INTRODUCTION. vii 
of the ambiens muscle and other leg muscles, and other inconstant minor 
internal features. The author of these articles unfortunately died before he 
was able to revise his early guesses as to the value of these items, and for the 
last thirty years these have been accepted at an abnormal valuation. It 
is now time to attempt a reasonable classification, but as previous schemes 
have quoted the above it is important that they should be cited. 
In connection with the skull the palate was used by Huxley as a distin- 
guishing character capable of differentiating groups of birds accurately. 
He indicated four styles of formation to which he applied the names— 
dromxognathous, desmognathous, schizognathous and egithognathous. 
Parker studied the matter more fully and added a fifth, saurognathous, at 
the same time putting on record many facts indicating the spurious nature 
of these divisions. As a matter of fact, all the forms are produced by con- 
vergence and divergence and cannot absolutely be used for the separation of 
groups. It has already been stated that none of the divisions are really 
satisfactory from the classificatory point of view by a professional anatomist, 
but we consider they are valuable, after the external structure of the bird has 
been well studied, in connection with other items. 
Garrod’s great idea was the usage of the formation of the nasal bones, 
the differences being termed holorhinal or schizorhinal. Here, again, there 
was soon seen to be pseudo-holorhiny present, and consequently the terms 
as originally proposed were proved to be futile. Once more, it will be found 
that as a subsidiary item the distinctions may be accepted as confirmatory 
evidence in many cases. 
Again, the presence or absence of basipterygoid processes was supposed 
to be of importance, but the variability of their presence seems to relegate 
these to a subordinate position. Regarded as indicating reptilian ancestry 
they recur in the most specialised avian forms and are absent from admittedly 
older groups, while in some groups they are present in the juvenile and absent 
in the adult, and in very closely allied species do not occur at all. The absence 
or presence of supraorbital grooves is sometimes quoted, but in a closely 
allied series of species great variation may beseen. The form of the pterygoids 
varies, but not much importance has been given to this character though it 
appears just as worthy of usage as some of the other items cited by osteologists. 
The quadrate is sometimes quoted, and it is a little variable so that it may 
be used later in connection with other skeletal features. The lower jaw seems 
so far to have been ignored but, we think, incorrectly. 
Owing to the exaggerated importance given to the palatal features the 
rest of the skeleton has been comparatively neglected. When extraordinary 
features were noted, through lack of knowledge these were unduly emphasised 
or unfairly minimised. The variability of the sternum may prove just as 
valuable as any feature of the skull when carefully studied. We have 
recognised phylogenetic affinity in its variation in a few cases we have recently 
studied ; the features sometimes quoted, the presence of spina externa and 
interna, do not appear to have much significance. The forms of the vertebra 
