INTRODUCTION. \' 
In the first place fossil birds are very rare ; in the second place they tell us nothing 
beyond a few facts concerning their osteology ; and in the third place the few facts 
they do tell us only make the darkness more obscure. The oldest fossil birds, of 
which numerous and nearly perfect examples have been found, were discovered in 
the Cretaceous beds of North America. Even at that remote period some birds 
had lost the keel to the sternum and acquired heteroccelous dorsal vertebrae. So 
far as we know the oldest birds had teeth, amphiccelous or doubly concave vertebrae, 
powerful wings, a keel to the sternum, four toes (three in front and one behind), 
webbed feet, a schizognathous palate, basipterygoid processes, and many other 
characters too numerous to mention. The geological record is, however, so imper- 
fect that many of these hypotheses are little more than guesses. 
Once for all it must be noted that any attempt to bring all fossil birds into the 
same system of classification as those now living is bound to fail. Between 
every two closely-allied groups of existing birds there must have been birds now 
extinct, the common ancestors of both, most probably differing from both, and 
partly resembling both, and incapable of being classified with either. To encumber 
the classification of existing birds with a few scattered links in endless chains of 
intergrading forms can only create confusion. The classification of fossil birds is a 
most interesting inquiry, and might be called the study of a vertical section of the 
bird-life which has existed in past geological ages. The classification of existing 
birds is the study of a horizontal section of the great bird mass of the world, and 
ought to form a different and distinct system confined to the horizon of the present 
time. 
If birds have all been derived from a common stock, which ean scarcely be doubted, 
“every species is in one sense of the same age; but some have developed slowly, 
possibly where the conditions of life have been easy, whilst others have developed 
rapidly where the struggle for existence has been more severe. The various races of 
man are presumably all descended from one family: but ‘‘ Quashy,” thanks to his 
“ pumpkin,” has had a very easy time of it in tropical and semitropical Africa, and 
has consequently lagged behind wofully in the race ; whilst John Bull has had a 
hard fight for existence, especially since he was crowded in his tight little island, 
and in the same time has made rapid strides. The struggle for existence increases 
the pace of development, if it be not too severe. Where the fight is too hard, as 
it has been for the Esquimaux on his arctic tundra, the fittest only just survive, or 
a less highly organized race is most adapted to struggle with such difficulties, and 
consequently the changes gradually introduced by natural selection are in a retro- 
grade direction. 
It is impossible in a work of this kind, where so many facts have to be ascertained 
