462 Evolution by Segregation. [August, 
The theory of Natural Selection being thus open to doubt 
under a variety of aspects, we may venture to call the atten- 
tion of our readers to a hypothesis which, in England at 
least, has scarcely received the attention it deserves, — the 
doCtrine of Evolution by Segregation, or, as it might be 
otherwise expressed, by isolation in space. According to 
this theory, the less the members of a species are crowded 
or pressed upon — whether by other species or by members 
of their own — the more likely they will be to strike out into 
variations which become permanent as good species. Segre- 
gation, as well as Natural Selection, presupposes a tendency 
to vary, whether from internal or external causes, but, as we 
shall have to show, it throws more light both upon the 
origin of such varieties and on their preservation. In most 
other respeCts the two principles are direCtly hostile, Segre- 
gation ascribing the origin of species to the absence rather 
than to the abundance of competition between individual 
and individual, or between form and form. 
It has been repeatedly urged, and even presented in a 
mathematical form, that if, among a numerous species inha- 
biting one common region, there arises spontaneously some 
modification which gives its possessor an advantage over its 
fellows, there is a very slender probability of such variation 
being transmitted unimpaired, not to say increased, to the 
next and succeeding generations. Whether the modification 
is superiority in speed, or in strength, or the possession of a 
colour favourable to concealment, it is almost a matter of 
certainty that in the course of a few years, by successive 
pairings with individuals of the ordinary type, it must be- 
come obliterated. Nor is this a mere theoretical supposition 
which breaks down in aCtual life. Experiments have been 
made to improve the half-wild cattle and horses of the 
pampas of La Plata, and of the llanos of Venezuela, by the 
introduction of superior Andalusian bulls and English and 
Arab stallions. But the results prove that a minority of 
individuals, whatever may be their physical superiority or 
their other advantages, if mixed with a vast majority of in- 
dividuals of the common type, and freely interbreeding with 
them, can effeCt no permanent modification of the race. In 
these experiments no natural selection occurred, though the 
struggle for existence was manifested under the most favour- 
able conditions. 
But let us suppose that among the animals of a certain 
species inhabiting a given region a few stray away from the 
ordinary haunts of their kind, or, in a word, emigrate. It 
may well happen, as Herr Wagner suggests, that individual 
