242 THE CANADIAN NATURALIST. [Aug. 



gerated value to their own specialties, and are unable to appre- 

 ciate those of others. Thus we find naturalists subdividing one 

 group more minutely than others, or raising one group to a position 

 of equivalency with others, to which, in the opinion of the stu- 

 dents of these others, it is quite subordinate. So also we have 

 some zoologists basing classification wholly on embryology or on 

 mere anatomical structure, or even on the functions of some 

 one class of organs. Secondly, there is a ftiilure to perceive 

 that, if there is any order in the animal kingdom, some one prin- 

 ciple of arrangement must pervade the whole ; and that our 

 arrangement must not be one merely of convenience, or of a 

 desultory and uncertain character, but uniform and homogeneous. 



The writer of these pages does not profess to be in a position 

 to escape from these causes of failure ; but as a teacher of some 

 experience, and as a student of certain portions of the animal 

 kingdom, he has endeavoured carefully to eliminate from his own 

 views the prejudices incident to his specialties, and to take a general 

 view of the subject; and is therefore not without hope that the 

 results at which he has arrived may be found useful to the young 

 naturalist. 



Classification in any department of Natural History is the 

 arranging of the objects which we study in such a manner as to 

 express their natural relationship. In other words, we endeavour 

 in classification to present to our minds such a notion of the 

 resemblances and differences of objects as may enable us to under- 

 stand them, not merely as isolated units, but as parts of the sys- 

 tem of nature. Without such arrangement there could be no 

 scientific knowledge of nature, and our natural history would be 

 merely a mass of undigested facts. 



At first sight, and to a person knowing only a few objects, such 

 arrangement may appear easy ; but in reality it is encompassed 

 with difficulties, some of which have not been appreciated by the 

 framers of systems. The more important of these difficulties 

 we may shortly consider. 



1, There are in the animal kingdom a vast number of kinds or 

 species. To form a perfect classification it would be necessary to 

 know the characters or distinctive marks of all these species. 

 To make even a tolerable approximation to a good system, re- 

 quires an amount of preparatory labour which can be estimated 

 only by those who have carefully worked up at least a few species 

 in these respects. 



