INTRODUCTION. 9 
manner in which they were constructed and the materials of 
which they were composed, but could also see what they could 
do—could see, in fact, for what work each is intended. The task 
of arrangement now becomes immensely easier. Our previous 
classification, founded simply upon the structure of the ma- 
chines, is now supplemented and rectified by our knowledge of 
what each is able to effect. One machine is found performing 
one set of actions, another a different set; and in this way not 
only is our classification rendered much easier, but we now get 
an insight into the meaning and nature of ntany points of 
structure which were formerly obscure. 
To make this illustration fully meet the case of the naturalist 
who deals with living beings only, we have simply to suppose 
that the machines to be examined are reasonably perfect in their 
parts and fit for work, and that our imaginary workshop is sup- 
plied with areasonable amount of light, not very brilliant, perhaps, 
and striking upon some objects more sharply than on others, but 
still upon the whole moderately steady and uniform. Far worse, 
however, is the case of the naturalist who has to deal with the 
remainsof extinct generations of animals and plants, whose work 
lies amongst those relics of a bygone world which are known as 
“fossils” or “ petrifactions”—objects in many cases more won- 
derful and more perplexing and more beautiful than the most 
ornate and elaborate productions of human skill. In his case the 
workshop is a vast and gloomy vault or charnel-house, with no 
internal source of light, and but fitfully illuminated by uncertain 
gleams from the world without. And what is worse than this, 
his machines are mutilated and defaced, in many cases wanting 
their most important parts, in all cases destitute of life and 
motion, and usually very unlike anything visible at the present 
day. Nevertheless it is almost incredible with what certainty 
and precision a mere fragment of a fossil, a single tooth or 
bone, can be referred bya skilled worker in this field of science 
to its proper place in the animal kingdom—with what exacti- 
tude the missing parts can be restored—and what splendid gen- 
eralisations can be drawn from what at first sight would appear 
to be the most fragmentary evidence. 
This imaginary illustration exactly expresses the points which 
are to be regarded as essential and fundamental in classifying 
and arranging animals. We have to look, namely, jrstly, to the 
plan upon which each animal is constructed; secondly, to the 
