cultivated plants. Ray does not speak of the appearance of new forms, but says that a known form changes into another already existing and known form, which is the reverse of that which the theory of descent requires.
In his development of the principles of his system, among other errors we encounter one that leads to very important consequences in his application of the dictum, 'natura non facit saltus,' which he interprets as though all affinities must present themselves in a series that would be represented by a straight line, an error which has misled systematists even in recent times, and was first recognised as an error by Pyrame de Candolle. Ray overlooked the fact that the dictum holds good even when the affinities arrange themselves in the form of branching series, that is, after the manner of a genealogical tree. Much more sound is his remark, that the framing of the true system had previously been impossible, because the differences and agreements of forms were not sufficiently known; and another saying of his, that nature refuses to be forced into the fetters of a precise system, shows the dawn of the knowledge which afterwards led in Linnaeus to a strict separation of the natural and artificial systems.
It excites no small astonishment after all Ray's judicious and clear-sighted utterances on the nature and method of the natural system to find him adopting the division into woody plants and herbs; nor is the matter improved by his making the distinctive mark of trees and shrubs to be the forming of buds, that is, distinct winter buds, which is a mistake into the bargain. Yet we feel ourselves in some degree compensated for this serious error by his dividing trees and herbs into those with a two-leaved and those with a one-leaved or leafless embryo, in modern language into Dicotyledons and Monocotyledons. Ray's system is undoubtedly the one which in the time preceding Linnaeus does most justice to natural affinities. The following synopsis of his Classes will serve to show the progress made since Cesalpino. The names in brackets are
