As to the origin and mode of introduction of successive floras, I am, for the reasons above stated, not disposed to dogmatise, or to adopt as final any existing theory of the development of the vegetable kingdom. Still, some laws regulating the progress of vegetable life may be recognised, and I propose to state these in connection with the Palæozoic floras, to which my own studies have chiefly related. Fossil plants are almost proverbially uncertain with reference to their accurate determination, and have been regarded as of comparatively little utility in the decision of general questions of palæontology. This results principally from the fragmentary condition in which they have been studied, and from the fact that fragments of animal structures are more definite and instructive than corresponding portions of plants. It is to be observed, however, that our knowledge of fossil plants becomes accurate in proportion to the extent to which we can carry the study of specimens in the beds in which they are preserved, so as to examine more perfect examples than those usually to be found in museums. When structures are taken into the account, as well as external forms, we can also depend more confidently on our results. Further, the abundance of specimens to be obtained in particular beds often goes far to make up for their individual imperfection. The writer of these pages has been enabled to avail himself very fully of these advantages; and on this account, if on no other, feels entitled to speak with some authority on theoretical questions. It is an additional encouragement to pursue the subject, that, when we can obtain definite information as to the successive floras of any region, we thereby learn much as to climate and vicissitudes in regard to the extent of land and water; and that, with reference to such points, the evidence of fossil plants, when properly studied, is, from the close relation of plants to those stations and climates, even more valuable than that of animal fossils. It is necessary, however, that in pursuing such inquiries we should have some definite views as to the nature and permanence of specific forms, whether with reference to a single geological period or to successive periods; and I may be excused for stating here some general principles, which I think important for our guidance. 1. Botanists proceed on the assumption, vindicated by experience, that, within the period of human observation, species have not materially varied or passed into each other. We may make, for practical purposes, the same assumption with regard to any given geological period, and may hold that for each such period there are specific types which, for the time at least, are invariable.