30 
Biologically, I take ita sub-genus is a genus in process of evolution 
from a parent stock, but our British Vanessids represent well-defined 
diagnosed generic divisions to which the term “sub-genus” is most 
distinctly inapplicable. It appears to me that if, when the Vanessids 
of the world are considered, certain genera or evolutionary groups are 
discovered with certain specialised characters in each group, and if it 
be further found that our various British species belong to several of 
these different evolutionary and _ structurally distinct groups, it is 
certainly right and proper to refer each of our species to its own 
generic group. It is reducing science to absurdity to say that the 
generic characters which it is necessary to recognise for the purpose 
of “classifying the numerous exotic species,” become simply specific 
characters when we come to consider the British fauna, because we 
possess so few species. 
It matters not at all if we have only one species in Britain belong- 
ing to a certain evolutionary group, that species must be put into the 
genus, and be known by the generic name which that evolutionary 
group bears in the faunas of the world. Genera are not mere 
matters of convenience, as such authors as we have quoted seem to 
imagine, nor should they be made up of just as many heterogeneous 
or homogeneous species as the mind can readily remember. 
There is no need to go into a detailed explanation of the various 
factors which have resulted in the formation of what we call well- 
defined genera. We have already hinted as to how the extinction of 
intermediate forms may be brought about. We have already pointed 
out what we consider is the line to be adopted when genera merge 
insensibly in various ways into each other. It cannot, however, be 
too strongly insisted upon that if genera are to be rejected simply 
because they are not abruptly terminated and distinctly isolated, we 
shall only have genera remaining which have become isolated through 
purely accidental causes; and to build up generic differences on 
groupings which are fer se exceptional, and not normal, is certainly 
not the way to develop a genealogical tree which shall show natural 
relationships, but is rather an attempt to point out the isolated groups 
which are conspicuous because of the absence of close relationship 
with each other or with any other group. 
I think I have pointed out, in a crude and incomplete manner, I 
am afraid, what I consider should be the basis of natural genera. 
That it differs absolutely from the genera that we are repeatedly 
treated to by systematists is perfectly obvious. In genera, as in 
species, we may detect a certain amount of permanence and insta- 
bility combined, even in the most ideal and theoretical aspects, the 
permanence predominating in the “species,” the instability in the 
‘“oenus.” There is, as it were, a general permanence combined 
with the instability of the component parts— 
“ Still changing, yet unchanged; still doomed to feel 
Endless mutation in perpetual rest.” 
