With regard to the conditions of deposition, Professor Lesquereux remarks, that the character of the leaves found in the Dakota group, and their analogy with species of our time, seem at first to refer them to a dry land flora; this is, however, not positively the case. The most abundant representative of the Cretaceous flora, the Sassafras, is remarkably similar to the present Sassafras officinalis, which inhabits every kind of ground and station, from the dry hills of Ohio to the low swamps of Arkansas. The numerous leaves of Laurus are comparable to those of L. caroliniana, a shore plant; as are also the Platanus, Magnolia, Populus, Salix, Menispermites; the essential types of the Dakota group being, therefore, those of low islands. or low shores, rather than hills or dry land. Professor Mudge says the characteristic of the local deposits of this group in the State of Kansas indicate that the forests were on small islands scattered over the Cretaceous ocean.*

In Cumberland and Nelson provinces, New Zealand, rocks referable to the Cretaceous age by their fossils overlie valuable coal-seams, locally altered into anthracite, and associated with the shales and sandstones is a rich dicotyledonous flora, both angiosperms and gymnosperms, as the Myrtaceæ, Fagus, Coprosma, Dammara, Phyllocladus, Podocarpus, Dacrydium, &c., many of the genera still existing in the New Zealand flora. There is one locality at Pakawu, where the genera Pecopteris and Tæniopteris are clearly associated with the dicotyledonous plants, the species however being different from those representing the same genera in an underlying sandstone, which is undoubtedly of Jurassic age, and which species are mostly identical with those found in the Rahmajal series of India.

Overlying the coal-bearing rocks are a series of conglomerates, sandstones, calcareous marls and chalk, with flints in some localities more or less developed, and there is evidence in the upper part of this series of fossils with an Eocene facies, succeeded by lignite beds indicating a land surface.†

The Cretaceous period may be divided into two distinct. periods according to the nature of its terrestrial vegetation, and these somewhat agree with the ordinary geological divisions; including the Wealden, Neocomian, and Gault in the lower; and upper green sand, lower, middle, and upper chalk in the upper division.

M. Schimper has divided the Cretaceous flora as follows: Etage Intérieur (Neocomien, Urgonien), Etage Moyen et Supérieur (Aptien to Danien).

The general Cretaceous flora includes: Thallogens, as Algæ; Acrogens, as ferns and equisetum; Gymnogens, as conifers

✦ "Trans. Kansas State Board of Agriculture," p.: 95.

† Dr. Hector, "Trans. New Zealand Institute," vols. ii. and vi.

and cycads, a few monocotyledons, as palms; dicotyledons, as the ordinary forest trees; but it is the different distribution of these classes that forms the marked feature of the two divisions. In the lower group, the gymnogens and cryptogams are abundant; in the upper, the dicotyledons are most numerous, with some conifers and ferns. This distribution has some relation to the floras preceding and succeeding the Cretaceous period. Thus the preceding Jurassic flora is composed of ferns, a few Equisetaceæ, some Coniferæ, and a great abundance of Cycadeæ. Three-fourths of all the fossil Zamic and half of the Cycadeæ, known from all the geological formations, are Jurassic. In the lower Cretaceous of Wernsdorf and Greenland, especially the latter, Heer finds a marked proportion of this family which is scarcely represented in the upper. Of the Conifera the lower Cretaceous rocks (Gault) contain the earliest known representative of Sequoia; this genus is interesting, not only as a persistent type ranging from the Cretaceous to the present, but from its wide distribution at former geological periods, i.e. the upper Cretaceous and Miocene, and being now represented by twc species only (the Redwoods), and those restricted to a comparative limited area in North America, the S. gigantea (Mammoth tree or Wellingtonia) to the western flank of the Sierra Nevada, and S. sem, pervirens along the coast range from the bay of Monterey to the frontiers of Oregon.* The absence of dicotyledons, the presence of a few monocotyledons, and the abundance of the Cycadeæ. approximate the lower Cretaceous to the previous Jurassic flora.

On the other hand, the Upper Cretaceous presents a distinct assemblage, and affords the first well-marked proof of the introduction on the earth of a vegetation allied to our fruit and forest trees. Among the ferns are the living genera Gleichenia, Lygodium, Asplenium, together with the earlier forms Sphenopteris and Pecopteris; scarcely any Zamiæ, but some Coniferæ, as Abietites, Phyllocladus (?) Sequoia, and its near relative Glyptostrobus, now living in China. Of the numerous genera of dicotyledons are the fig, willow, poplar, beech, oak, plane, sweet-gumtree (Liquidambar), sassafras (abundant in the Dakota group), Magnolia, Liriodendron (tulip-tree), some Proteaceæ, and other living and extinct genera, including Credneria, many of them having at present a very different geographical distribution. Among others, the Magnolia and Liriodendron belong to North America by origin, succession, and presence. Of the eight species

* See Lesquereux, "Cretaceous Flora," p. 116; and Professor A. Gray, "Address to the American Association at Dubueque, Iowa," Aug. 1872.

† Lesquereux remarks that "the essential types of our actual flora aro marked in the Cretaceous period, and have come to us after passing, without notable changes, through the Tertiary formations of our continent."

of true Magnolia now known to botanists, seven belong to the western slope of the temperate zone of North America. Liriodendron, the Tulip-tree, has in its characters, its distribution, and its life, a great degree of affinity with Magnolia; the American species is the only one now known in the vegetable world, and its habitat is strictly limited to America. Either considered in its whole or its separate characters, the tulip-tree is a constant subject of admiration and wonder. It could be named, not the king-it is not strong enough for that—but the queen of our forests, if the Magnolia was not there with it to dispute the prize of perfection by the still grander majesty of its stature, the larger size of its foliage, the elegance and perfume of its flowers. Our sense of admiration for these noble trees is heightened still by the dignity of their ancient origin.*

The upper Cretaceous rocks, therefore, by their numerous forms of dicotyledons strongly foreshadow the Tertiary and succeeding floras. As partly bearing on this point, Professor Lesquereux states, as regards the Western Territories, "In ascending from the lower lignitic measures, where the essential types of the Cretaceous flora (the Dakota group) have no representatives, we see these Cretaceous types re-appearing, a few in the Upper Eocene, more, in the Carbon group above, still more in the Upper Tertiary, following thus an increasing degree of predominance, culminating, it seems, at the present time, in the flora of the eastern slope of the North American continent."†

A character of the Cretaceous floras is their apparent want of homogeneity; even when probably synchronous, they are so diversified, when compared to each other, as to appear not to belong to the same epoch and the same country. On this point MM. Saporta and Marion remark, in alluding to the localities which have been carefully studied, "What point of analytical connection can be established between Niederschoena in Saxony, Moletin in Moravia, Quedlinburg and Blankenburg in the Hartz, Halden in Westphalia, the sands of Aix, the Senonian of Bausset, the Santonien of Fuveau, and the North American Cretaceous of Nebraska?" Professor Lesquereux also shows that the Dakota flora, with scarcely any forms referable to species known from coeval formations of Europe, presents in its whole a remarkable and as yet unexplained case of isolation.

Contemporary, therefore, with the extensive marine deposits of this period-some of a peculiar nature, great thickness, and uniform character, as the white chalk, indicating deep-sea conditions-there were evidently land surfaces either distributed as islands in the cretaceous ocean, or forming more or less low • Lesquereux, "Cretaceous Flora," p. 121.

† Ibid. p. 117.

Essai sur l'état de la végétation à l'époque de marnes heersiennés de Gelinden." Par Count Saporta et Dr. F. A. Marion. P. 14.

shores bordering the same, upon which grew a vegetation varying in character, and sometimes localized, according to climatal conditions; in some places sufficiently luxuriant to become the source of future beds of coal; in other places, to be carried by rivers or other means and imbedded with the marine remains

of the adjacent sea. This vegetation was not, however, of uniform character throughout the Cretaceous period, but became modified and varied in the various stages, and with a marked difference in the lower and upper periods. In the lower, with the incoming of a new marine fauna, the plant-life resembled that of the previous Jurassic age; whilst in the upper, with a comparative similar aspectal fauna, the cotemporary vegetation was different from that of the lower Cretaceous period, and presented a land facies resembling that characteristic of the subsequent Tertiary period, in which the forms of the cotemporary marine life were entirely changed.

In recapitulating the evidence derived from our present knowledge of the distribution and character of the Cretaceous flora, it appears-1. That in the different localities from which the plants have been described, while the floras have some points in common, they are more or less localised in character, probably in consequence of different physical and climatal conditions. 2. That the Upper Cretaceous flora differs considerably from the Lower, inasmuch as the latter, by the predominance of the gymnosperms, is related to and continuous with the previous Wealden and Jurassic flora; while the former, containig a few of the older forms, is marked by the abundance of the dicotyledons, thus foreshadowing the subsequent Tertiary flora. 3. That while in some areas the Cretaceous flora yields generic forms, which are now represented and living in the same area— as for example in New Zealand, Europe, and America—in other places, as in Greenland, there are no such living representatives; and again, some genera, as Sequoia, which were widely distributed during the Cretaceous period, are now restricted to a single locality. 4. That the appearance of a phænogamous flora in the upper Cretaceous rocks affords a parallel to the occurrence in the same formation of the teleostian fishes, both of which classes not only increase in the subsequent Tertiary period, but are the dominant forms at the present day, as shown in the accompanying Pl. CXXX.

These occurrences may indicate either distinct creations, adapted to the then existing inorganic conditions, or, when our knowledge of the earlier forms is more extended, it may be shown to have resulted by modification during long periods of time of previous generalised or ancestral types, of which there is some evidence in the underlying Jurassic and still older formations.