42 
SYSTEM OF NATURE. 
It seems scarcely possible to imagine a group of animals 
more uniform, more perfectly distinct, more decidedly 
marked, than birds. The separation of the marsupials from 
the placentals was a step that required much consideration, 
and when once decided on, it appeared necessary to defend 
it at considerable length, because, in their appearance, 
marsupials are essentially quadrupeds. With birds the case 
is altered : there is no child but can instantly distinguish 
between a bird and a quadruped, therefore the group is on 
all hands admitted as natural, and requires no argument in 
support of its distinctness. Were I to attempt to devise 
laws for nature instead of studying nature's laws , I should 
most certainly consider the whole of Mammalia a central 
and indivisible group, and treat birds, fishes and reptiles, 
each as a double group.^ How easy of accomplishment does 
such an arrangement appear, and how harmonious when 
accomplished ! but whoever gives due weight to structural 
difference must divide the Mammalia; whoever gives due 
weight to structural similarity must retain the birds entire. 
A dichotomous division, apparently a leading character¬ 
istic of external groups, is not practicable with birds : we 
are compelled to treat them as a group perfect in itself. 
These considerations do not, however, interfere with our 
considering birds to form the abnormal portion of an ex¬ 
ternal group, and it is in that light that I regard them. 
The position I have attempted to point out as that of the 
marsupials, requires that an abnormal group should be 
closely united to it,—a group possessing in a marked man¬ 
ner certain of its characters, yet combined with others 
widely different. Birds have the same warm blood and 
terrestrial habits as the marsupials; the same peculiarities 
in the systems of circulation and sensation ; and the mode 
