^j- INTEODUCTION. 
mixed race into groups on minor structural points must outrage a natural arrangement 
based on affinity, and on the other hand a truly natural arrangement must necessarily 
o-roup together, in a very perplexing manner, individuals differing in points of structure 
and feature,-and, further, that though, on account of the long lapse of time required to 
modify an important structural detail, species differing in this respect may be safely assumed 
to be very distantly related, if at all, yet the converse, that similarity in essential details of 
structure necessarily implies close relationship, does not hold good : and in this lies a per- 
petual stumbling-block in the way of classification based on affinities; for any species that 
is capable of divergence from the type, must also be capable of convergence to it, and the 
offspring of a single pair most divergent at one time, may have again at a later period both 
had a tendency to revert to the original type and thus approach each other, and, though in 
the present generation absolutely identical, may possibly be the direct descendants of the 
most dissimilar forms ever assumed by their ancestors; so that we have really no more right 
to take for granted that two different kinds of Barbet are less remote, in derivation from a 
parent stock, from each other than a AVoodpecker is from either of them; and it is an 
assumption for which we have no warrant at all to say that one genus or family is more 
nearly related to any other particular genus or family than to the remainder, on account of 
the greater similarity in structure : and this point strikes at the root of all classification 
based on affinity. 
With reference to this it is worthy of remark that, as long as the various races of mankind 
are kept separate, each geographical district presents its own peculiar type of feature and 
form, easily to be identified, and yet comprising individuals varying widely as to form as well 
as hLbit; but as soon as new blood is introduced into any country by mterbreeding, the 
hard line of separation ceases, and a hybrid race is produced, the characteristic feature of 
which is non-uniformity— that is, that among the half-bred progeny the original types 
reassert themselves in the most aberrant manner, and with a vagueness that defies prediction 
as to the nature of the offspring from a study of the parents. This view of the case would 
lead us to infer that in classification feature is a surer guide than habit, an almost certain 
guide in a pure species, but deteriorating greatly in value with a mixed race, and, secondly, 
that geographical distribution is effective in modifying both colour and feature. The 
vagueness of our knowledge of the past history of the animal kingdom renders it impossible 
to°diBtinguish clearly between cause and effect; and naturalists are thus enabled to take 
whichever seems best adapted to bear out their particular views : one, assuming habit as 
the cause, explains by this means peculiar modifications of structure otherwise unaccount- 
able ; another, equally anxious to arrive at the truth, assumes the points of structure as the 
cause, and points out the habit as a natural consequence. It will be only by observations 
extending over long series of years that any definite conclusion on these points will be 
obtained ; and for the present (in this particular the scientific world can never too highly 
appreciate the services of Mr. Darwin in his life-long series of careful experiments) orni- 
thologists cannot perhaps be better employed than in clearing up the existing confusion in 
synonymy, and thus paving the way for a comparison of the results of observations in all 
parts of the world. 
