in Natural History. 485 



neric and other abstract terms. In mathematics they admit of 

 exact (or I would rather say more exact) previous definition ; 

 and hence arises the accuracy of deductions the most recondite 

 and remote in that science. In the other sciences, which are of 

 a speculative and contingent nature, these terms are employed 

 not with the same precision, but seem to be the result of our 

 necessities, borrowed from sensible objects and analogy, and fre- 

 quently indeed from accidental coincidences. They derive their 

 force rather from the character of the mind that employs them, 

 than from any exact definition they may have received ; and it 

 seems impossible to make men use such words in a common ac- 

 ceptation. Hence it is, I apprehend, that knowledge of a specu- 

 lative kind so soon finds its limits ; and where at its outset it has 

 promised such glorious results to mankind, as long as it floated 

 in general propositions, the same subject eludes the grasp of the 

 human faculties when it is attempted to be reduced to exactness, 

 and leaves something always to be desired. We are constantly 

 approximating to the truth, yet never reaching it. 



It is sometimes asserted, but not correctly, that Natural Hi- 

 story, by the aid of its terms, partakes of the nature of mathema- 

 tical truth ; or that it lies intermediate between that science and 

 speculative knowledge. The situation of the naturalist is rather 

 this. He finds himself placed amidst an infinite number of un- 

 known particulars ; and in order to facilitate an acquaintance 

 with them, he at once, without regarding individuals with much 

 minuteness, throws together a number of them, which he calls a 

 species, according to an assumed hypothesis. These he attempts 

 again to combine by certain external characters, and calls them 

 a genus. By these means he is enabled to contemplate and treat 

 of them, without being utterly bewildered in the labyrinth of 

 unarranged individuals. Classification is his fihim Ariadneiun. 

 It was but imperfecdy understood by the ancients; and has 



enabled 



