so neatly and perspicuously constructed, that the mind may comprehend it, and compare it with others, at one view; but this can hardly be done vhen the words much exceed twelve. This rule, of course, can be strictly applied to Latin definitions only, though it should be kept in view in any language, as far as the genius of that language will allow. Linnæus says, "Genuine specific distinctions constitute the perfection of natural science;" which is strongly confirmed by the great inferiority of most botanists, in this department, to that great man, and especially by the tedious feebleness and insufficiency displayed among those who court celebrity by despising his principles.
In constructing generic and specific characters, the arrangement of the different parts on which they are founded is to be considered. Such as are most important in the natural order, or genus, are to stand first, and the subordinate, or more peculiar marks of the object before us, ought to close the sentence. On the contrary, in drawing up natural characters of a genus, as well as full descriptions of particular plants, it is proper
