x INTRODUCTION. xill 
The binomial name will probably generally be used as a contraction ; but 
it must never be forgotten that it is only a contraction. The difference 
between a species and a subspecies, though in some cases not very clear, 
is far too important a fact to be sacrificed to a craze for a uniform 
binomial nomenclature. 
The grouping of individuals into subspecies and species is the first 
step in 
Tur CLASSIFICATION OF BrIRDs. 
The second step is to group species into genera and subgenera ; the third 
is to group genera into families and subfamilies, and the fourth to group 
families into orders and suborders. The use of the terms subspecies, 
subgenera, &c. implies that all these divisions and subdivisions are more 
or less artificial, and that our systems of classification attempt to draw a 
hard-and-fast line where Nature has drawn none, or only a few here and 
there. 
Looked at from one point of view, Nature may, however, be said to have 
drawn some very hard-and-fast lines. If it were possible to examine every 
species of bird which exists or has existed, we might find that all birds 
were descended from one common ancestral species, and that, consequently, 
every species of bird was connected with its nearest allies by an unbroken 
series of intermediate forms; in which case we should be obliged to admit 
that there was only one species of bird, divisible into an immense number 
of subspecies. Or we might find that birds are descended from several 
ancestral bird-reptiles (so to speak), and that consequently there were 
several species of birds, each divisible into an immense number of sub- 
species. We have, however, only to deal in our classification with existing 
species ; and we at once perceive that by the extinction of species and 
genera, to say nothing of families and orders, Nature has drawn some very 
hard-and-fast lines, sometimes only narrow lines, but in many cases very 
broad ones. 
When we come to deal with genera, the artificial character of our 
classification at once reveals itself. The old-fashioned notion that species 
were separated by differences of colour, and genera by structural differences 
(that is, difference in the shape of the bill, feet, wings, or tail), is a pre- 
Darwinian ornithological superstition, which is pure theory, and is entirely 
unsupported by facts. There is no evidence of any kind that the leopard 
can change his spots in a shorter time than it takes him to change the 
shape of his skull. On the other hand, there is strong evidence to prove 
that in many genera of birds colour or pattern of colour is more constant 
than many of the so-called structural characters. The principal causes of 
the change of colour in birds are supposed to be to ensure protection from 
enemies and to please the taste of the females, whilst the changes in the 
