254 ELEMENTS OF ORNITHOLOGY, 
small, it is especially difficult to make this discrimination. 
Birds, moreover, are creatures which leave behind them com- 
paratively few fossil remains, while every one now sees that the 
number of species which have become extinct must be enormous. 
The best Ornithologists, those even who are the most ardent 
evolutionists, have come to despair of being ever able to deter- 
mine satisfactorily what the exact genetic relations of different 
groups of Birds really are. We think it then not only the wiser 
course, but the only course consistent with the interests of the 
student of science, to abstain from making any positive asser- 
tions as to the genetic affinities of different Bird-groups. On 
the other hand, we do not by any means deny the truth of the 
theory which would ascribe real blood- relationship to different 
groups of Birds. We desire to keep an open mind with respect 
to questions of this kind, and we would advise our readers to do 
the same. In the meantime we wish to avail ourselves of the 
most recent labours of Ornithologists in this cause, and to give 
the greatest weight to characters which may fairly be supposed 
to indicate real relationship by descent. Characters of the 
kind would have been regarded as essential and fundamental 
ones even before the theory of evolution became popular— 
we mean characters derived from the form of the skeleton and 
other anatomical peculiarities and from the mode of the process 
of the development of the young. 
We may thus legitimately speak of “real affinity,” “true 
relationship,” and ‘ essential connection,” as existing between 
certain Birds, whether or not a real genetic affinity exists 
between them. If such genetic affinity does exist between 
such Birds, then such a mode of speech has a plain and obvious 
truth and fitness. But if such genetic affinity does not exist 
between them, then such expressions must be understood to 
denote resemblances such as were recognized as being of a 
‘“‘ deeper ” kind than some other resemblances, before questions 
of descent had begun to be discussed. A ‘ deep resemblance ” 
of this kind is one which is the sign of a great many other 
resemblances, whereas a ‘superficial resemblance ” has no such 
significance. 
This great uncertainty as to the full significance of characters 
which no one can call ‘ superficial,” causes the classification of 
Birds to be a more or less arbitrary one. The arbitrary nature 
of Ornithological grouping’is intensified by the habit which has 
so long and widely prevailed of not even attempting to define the 
groups by any constant and universal anatomical characters. 
