in Natural. History. 495 
mals set at naught all our classification by their osteology arid 
mode of bringing forth. | 
The adoption of these numerous terms, intended to express 
fixed ideas, must be looked on with suspicion. The terms 
species and genus are too well established by custom, and are 
so clearly the result of convenience, and moreover conform so 
closely to the ordinary use of these words, that their utility 
cannot be questioned ; but those numerous subdivisions current 
among our neighbours, and sensibly increasing among ourselves, 
may well be doubted as unphilosophical language. To each of 
them is attempted to be assigned a definite value beforehand, 
and an impracticable degree of precision ; and we deceive our- 
selves by fancying that we can deal with these delicate and 
fleeting instruments of thought differently from the rest of the 
world. But are we to attempt to fetter nature by our systems 
and terms ? ** Books should follow sciences, not sciences books," 
says the immortal Bacon; yet the adoption of systems and 
technical expressions, which have received their definition be- 
forehand, cannot be employed without the danger of perpetuating 
false hypotheses, and an apprehension on the part of the ignorant, 
that these inventions give us some power over nature not be- 
longing to ordinary language. 
The more correct mode would be to exclude from the natural 
method most of these terms, and to employ in their place some 
convertible words of looser import, as indeed M. Cuvier has to 
some extent done; such for instance, as group, section, division, 
to express those larger assemblages of approximations to as- 
signed forms, which are rather predicated than proved; and in 
many cases to point them out by mere signs, such as are used 
in printing. Thus, for instance, the word section, or any similar 
word, might be employed to express the plants severally com- 
prehended in the order Graminee, the class Composite, and the 
VOL. XV. 3s division 
