Remctrhs on the Flora Scotica of Dr Hooker. 149 
constitute more than a divisional character. Pentandria Digy- 
nia, consisting of the Umbelliferae, labours under similar objec-' 
tions : but the fault in both cases is more dependent upon the 
difficult nature of those orders^ than on Dr Hooker’s arrange- 
ment. However, this in no degree takes from the truth of the 
assertion, that where points of contrast have ‘’been chosen, they 
should be strictly adhered to, and exhibited with clearness. 
In regard to Specific Characters, the same remark acquires 
the same force : and here a deviation from order is more blame- 
able, as there are generally fewer causes of distraction, and the 
characters are more readily comprehended by one sweep of the 
eye. In the species of Poa, for instance, there is not sufficient 
adherence to regularity. The characters here might be taken 
from the panicle, spikelets, florets, culm, and root, unless a 
more compendious method should be devised -and let it not 
be imagined that specific characters are the more correct that 
they are verbose, and taken from many circumstances. Linnaeus 
confined himself to a few, and made his characters concise ; and 
though instances occur in which a greater profusion of words is 
necessary than he was in the habit of using, it should be re- 
membered that the strength of a character is in general propor- 
tioned to its concentration. The idea of limiting the number of 
words to twelve may be ridiculed, but the sense of propriety 
which led to it must not be censured ; and as Smith justly ob- 
serves, his remark, that Genuine specific distinctions consti- 
tute the perfection of natural science, is strongly confirmed by 
the great inferiority of most botanists, in this department, to that 
great man, and especially by the tedious feebleness and insuffi- 
ciency displayed among those who court celebrity by despising 
his principles.” Let us compare the characters of the species of 
Hierackim^ for example, in the Flora Scotica, some of which 
have from thirty to fifty words, with the characters commonly 
given by Linnseus ; and we will be convinced that the latter are 
preferable, and that the former, though assuming the form ra- 
ther of descriptions than of essential characters, are deficient in 
point of strength and perspicuity 
In examining the Flora of a country, one of the most impor- 
tant criteria exists in the completion or defection of the number 
