248 Robertson : Theories of Evolution. 
appeared in Massachussetts in 1791 as a sudden variation ; it 
was kept and bred from because it had bent legs which pre- 
vented its jumping into neighbouring fields, and it has formed 
a distinct and permanent race ever since. In 1761 Duchesne 
found a strawberry plant in his garden with simple instead of 
ternate leaves; this bred true and is still in cultivation. I 
mention these cases because they are so very well marked, but 
the advocates of the mutation theory do not demand such large 
jumps as these for the evolution of species. The question, ‘What 
is a species?’ is one for which no biologist has yet found a 
satisfactory answer, but still we have a working notion of what 
we mean when we speak of a species. We mostly use the term 
in the sense of classifactory binomial unit of Linnzeus [e.g., 
Veronica chamedrys=the Germander Speedwell]. But later 
investigations have shown that many systematic species are not 
really units at all, but comprise a number of distinct races, 
which are found on cultivation to breed true indefinitely. In 
some extreme instances the number of races lumped together 
under one specific name is very large. Dvraba verna, the little 
Whitlow-grass, has over 200 local races, which are uniform 
and come true from seed! These sub-divisions of a species 
have received the name ‘elementary species,’ and the ‘mutation’ 
of a species generally does not imply a greater jump than that 
from one elementary species to another. 
The great worker on the subject of mutations has been Prof. 
Hugo de Vries. There is a kind of Evening Primrose, Gnothera 
Lamarckiana, which is unknown in the wild condition. Not 
very long ago De Vries discovered it growing in thousands, 
apparently as an escape, in a field near Amsterdam. He 
noticed that while the majority of the plants were of the true 
CG. Lamarckiana type, others differed widely from it in such 
characters as height, leaf-form, leaf-surface, pigmentation, and 
soon. For the sake of more satisfactory observation he dug 
up, in 1886, nine large rosettes of the type form, and transferred 
them to his own experimental garden. The seeds of these 
produced 15,000 young plants, all but 10 of which were of the 
true G. Lamarckiana type, while the divergent ten belonged 
to two quite new types, to which De Vries gave new specific 
names. In the same way in later generations a great majority 
of normal and a small minority of divergent types were produced. 
Seven new forms were thus obtained, and were found to come 
true from seed. These new forms De Vries regards as elementary 
species, and considers that in Lamarck’s Evening Primrose we 
Naturalist, 
