Robertson : Theories of Evolution. 249 
have caught a species in the mutating period, in other words, 
in the act of giving birth to a number of new species. De 
Vries imagines that all species have periods of stability during 
which they only vary according to the laws of fluctuating 
variability, and then pass through a mutating period, to which 
a period of stability again succeeds. 
De Vries supposes that the struggle for existence is not so 
much between individuals of a species as between different 
elementary species. When a group of elementary species 
arises by mutation, natural selection determines which of these 
elementary species shall survive. 
On general grounds the chief point in favour of the mutation 
theory is that it does not require so long a period for the 
evolution of the organic world as that demanded by the 
Darwinian theory, and is thus more in harmony with the 
calculations of physicists as to the age of the earth. 
De Vries’ theory has been much criticised. A great deal 
depends on the previous history of G2. Lamarckiana. De Vries 
assumes that it is an ordinary pure species, but it has been 
suggested that it is possibly a hybrid. The fact that it is 
unknown in the wild state, and the sterility of some of the 
pollen and ovules, lend support to this suggestion. The 
‘mutations’ isolated by De Vries strongly recall the ‘ analytical 
variations’ of Bateson, obtained by hybridising sweet-peas, etc. 
So it seems that the question of the real existence of mutations 
must remain an open one till further experimental evidence is 
brought forward. 
VII.—CONCLUSION. 
There are a number of important branches of evolutionary 
work which I have not even touched upon in this brief outline 
—amongsst others the contributions of Weismann and the work 
of the modern biometrical school. I have confined myself to the 
briefest possible mention of those lines of research which seem 
to me most hopeful for the future. It is experimental work on 
heredity, such as that undertaken by those who have confirmed 
and extended the conclusions of Gregor Mendel, which gives 
the clearest promise of further light on these huge and com- 
plicated problems.* 
.* I should like to refer any reader who wishes to follow up the subject 
to ‘Recent Progress in the study of Variation, Heredity, and Evolution,’ R. 
H. Lock (Murray, 1906). 
1907 July 1. 
