As the discriminating characters of the Linnæan system are founded in nature and fact, and depend upon parts essential to every species of plant when in perfection; and as the application of them to practice is, above all other systems, easy and intelligible; I conceive nothing more useful can be done than to perfect, upon its own principles, any parts of this system that experience may show to have been originally defective. This is all I presume to do. Speculative alterations in an artificial system are endless, and scarcely answer any more useful purpose than changing the order of letters in an alphabet. The philosophy of botanical arrangement, or the study of the natural affinities of plants, is quite another matter. But it would be as idle, while we pursue this last-mentioned subject, so deep and so intricate that its most able cultivators are only learners, to lay aside
