the most accordant with the known facts in palaeontology, geographical botany and zoology, of anatomical structure and classification:  but direct proof of it is wanting, and moreover, if true, it must have taken place very slowly; so slowly, indeed, that its effects are discernible only after a lapse of time far longer than our historic epoch.”  In contemplating the present state of the species of Cupuliferae in Europe, De Candolle comes to the conclusion that, while the beech is increasing, and extending its limits southward and westward (at the expense of Coniferae and birches), the common oak, to some extent, and the Turkey oak decidedly, are diminishing and retreating, and this wholly irrespective of man’s agency.  This is inferred of the Turkey oak from the great gaps found in its present geographical area, which are otherwise inexplicable, and which he regards as plain indications of a partial extinction.  Community of descent of all the individuals of species is of course implied in these and all similar reasonings.

An obvious result of such partial extinction is clearly enough brought to view The European oaks (like the American species) greatly tend to vary that is they manifest an active disposition to produce new forms Every form tends to become hereditary and so to pass from the state of mere variation to that of race and of these competing incipient races some only will survive.  Quercus Robur offers a familiar illustration of the manner in which one form may in the course of time become separated into two or more distinct ones.

To Linnaeus this common oak of Europe was all of one species.  But of late years the greater number of European botanists have regarded it as including three species, Q. pedunculata, Q. sessiliflora, and Q. pubescens.  De Candolle looks with satisfaction to the independent conclusion which he reached from a long and patient study of the forms (and which Webb, Gay, Bentham, and others, had equally reached), that the view of Linnaeus was correct, inasmuch as it goes to show that the idea and the practical application of the term species have remained unchanged during the century which has elapsed since the publication of the “Species Plantarum.”  But, the idea remaining unchanged, the facts might appear under a different aspect, and the conclusion be different, under a slight and very supposable change of circumstances.  Of the twenty-eight spontaneous varieties of Q. Robur, which De Candolle recognizes, all but six, he remarks, fall naturally under the three sub-species, pedunculata, sessiliflora, and pubescens, and are therefore forms grouped around these as centres; and, moreover, the few connecting forms are by no means the most common.  Were these to die out, it is clear that the three forms which have already been so frequently taken for species would be what the group of four or five provisionally admitted species which closely surround Q. Robur now are.  The best example of such a