rs) INTRODUCTION. 
Mammals, the Birds, the Reptiles, the Amphibia, and the Fishes. 
In fact, our group of Quadrupeds now agrees exactly with the 
great and natural division of the Vertebrata, or vertebrate 
animals. It is true that a// vertebrate animals have not got four 
limbs, or not obviously so, but they never have more than four 
under any circumstances; and a closer examination soon shows 
us that they agree with one another in many other characters 
which are of much greater importance than the characters of 
the limbs alone. 
We have arrived, then, at the grand principle of all good 
classification—namely, that we should group together those 
objects only which are united by essential and fundamental 
points of similarity, and that in so doing we should ignore 
all mere superficial resemblances. The question now arises, 
What are these essential and fundamental points in the case of 
animals? . 
If for the moment we look at animals simply as so many 
machines, we shall not find much difficulty in answering this 
question. Let us suppose ourselves placed in a gigantic work- 
shop full of an immense number of complicated and curiously- 
constructed machines of different sorts, and asked to put them 
in order—to put those of one kind in one place, and those of 
another kind jin a different place. How should we proceed to 
act? Supposing, in the first place, that all the machines were 
at a stand-still, all that could be done would be to examine 
carefully the external form and internal structure of each, and 
to do our best to pick out some peculiarity which would distin- 
guish some from all the others. In this way, if our mechanical 
knowledge were sufficiently extensive, we should no doubt 
ultimately succeed in classing all our machines into something 
like a rough natural arrangement. We should, for instance, 
have those made on the principle of the lever in one place, 
those on the principle of the inclined plane in another, and | 
those on the principle of the pulley in a third. Still our classi- 
fication would most certainly be imperfect, and in some cases 
altogether incorrect. In some instances the parts of the ma- 
chine would be so complex as to be utterly incomprehensible, 
and in many cases our ignorance of what each was intended 
to effect would be an insuperable bar to our arriving at any 
arrangement. Suppose now, however, that all the machines 
were suddenly set in motion, so that we could see not only the 
