536 Colonel It. Meinertzhagen on [Ibis, 
We find many examples o£ this, the most remarkable 
being the case of the Chaffinch (Fringilla coelehs). In the 
Canary Islands, as already pointed out by Bannerman (Ibis, 
July 1920), we find undoubted geographical races of the 
Chaffinch living alongside Fringilla teydea , and tending to 
vary in the same direction as a geographical race, as F. teydea 
does as a species. We believe that F. teydea was once but a 
geographical race of F. coelehs , but that isolation has pro¬ 
duced a species. The more recent invasions of Chaffinches 
to the Canary Islands are moving along the same path of 
evolution as did F. teydea , and are undoubtedly incipient 
species. 
Though, therefore, we accept the fact that isolation pro¬ 
duces a true species from an original geographical race, such 
can only occur under conditions of complete isolation. By 
far the majority of geographical races are not incipient 
species, and stand little chance of gaining that status so 
long as they are in actual touch with the parent stock. 
I further wish to endorse the old truth that isolation is the 
strongest factor working for variation, and is the main factor 
in the evolution, not only of geographical races but of 
species. 
4. What is the value of a subspecies ? 
We are not clear as to what is intended by this question. 
To him who asks what is the value of trinomial nomen¬ 
clature, we reply that it has been found the shortest, most 
convenient and scientific way of referring to geographical 
variation. Its critics have not yet suggested a better way. 
To him who asks of what value to science is the recogni- 
tion of slight geographical variation, we reply that it helps 
to solve the problems of migration, evolution, and distribu¬ 
tion among birds, and that such value has been recognised 
by the leading ornithologists of the world. 
But there are still a few who regard the trinomial system 
as a simple and quick way of gaining notoriety, whilst others 
look on the method as a confusing and unnecessary inven¬ 
tion of the Devil, 
