied 
to 0. Lamarckiana. 
As a result of the observations detailed in this section of the book, 
De Vries comes to the following conclusions :— 
(1) The new species arise suddenly, without transitional forms. 
(2) They are usually fully constant from the first moment of their 
origin. 
(3) The characters of the new forms agree in their nature with those 
which distinguish the older species allied to 0. Lamarckiana, e.g., 
O. biennis. Only one new form, namely, 0. nanella, a dwarf type, is 
analogous with any usual kind of garden variety. 
(4) A considerable number of like individuals arise at the same 
period. 
(5) Although the new types vary normally and frequently transgress 
the limits dividing them from the parental type, yet their appearance 
has nothing to do with normal variability. 
(6) Mutations occur indefinitely in any direction. 
(7) The tendency to mutate recurs periodically. 
Any direct observation of the origin of a species was formerly 
supposed to be out of the question. There is therefore much truth in 
De Vries’s observation that the most important conclusion to be drawn 
from this section is the proof that this process is capable of experimental 
treatment. Of great importance also is the fact that towards the end of 
the experiments the ancestry of the mutating plants was exactly 
known for several generations. It'is therefore most unfortunate that, 
as pointed out by Bateson and Saunders, 0 clear proof of the freedom 
of the original strain of plants from a cross with some other species of 
(Enothera is not forthcoming. The fact that plants tracing a pure 
descent through seven generations of 0. Lamarckiana were still able to 
mutate in the same manner as earlier examples seems, however, to 
support the author’s view ; and so to some extent do the results of 
crossing experiments detailed in the second volume. In the meantime 
we can only hope that the point will before long be settled by experi¬ 
ments with other strains of 0. Lamarckiana crossed with allied species. 
But even if it can be shown that the original plants were of impure 
ancestry, the value of these observations and their bearing upon the 
* Report to Evolution Committee of the Royal Society, 1902. 
