138 DISCOURSE ON THE STUDY 



mostly where the species to be distinguished are 

 separated from each other chiefly by difference in 

 degree, of certain qualities common to all, and 

 where the degrees shade into each other insensibly. 

 Perhaps such subjects can hardly be considered 

 ripe for systematic nomenclature ; and that the at- 

 tempt to apply it ought only to be partial, embracing 

 such groups and parcels of individuals as agree 

 in characters evidently natural and generic, and 

 leaving the remainder under trivial or provisional 

 denominations, till they shall be better known, and 

 capable of being scientifically grouped. 



(132.) Indeed, nomenclature, in a systematic point 

 of view, is as much, perhaps more, a consequence 

 than a cause of extended knowledge. Any one 

 may give an arbitrary name to a thing, merely to 

 be able to talk of it ; but, to give a name which 

 . shall at once refer it to a place in a system, we 

 must know its properties ; and we must have a 

 system, large enough, and regular enough, to receive 

 it in a place which belongs to it, and to no other. 

 It appears, therefore, doubtful whether it is de- 

 sirable, for the essential purposes of science, that 

 extreme refinement in systematic nomenclature 

 should be insisted on. Were science perfect, in- 

 deed, systems of classification might be agreed 

 on, which should assign to every object in nature 

 a place in some class, to which it more remarkably 

 and pre-eminently belonged than to any other, and 

 under which it might acquire a name, never after- 

 wards subject to change. But, so long as this is not 

 the case, and new relations are daily discovered, 

 we must be very cautious how we insist strongly 



