RACIAL AND SUBSPECIFIC NAMES. - 169 
clusive proof that the separated race is a subspecies, and if this be so, 
then it seems unnecessary to find any varietal differences, though as a 
matter of fact, such differences practically always exist where races are 
so separated. The real proof that geographical races are subspecies, 
or that any two different races of a species are subspecies, is not to be 
found in their segregation, or in the amount of their differences, but 
in the circumstance, that these differences have a permanence under 
any disturbance as of habitat, etc., that makes some approach to the 
permanence under such disturbance of a species. ‘That this can rarely 
be ascertained makes it necessary that we should observe a doubtful 
attitude in most cases, admitting that we are unable to decide in 
either way, unless so strong a fact as geographical separation obtains. 
To understand the nature and causes of such local races, to find 
out whether they be subspecies or no, t.e., whether under disturbance 
they retain their racial characteristics, or in a limited time revert to 
some other ordinary form of the species, is at least as important a 
question as any relating to varieties, and consequently if they have to 
be discussed in any case, the necessity of recognising them without a 
description, which Mr. Wheeler accepts as the criterion of the neces- 
sity for a name, is at least as great as in the case of varieties. 
Mr. Wheeler’s contemptuous reference to the numbers and pro- 
portions of different forms in any race must arise from not under- 
standing my statement. I don’t want to meddle with varietal or 
aberrational names in any way, but I again assert that two races of a 
species differ in a subspecific manner and possibly (until it is proved 
or disproved by experiment) to a subspecific extent, even if the only 
difference between them is that the varieties of which they are com- 
posed occur in decidedly different proportions in the two races. A 
composite photograph of either race would differ appreciably from that 
of the other. 
The diagram (and corresponding text) in Rothschild’s and Jordan’s 
Sphingidae, p. xxxv., precisely asserts my proposition in pointing out 
that subspecies may differ from each other merely in the proportions 
of the varieties of which they consist, which I take it is, nevertheless, 
the item of unsound reason that is beyond Mr. Wheeler’s imagination. 
In passing, | may note that they drop any very definite use of the 
word “variety,” making it cover all variation. They use ‘‘ subspecies” 
in the sense I adopt, and form for what I call a variety. 
There is much to be said for this attitude, considering the am- 
biguity that varied usage has attached to the word “ variety.” 
Syngrapha, to take one of Mr. Wheeler’s illustrations, is practically 
absent in most races of coridon, an aberration in many, in the Charente 
Inférieure it is a very predominant variety. This race, therefore, de- 
serves a name, but the name synyrapha is not interfered with in any 
way. 
The racial name includes both varietal and non-varietal forms of the 
race, but does not interfere with the varietal names. The illustration 
drawn from Lycaena arion by Mr. Wheeler may serve to explain the 
position. I know something of the life-history of L. arion, but of its 
subspecies, varieties, and aberrations, I know very little, therefore I 
deal with the matter somewhat hypothetically. Mr. Wheeler mentions 
ligurica as a racial form, i.e., as a subspecies, but he implies also that 
it is the name of what I call a variety. If liywrica is a subspecies, 
