^1';

nbsp;

i

UNIVERSITEITSBIBLIOTHEEK UTRECHT

3200 662 3

DEPARTMENT OF THE INTERIOR.

UNITED STATES GEOLOGICAL SURVEY

THE TERRITORIES

Bibliotheek

Instituut voor aardwetenschat^pe� Rudapestlaan 4nbsp;� CO Utrecht

F. V. HAYDEISr,

UNITED STATES GEOLOGIST-IN-CHAEGB

BIBLIOTHEEK OER RIJKSUNIVERSITEIT

UTRECHT

Department of the Interior,

United States Geological Survey,

November 1, 1883.

On the 27th of September, 1882, at the request of Dr. F. V. Hayden, the completion of the publications of the United States Geological andnbsp;Geographical Survey of the Territories, formerly under his charge, wasnbsp;committed to the charge of th� Director of the Geological Survey by thenbsp;following order from the honorable the Secretary of the Interior:

Department of the Interior,

Washington, September S7, 1882.

Maj. J. W. Powell,

Director U. S. Gteological Survey, Oily:

Sir: The letter of Prof. F. V. Hayden, dated June 27, bearing your indorsement of July 20, relating to the unpublished reports of the surveynbsp;formerly under his charge, is herewith returned.

You will please take charge of the publications referred to in the same, in accordance with the suggestions made by Professor Hayden.

It is the desire of this office that these volumes shall be completed and published as early as practicable.

Very respectfully,

H. M. TELLER,

Secretary.

Of the publications thus placed in charge of the Director of the Survey, the accompanying volume is the second to be issued. The firstnbsp;was entitled �The Vertebrata of the Tertiary Formations of the West, bynbsp;Edward D. Cope.� On the 12th day of October, 1882, the manuscript ofnbsp;the present volume was received at the office of the Geological Survey, andnbsp;through the hearty co-operation of Professor Lesquereux, the work hasnbsp;been pushed to rapid completion. The volume is an important contribution to the ancient botany of North America, and will be heartily welcomednbsp;by paleontologists.

Director.

iii

LETTER TO THE SECRETARY.

Washington, November 1, 1883.

Sir: I have the honor to transmit, for your approval, the eighth volume of the final reports of the United States Geological and Geographical Survey of the Territories, prepared by the eminent paleontologist. Prof.nbsp;Leo Lesquereux.

A brief synopsis of the contents of the volume may be given as follows:

In the first part�the Cretaceous Flora�are described a large number of new species, some representing rare and very remarkable types, all ofnbsp;which are figured on the first seventeen plates. Besides the descriptionnbsp;of the species, there are some general remarks on the geology of thenbsp;Dakota group, and on the character of the plants in regard to climate andnbsp;their affinities with plants of succeeding geological periods. A table ofnbsp;distribution is added, enumerating all the species known up to the presentnbsp;time, pointing out the relations of the plants of Europe and various partsnbsp;of North America with those of the Dakota group in Nebraska, Kansas,nbsp;and Colorado. The number of species enumerated in this table is 443,nbsp;of which 200 are from the Dakota group.

The second part contains a revision of the plants of the Laramie group. The introduction considers the relations of these plants to thosenbsp;of Europe, for the purpose of fixing the age of the formation. Thennbsp;follows a description of a few new species from very fine specimens onnbsp;three plates, and a table of distribution including only the species of thenbsp;Laramie group, which in the seventh volume of the series were mixednbsp;with those of the other stages of the Tertiary and were not grouped clearlynbsp;enough for the proper appreciation of the general characters of the flora.

Up to the present time the author has been unable to find a single species that he could identify with any from the Dakota group. He hasnbsp;now in his possession very large collections of plants from this group,nbsp;which have not been reported upon, collected in Colorado and Wyoming;nbsp;yet after a careful examination he fails to find any form even related tonbsp;those of the Dakota group.

The third part reviews the flora of the White and Green River regions, which he separates into two groups. The plants of Green River and Alkali

LETTBE TO THE SECEETAEY.

Stations and Randolph County, Utah, are most of them different from those of Florissant, Mouth of White River, and Elko. These plants are represented by twenty-one plates, and their relation is indicated with the flora ofnbsp;the Gypses of Aix in France, which is generally regarded as lowest Miocenenbsp;or Oligocene.� The table of distribution of these plants includes, in America,nbsp;those of Florissant, Elko, Green River Station, Alkali Station, Sage Greek,nbsp;and Barrel! Springs as compared with the Miocene of Greenland, Alaska,nbsp;the Oligocene of France and Germany, and the Miocene of Europe,

The fourth part relates to Miocene plants described from specimens obtained from the Bad Lands, California, and Oregon, and from Alaska,nbsp;and they occupy fifteen plates. There is also a table of distribution thatnbsp;indicates the relations of these species of Alaska, Carbon, Washakie, thenbsp;Bad Lands, Oregon, California, and Fort Union with the Arctic Miocene,nbsp;Greenland, Spitzbergen, and those of Europe. This eighth volume formsnbsp;a kind of supplement to the two preceding volumes, inasmuch as in it arenbsp;figured and enumerated all the plants which have been found since theirnbsp;publication, in the formations of the Mesozoic and Genozoic periods ofnbsp;North America, and therefore forms a broad basis in vegetable paleontology for the direction of future researches and the classification andnbsp;determination of the fossil flora of the Continent. The three volumes ofnbsp;this series, on vegetable paleontology, form a grand monument to thenbsp;industry and fame of the author.

I take pleasure in acknowledging my obligations to the Director of the U, S. Geological Survey, who has with great kindness superintended thenbsp;printing of this Report.

The plates were engraved by the well-known firm of Thomas Sinclair amp; Son, of Philadelphia, and are fine examples of their work.

I have the honor to remain, with great respect, your obedient servant,

F. V. HAYDEN,

United States Geologist.

To the Honorable the Secretary of the Interior.

FNITED STATES GEOLOGICAL SURVEY OF THE TEEiaTOEIES.

WESTEEN TEEEITOEIES.

Part III.

THE CRETACEOUS AND TERTIARY FLORAS.

By LEO LESQUEREUX.

Page.

XoTE OF Maj. J. W. Powell, Director of the Geological Survey�...................... iii

Letter of Dr. F. V. Haydex to the Secretary of the Ixterior........................... v

Letter of ................................................................................... xi

Introduction.................................................................................. 1

I. �The Flora of the Dakota Group........................................................ 2

General remarks........................................................................ 2

Description and enumeration of species of the American Dakota Group formation................ 25

Table of distribution of the plants of the Cretaceous Cenomanian formation...................... 93

The relationship of the flora of the Dakota Group.......-.............................-..... 105

II. �The Flora of the Laramie Group...................................................... 109

Table of distribution of the species of the Laramie Group..................................... 115

Description of species added to the flora of the Laramie Group................................ 121

III. �The Flora of the Green River Group................................................. 127

Geological distribution of the measures...................................................... 127

Enumeration and description of the species of fossil plants known from the Green River Group � 135

General remarks......................................................................... 205

Table of distribution of the plants of the Green River and White River Groups.................. 206

Relationship of the local groups indicated by correlation of species............................. 213

IV. �The Miocene Flora...................................................................... 219

Description of the Miocene species from specimens obtained in the so-called Bad Lands of Dakota. 221

Description of Miocene species of California and Oregon...................................--- 239

Contribution to the Miocene flora of Alaska.................................................. 257

Species of plants from the Chalk Bluffe of California.......................................... 265

Table of distribution of the North American Miocene fossil plants............................. 266

Remarks on the species of Miocene plants...................................................273

Index........................................................................ 279

Descriptions of plates.........................................................1____follow page 283

LIST OF ILLITSTKA-TIOISTS.

Plates I-XVIII.�Fossil Plants from the Dakota Group�Cretaceous.

Plates XIX-XX.�Fossil Plants from the Laramie Group.

Plates XXI-XLV (A).�Fossil Plants from the Green River Group.

Plate XLV (B).�Fossil Plants from the Chalk Blufls of Nevada County, California. Plates XLVI-XLIX.�Fossil Plants from the Bad Lands of Dakota.

Plates L-LIX.�Fossil Plants from the Miocene of California and Oregon.

LETTER OF TRANSMITTAL.

Columbus, Ohio, September 30, 1882.

Dr. F. V. Hayden, Philadelphia.

Dear Sir: I send herewith the manuscript of the eighth volume of the Reports of the United States Geological Survey of the Territories, madenbsp;under your direction. Besides a short introduction, this volume contains:

1st. A review of the Cretaceous Flora of the Dakota Group, or of what has been published in volume VI, with descriptions of a large number of new and remarkably interesting species illustrated by 17 plates.-

2d. Some remarks on the Flora of the Laramie Group, which I consider as Eocene, with descriptions of a few new species, illustrated by 3 plates.

3d. The more valuable part of the volume, viz: the descriptions of the plants of the Oligocene, a flora of which little was known before, andnbsp;which is now richly represented by a large number of specimens, especiallynbsp;from Florissant, Colorado. This Flora will be quite as well received bynbsp;paleontologists as has been the Cretaceous Flora of volume VI. It isnbsp;illustrated by 24i plates, which are all very finely made.

4th. Half of one plate serves for illustrations of a few plants from the oldest Pliocene, or upper Miocene of California.

5th. Descriptions with figures of Miocene plants of the Bad Lands, with 5 plates. The plants, clearly of Miocene type, are very interestingnbsp;from their relation to species of the Arctic Flora.

6th. Descriptions of species of Miocene plants of California and Oregon from specimens pertaining to the State Museum of Oakland, California.nbsp;They are illustrated by 10 plates, the whole number of the plates being 60.

7th. A short account and description of new species found in a collection of fossil plants made in Alaska by W. H. Dali, of the United States

LETTEE OF TEANSMITTAL.

Coast Survey, for the Smithsonian Institution. The specimens were sent to me for determination, and si was allowed to give in volume VIII a shortnbsp;description of the new species added to the Alaskan Flora already partlynbsp;known hy the works of Heer. These new species have heen figured innbsp;the Proceedings of the National Museum, vol. v, pi. vi-x.

It is not unnecessary to remark that all the plants described in volume VIII are considered in separate groups according to their relation to the age of the formation which they determine. Comparisons are established with the European Floras by tables of distribution, etc.

I truly believe that this volume will prove to be a very valuable contribution, not merely to the paleontology but also to the geology of this country.

Very truly and respectfully yours,

LEO LESQUERE�X.

CRETACEOUS AND TERTIARY FLORA.

By Leo Lesquereux.

INTRODUCTION.

The present volume contains:

1st. The materials referable to the Cretaceous Flora.

The species recognized from specimens received since the publication �of the Annual Report of Dr. F. V. Hayden, 1874, are of course describednbsp;here, but it has been found advisable to add to them and to consider againnbsp;part of what has been published in that report as a Review of the Cretaceous Flora of North America; mentioning also the species describednbsp;by Professor Heer and Dr. Newberry from specimens obtained from thenbsp;Dakota Group.

It is well known that the plants of the Cretaceous epoch, at least those of a higher class, the Dycotyledons, have been barely discovered andnbsp;described in Europe, while the profusion of these vegetables in the Dakotanbsp;Group constitutes an original illustration of a peculiar vegetation which,nbsp;for reasons explained hereafter, will be of great significance in the future.nbsp;From this consideration the exposition, in the same work, of all that isnbsp;known to this time of the North American Cretaceous Flora is greatly to thenbsp;advantage of vegetable paleontology both in this country and in Europe.

2d. A description of a few species of plants of the Laramie Group, which I persist in considering as Eocene.

These species, added in this volume to the list of the plants already described from the same formation, were all obtained at Golden, Colorado,nbsp;from the locality where most of those published formerly were found bynbsp;myself. One, Oreodoxites plicatus, a fine Palm, represented by a number ofnbsp;well-preserved though more or less fragmentary leaves, is of a peculiar type,

2 nbsp;nbsp;nbsp;CEETACEOUS AND TEETIAEY PLOEA.

and finds its affinity only in Ludoviopsis geonomcefolia, Sap., of the Eocene of S�zanne, A second, Sterculia modesta, Sap., also of S�zanne, is represented by a beautifully preserved specimen whose identity has been recognized by the author. A third, Aralia pungens, is remarkable for its verynbsp;close relation, perhaps identity, to four species described by Massalongonbsp;as Sylphidium from the Eocene of Italy. And still a fourth, ZizypJius Beck-withii, is evidently allied to Z. Harcourtii of S�zanne. These, on sevennbsp;species only, added to the flora of the Laramie Group, tend to confirm thenbsp;conclusions which I have admitted on the age of the flora of the greatnbsp;Lignitic, or Laramie, Group.

3d. A large number of species described from what I called in Volume VII the Green River Group No. 4, which I considered as probably Miocene.

When that volume was published this flora was known only by a very few species. Since that time a large number of specimens have beennbsp;procured from the same formation, especially at Florissant, Colorado. Thenbsp;species which they represent are very interesting as indicative of a geological period older than the Miocene, or preceding in age the Carbon andnbsp;Alaska floras.

4th. A new contribution to the Miocene Flora from specimens procured from various localities of the Bad Lands of California and Oregon, with mention of new species recently obtained from Alaska, and a note uponnbsp;a few specimens from the Chalk Bluff of California, a Pliocene formation.

I.�THE FLORA OF THE DAKOTA GROUP.

GENERAL REMARKS.

All that refers to the geology of the Cretaceous Dakota Group�its immediate superposition upon rocks of Permian age; its relation to thenbsp;strata overlying it in an uninterrupted series of marine deposits up to thenbsp;base of the Tertiary measures; its thickness, the superficial expanse of itsnbsp;area�^has been recorded in the general remarks of Volume VI of thesenbsp;reports. Since that time very little has been added to what was knownnbsp;and published on the subject.

One fact only should be mentioned now. It is the discovery of numerous specimens of Cretaceous plants at the base of the Rocky Mountains in

INTRODUCTION. nbsp;nbsp;nbsp;3

Colorado. The plants, by the identity of a number of them and the close affinity of character of some others with species of the Dakota Group,nbsp;have positively confirmed the supposition that this formation, passingnbsp;westward in Kansas under the Tertiary measures, is prolonged under themnbsp;and continues to the Rocky Mountains.

Already, in 1873, Dr. A. C. Peale had procured from Colorado fragments of poorly preserved leaves which had been recognized as identical withnbsp;Proteoides acuta, Heer, a species commonly found in the Dakota Group ofnbsp;Kansas and Nebraska. From this. Nos. 14-16 of the section of Southnbsp;Platte RiveP had been then considered by Dr. Hayden as referable to anbsp;Cretaceous formation. More recently. Passed Assistant Engineer H. C.nbsp;Beckwith, United States Navy, and Rev. Arthur Lakes, have got, nearnbsp;Morrison, a few miles west of Denver, numerous specimens of some of thenbsp;more predominant species of the Dakota Group�Sassafras {Araliopsis)nbsp;cretaceum, Magnolia Capellini, Aralia, Salix protecefolia, etc., with somenbsp;others, which though new are related species which tend to identify thenbsp;Cretaceous formation at the base of the Rocky Mountains with that ofnbsp;Kansas. Admitting, therefore, the prolongation of the Dakota Groupnbsp;under the Tertiary measures to the base of the mountains, the width ofnbsp;the area covered by this formation should be estimated from east to westnbsp;at 450 to 500 miles.

Perhaps, also, I should omit here any remarks on the flora of the North American Cretaceous as represented by the plants of the Dakotanbsp;Group, having already, in Volume VI of the United States Geologicalnbsp;Survey of the Territories, by Dr. F. V. Hayden, considered the generalnbsp;character of this flora and its relation to plants living at our time, or tonbsp;analogous or identical species observed in the formations succeeding thatnbsp;of the Cretaceous. But the materials which I had then for considerationnbsp;were few and local; they have since been greatly increased, and also newnbsp;points for comparison have been furnished to phytopaleontologists by thenbsp;works of Heer on the recently discovered Cretaceous plants of Greenland.nbsp;Fr-om this, some of the conclusions formerly admitted have been morenbsp;or less modified, while others have received a higher degree of precision

� Dr. F. V. Hayden, Annual Report, 1873, pp. 195, 196.

OEETAOEOUS A�fD TEETIAEY FLOEA.

if not of actuality. It is thus advisable to look again over what is known to the present time of the characters of the North American Cretaceousnbsp;flora and to record the deductions legitimately derived from that knowledge.nbsp;This kind of work is a necessity for the present, as it will be also for thenbsp;future, not only because what is known now is, probably at least, a merenbsp;fraction of the elements constituting the North American Cretaceous flora,nbsp;but because the determinations of the plants are still and must be for anbsp;long time to come unreliable to a certain degree.

The plants of the Dakota Group, as known mostly by detached leaves, are striking from the beauty, the elegance, the variety of their forms, andnbsp;from their size. In all this they are fully comparable to those of any geological epoch as well as to those of our time. From entirely developednbsp;leaves, less than one inch in size, they show all the gradations of size tonbsp;one foot, even to a foot and a half in diameter. The multiplicity of formsnbsp;recognized for a single species is quite as marked as it might be upon anynbsp;tree of our forests; and to show the admirable elegance of their formsnbsp;it suffices to say that, at first sight, they forcibly recall those of the mostnbsp;admired species of our time�the Tulip-tree, the Magnolia, the Sassafras,nbsp;the Sweet-gum, the Plane-tree, the Beech, the Aralia, etc. The leaves ofnbsp;Protophyllum Sternhergii have the size and the aspect of those of the Catalpa,nbsp;one of our finest ornamental trees. Those of Menispermites obtusilobus, ofnbsp;Protospermum quadratum, represent in the same manner some of the rarestnbsp;shrubs, Menispermum, Ferdinandia, etc., carefully raised in conservatoriesnbsp;for the graceful forms of their leaves or the richness of their vegetation.nbsp;It is, indeed, the first impression received from the beauty of forms of thenbsp;leaves of the North American Cretaceous, and the evident likeness of theirnbsp;facies to that of the finest vegetable types of our time, as we see them aroundnbsp;us, which strikes the paleontologist, and may lead him into error in forcingnbsp;upon the mind the belief of a typical identity where possibly there may be anbsp;mere likeness of outlines, a casual similarity of forms in the leaves. For,nbsp;really, when we enter into a more detailed analysis of these Cretaceousnbsp;leaves, we are by and by forcibly impressed by the strangeness of the characters of some of them, which seem at variance with any of those recognizednbsp;anywhere in the floras of our time, and unobserved also in those of thenbsp;geological intermediate periods. Not less surprised are we to see united in

IKTEODTJOTION.

a single leaf, or species, characters which are now generally found separated in far distant families of plants. The leaves of Bremophyllum, so strikingnbsp;by the peculiar appendages of their borders; those of Anomophyllum, referable to Platanus by one-half, to Quercus by the other; those of Platanusnbsp;oltusiloha, half Acer, etc., are of this kind.

On another side, the characters of some of the Cretaceous species are sometimes of such a transient or indefinite order that it is scarcely possiblenbsp;to take hold of them and to describe them with any degree of reliance. Atnbsp;first sight they appear very distinct, but, in comparing a number of specimens, the differences dwindle by unmistakable transitions and disappear.nbsp;In other leaves, on the contrary, visibly identical by their outlines, thenbsp;nervation is so different that they are forcibly separated and referred tonbsp;far distant generic divisions. Hence this flora does not leave any satisfaction, any rest, to the mind. Even the most clearly defined types becomenbsp;doubtful in regard to their integrity when we see others, which, at first,nbsp;were recognized as positively fixed, manifesting instability and pointing tonbsp;diversity of relation by the discovery of new specimens. The leaves considered first as Sassafras, for example, seemed evidently referable to thisnbsp;genus; but when leaves of the same type were found with dentate borders,nbsp;though bearing, besides, all the characters of a genus which belongs to thenbsp;Laurinece, a family where, as yet, no representative has been found withnbsp;dentate borders of leaves; when others were obtained with subdivisionsnbsp;of the lower lobes in two or three, thus showing the palmate shape ofnbsp;Aralia leaves, the confidence in the value of the characters at first recognized had to be abandoned.

The first exposition of the Dakota Group flora shows four species of Ferns, six species of Conifers, and one of Oycadece only. To this smallnbsp;number we have added in this volume one species of GrleicJienia, six speciesnbsp;of Conifers, and five of Oycadeoe. The specific values of some of the vegetable remains referable to the Conifers is, however, doubtful, especiallynbsp;for those which are represented by cones only. Abietites Ernestines,nbsp;Sequoia formosa. Sequoia Beichenbaclii, and the fragments described asnbsp;Inolepis are of this kind; all, however, though their specific or genericnbsp;relation may be uncertain, are evidently representatives of some species of

OEETACEOCrS AJSD TEETIAEY PLOEA.

Conifer. The fragments referable to this group are difficult of determination, for the organs represented upon the coarse shale or hard ferruginous sandstone of the formation merely expose some traces of their more prominent outlines, originally printed upon the soft embedding matter. We donbsp;not find, therefore, any flattened cones with the scales, nor any flattenednbsp;branches with leaves, but impressions only, more or less deeply carvednbsp;into the stone, the cones even passing vertically or obliquely through thenbsp;shales and showing the space originally occupied, as a mere cylindricalnbsp;hollow, around which the forms of the scales are more or less clearlynbsp;molded. The numerous leaves of Pinus spread upon the surface have dugnbsp;in the same way, and by their hard substance, narrow linear channels,nbsp;representing the back of these leaves, with an indistinct midrib; andnbsp;branchlets of Segmia also are seen as longitudinal grooves, bearing onnbsp;both sides the same impressed form of their leaves. This cannot be considered a very distinct representation of characters, the minute detailsnbsp;desirable for an exact determination being more or less obsolete.

Among the specimens recently examined, a second fragment has been found referable to Phyllocladus} The presence of this genus in the Cretaceous flora is thus sufficiently ascertained. We may, therefore, recordnbsp;as recognized in the flora of the Dakota Group, for the Ferns, the generanbsp;Lygodium, Sphenopteris, HymenopJiyllum, and GleicJienia, the first three bynbsp;each one species, the last by two; in the Oycadece, Podozamites by sixnbsp;species, and in the Conifers, Sequoia by three species, Pimis by one, PMl-locladus by one, Torreya and Thuites each by one, leaving out as of uncertain generic relation with the cones mentioned above, Glyptostrohus (?)nbsp;gracillimus, rwhicli is perhaps identifiable with Sequoia condita, or withnbsp;Frenelites, and Geinitzia ij), known merely by the impressions of somenbsp;detached scales. To this should be added Araucaria from a speciesnbsp;described in �Extinct Floras of North America� by Dr. Newberry, fromnbsp;Nebraska specimens.

The first dicotyledonous leaves described in the �Cret. FI.,� under the name of Liquidamhar integrifolium, have been considered by some

^ Since this was written, Heer, in part 2d of Vol. VI of the Arctic Flora,has described this species under the name of Tkinjieldia Lesquereuxiana, as a plant of uncertain relation.

INTEODUGTION.

authors as uncertain in regard to their generic relation merely on account of their entire borders. The form of the leaves, however, especially asnbsp;figured (pi. xiv, fig. 3), with the lobes slightly enlarged above the sinuses,nbsp;then gradually narrowed to a slightly obtuse point, and the nervation also,nbsp;have the same character as those of the living Liquidambar Styracijlua.nbsp;It is true that the four species of this genus known in the present floranbsp;have serrate borders of leaves. But three fossil species represented bynbsp;leaves with entire borders have been described as Liquidambar from thenbsp;Tertiary of Europe; and, though this reference is more or less hypotheticalnbsp;and controverted, it shows, nevertheless, that botanists of high standing�nbsp;Unger, Watelet, Massalongo�have considered it, at least, as probable. Itnbsp;is easily seen that the leaves of Aralia Towneri (pi. vi,fig.l4) have a relationnbsp;in shape or general outline to those of Liquidambar integrifolium, and thisnbsp;apparent similarity can but suggest the possible relation of all these andnbsp;like forms to the genus Aralia. I may admit this relation as probable fornbsp;the two leaves figured in �Cret. FI.,� pi. xxix, figs. 8 and 9, which are comparable, by their primary nervation, to those of Aralia concreta (pi. ix,nbsp;figs. 3, 5). But though we have now a large number of specimens referable to diverse Araliaceous types, there is none as yet with leaves dividednbsp;into lanceolate acute lobes like those which are figured in pi. ii, �Cret. FI.,�nbsp;and with five primary nerves from the base. The reference of these leavesnbsp;to Sterculia has been proposed also, from analogy of forms. But accordingnbsp;to the definition of this genus as I admit it for the fossil leaves of thenbsp;Dakota Group, I refer to it merely tripartite leaves with narrow linearnbsp;lobes, comparable to those of Sterculia labrusca, like those of the few speciesnbsp;described in this volume.

A number of vegetable remains of the Cretaceous are evidently referable by their characters to Populibs. The only fragments of dicotyledonous leaves recognized by Heer, in the specimens which he studied from thenbsp;Lower Cretaceous formations of Greenland (Kome), represent a Populus,nbsp;appropriately specified by the name of P. primceva. From a higher stage ofnbsp;the same Cretaceous formation of that country (Atane) the celebrated Swissnbsp;paleontologist has described four other species of Populus. In his � Phyllitesnbsp;Cr�tac�es du Nebraska,� and from specimens of the Dakota Group, he hasnbsp;recognized Populus litigiosa, Populus (?) Pebeyana, and another species still.

CEETAOBOUS AND TEETIAET FLOEA.

P. cyclopTiylla, described in Proc. Acad. Nat. Sci., Philadelphia. Professor Newberry, in his paper �On the Later Extinct Floras of North America,�nbsp;has described, also, besides the doubtful P. (?) Peheyana, three new species:nbsp;Populus (?) cordifolia, P. elliftica, and P. microphylla. The specificationnbsp;and the interrogative punctuation applied to some of these names shownbsp;that the authors themselves do not consider the generic reference as definitive, the character of some of the leaves being somewhat in disagreementnbsp;with those generally recognized in species of Populus of our present time.nbsp;Indeed, species of this kind, like the present P. alba, for example, havenbsp;such multiplied and diversified forms of leaves, such great variability innbsp;their nervation, the mode of attachment, the length of the petiole, etc., thatnbsp;they readily offer, by comparison with fossil leaves of obscure relationship,nbsp;some points of affinity which, not being found elsewhere, have to be considered by the authors. Hence the doubtful references which may be,nbsp;and are often, rectified by subsequent discoveries, as is proved by the greatnbsp;proportion of synonyms appended to the enumeration of Populus species.nbsp;To obviate this inconvenient multiplication of fluctuating species of Pop-ulus I proposed a new generic division, under the name of Populites, fornbsp;the classification of those Cretaceous leaves, numerous indeed, which,nbsp;partaking of some of the characters of Populus, are nevertheless removednbsp;from this division by some others, as remarked in the first memoir whichnbsp;I published on some Cretaceous plants from Nebraska.

This paper had to be prepared on short notice from a limited number of specimens, but since its publication I have had opportunity to studynbsp;the specific forms of the Cretaceous Flora by comparing a very large number of specimens, and have thus been able to recognize a more evidentnbsp;affinity of some of those leaves referred to Populites with other genericnbsp;divisions. Populites Lancastriensis, P. elegans, which Schimper admits asnbsp;a true Populus, and Populites cyclophyllus are the only species preserved innbsp;this genus. P. ovatus, considered as possibly referable to Celtis in Cretaceous Flora, being rather related by its characters to the Ampelidece, isnbsp;described under the new generic division of AmpelopTiyllum. The affinitynbsp;of P. guadrangularis being more evident with Alnus, has been describednbsp;as Alnites. P. fiabellata, as seen from other specimens, appears to be anbsp;deformed leaf of G-reviopsis Haydenii, and P. Salisburioefolia, being related tonbsp;Cissus, is described as Cissites.

INTEODUOTIO�T.

In regard to the distribution of Populus, to which are referred the most ancient dicotyledonous leaves known as yet, from the Lower Cretaceous of Greenland, the genus has, as said above, three species knownnbsp;already in the Upper Cretaceous of that same country, and five or six innbsp;the Dakota Group, It has, however, not been remarked in any Cretaceousnbsp;Flora of Europe. It is not mentioned in the review of the genera represented by the, as yet, undescribed species of Aix-la-Chapelle,^ and no formnbsp;even distantly related is described in the Lower Paleocene Flora of Gelin-den. It has, however, one species in the Eocene Flora of S�zanne, andnbsp;increases in the number of its representatives in all the stages of thenbsp;European Miocene. As far as we know it, till now, it has few species innbsp;our Lower or first American Tertiary Group�the Eocene; has a large proportion, eight per cent, of the species, in the Evanston Group; still more, ornbsp;twelve per cent., in the Miocene of Carbon, and is present in the second^nbsp;the Green River Group in four species, three of them of peculiar types,nbsp;one of which is very abundant.

The presence of Willows {Salix) in the Flora of the Dakota Group cannof be controverted, though neither seeds nor scales of this genusnbsp;have been found as yet. As it is seen in �Gret. FL,� p. 60, pi. v, figs.nbsp;1-T, I have described as referable to one species only a number of leavesnbsp;somewhat different in size and shape. As the specimens representingnbsp;them are from the same locality, and as I recognized upon some numerousnbsp;fragments of leaves a unity of character, size, form, and even texture andnbsp;color, I considered them as mere varieties of leaves of the same tree. Dr.nbsp;Newberry has, from the same formation, four species which, he says, henbsp;has chosen to regard as distinct, for geological convenience. No Salix hasnbsp;been recognized as yet in any stage of the Cretaceous of Greenland; butnbsp;one species, Salicites Hartigii, Dkr., is from the Quader-sandstein of Germany, and another, Salix Goetziana, Heer, from Quedlinburg. The genusnbsp;is therefore sparingly represented in Europe and North America in Cretaceous Floras which are considered as nearly synchronous.

The other genera of the Amentaceoe, Betula, Alnus or Alnites, Myrica,

^Dp. M. Debey has recently published a fine memoir on some querciform leaves found in the sand rocks of Alx-la-Chapelle^ Khenish Prussia.

CEEaiACEOUS AND TEETIAET FLOEA.

Quercus, Fagiis, and Ficus, to which leaves have been referred in the Cretaceous Flora, do not require any observations. In this case, as in all the determinations of fossil plants, the characters of the species are not

�

always satisfactorily established, but the generic affinities have been recognized or passed by authors without any marked criticism. The generic relation is specially positive for the remains referable to Myrica; one fragmentary leaf and some seeds have been already described in the �Cret. FL,�nbsp;w^hile two fine new species are added in this memoir. It seems equally sonbsp;for Quercm or its peculiar division, Bryophyllum, of which we have twonbsp;new species, and for Ficus, to which three species are added.

Specimens of leaves referable to Platanus have been found in moderate proportion both in Nebraska and Kansas. The first was described by Heer, in the �Phyllites Cr�tac�es du Nebraska,� as Platanus Newherryi,nbsp;from a very incomplete fragment. The accuracy of this determinationnbsp;was, however, subsequently verified by the discovery of more completenbsp;leaves, figured in �Gret. FL,� pi. viii, figs. 2 and 3, and pi. ix, fig. 3,nbsp;which show the narrowed base descending along the petiole lower thannbsp;the point of union of lateral primary veins, and also the tendency to anbsp;three-lobed division, characters which are not observable in the fragmentnbsp;which Professor Heer had for his examination. To this fine species havenbsp;been added: Platanus primceva, described from leaves so remarkably similar to those of P. aceroides of the Miocene that I was at first disposed tonbsp;consider them as identical. I have lately received numerous large leavesnbsp;of this species with specimens bearing fruits, which, very small, show anbsp;great difference from those of the living species; then, P. Heerii, rare, likenbsp;the former, and found, as yet, only along the bluffs of the Salina River;nbsp;P. ohtusiloha, from a number of somewhat fragmentary specimens fromnbsp;Beatrice, Nebraska, representing leaves of about the same size and ofnbsp;the same characters; and P. diminutiva�all species described and figurednbsp;in �Cret. FI.� The last one, as remarked in its description, may be anbsp;dwarfed form of P. primceva or P. Heerii. The leaf appears as gnawednbsp;along the veins by insects or perhaps by a parasite fungus. Its specification is not positive and is subject to criticisms. The base of the leaf isnbsp;rounded to the petiole, a character as yet unique for a species of this kind.nbsp;P. recurvata should, following the opinion of my honored friend Saporta, be

'A

INTEODUCTION.

referred to the Araliacem by a more intimate affinity to Araliopsis species;^ and Platanus affinis seems now, after the examination and comparison ofnbsp;a number of specimens from Kansas, more evidently referable to thenbsp;Am;pelidece than to the Platanece. Therefore these last two species arenbsp;now eliminated from this generic division. The first is now Araliopsisnbsp;recurvatus, the second Oissites affinis.

I persist in considering Platanus Heerii and P. obtusiloba as two different species, though it has been suggested that the last was probably a mere variety of the first. The identity is denied not only by the size,nbsp;the facies, and the nervation of the leaves, but especially by the thinnernbsp;texture of those of P. obtusiloba. The fact that the numerous specimensnbsp;representing it are all from the same place in Nebraska, and that P. Heeriinbsp;has not been found in that State thus far, confirms this separation. Innbsp;regard to this last species Professor Geinitz has remarked in �Isis,� 1875,nbsp;p. 558, that paleontologists might, perhaps, recognize in it a Credneria.nbsp;There is, indeed, some similarity in the general outline of the leaves. Butnbsp;this might be said of many of the generic forms of the Cretaceous, whichnbsp;seem referable to a few different types, or to present in one leaf the characters which are now generally found isolated in separate vegetable groups.nbsp;The genus Credneria, known as it is to me by what is described by Stiehler,nbsp;Vol. V of the � Paleontographica,� includes species with cordate or sub-cordate leaves (none narrowed to the petiole), and bearing above the basenbsp;two or three true secondary veins at right angles to the midrib. In P.nbsp;Heerii the leaves are cuneate at the base, even gradually narrowed ornbsp;decurrent on the petiole, which thus becomes slightly winged, and thenbsp;veins under the primary nerves are mere marginal veinlets. Perhaps thenbsp;relation of this species is more marked to the genus EttingsJiausenia, which,nbsp;I regret to say, is known to me only by supposed synonyms ChondrofJiyllumnbsp;grandidentatum, as represented by Heer in the Cretaceous Flora of Moletein,nbsp;and by Phyllites repandus, Sternb., two forms which have no affinity tonbsp;Platanus.

The typical character of the Cretaceous species of Platanus is more evidently related to the Aralieae than to any other. This is proved by thenbsp;reference to that genus of leaves now generally admitted as species of

* Hekr, in � Arctic Flora,� vol. vi, part 2, admits it as Sassafras.

CRETACEOUS AND TERTIARY FLORA.

Aralia, as Platanus grandifolia, P. digitata, P. Jatropcefolia, P. Hercules, Ung., and P. latiloha, Newby. The leaf of Sassafras {Araliopsis) Platanoidesnbsp;(pi. vii, fig. 1) has the facies and some of the characters of Platanus morenbsp;distinctly defined than any other of the group; the same characters arenbsp;even reproduced in Aspidiopliyllumplatanifolium (pi. ii, fig. 4).

The geological distribution of the genus Platanus is truly remarkable. No trace of it is recorded as yet in the Cretaceous of Europe, not even innbsp;the Paleocene and Eocene of France, so rich in fossil vegetable remains.nbsp;Its first appearance in Europe is in the Upper Miocene of Oeningen, andnbsp;of Austria and Italy, where it is represented by two very similar forms.nbsp;Platanus Gruillelmce and P. aceroides, two species present in the samenbsp;formation from the northern parts of the arctic lands to Italy. It is followed in the Upper Tertiary, or Pliocene, of this last country by Platanusnbsp;Academice, Gaud., related as progenitor, perhaps, to the living P. orientalis.nbsp;I have remarked above that the relation of leaves of the Dakota Group tonbsp;Platanus has been considered as doubtful by some European paleontologists. This doubt may have been induced by the understanding of thenbsp;total absence of Platanus leaves in the Cretaceous and Lower Tertiary ofnbsp;Europe. If so, it is certainly removed by the presence in our ligniticnbsp;Eocene of some very beautiful and well characterized species of this genus:nbsp;Platanus Haydenii and P. Beynoldsii, Newby. These species, discoverednbsp;first in the Tertiary of the Upper Missouri River, near Fort Union, are predominant at Golden, Colorado, and are also found at Black Butte Station.nbsp;The third Tertiary Group, that of Carbon, has, for the more numerousnbsp;representatives of its Flora, leaves of Platanus aceroides and P. Guillelmoe.nbsp;No species of this genus has been described from the Oligocene Greennbsp;River Group; but we have from the Upper Tertiary {Pliocene) of California very fine specimens of leaves of two species, P. appendiculata andnbsp;P. dissecta, closely related by their characters to the living P. occidentalismnbsp;Therefore, and considering the geological records, we may trace the originnbsp;of Platanus as far down as the North American Cretaceous, and follow itsnbsp;development through nearly all the stages of its Tertiary to our presentnbsp;time, by a number of closely allied intermediate forms.^

'Platanus Seerii, L. and P, affinis L. are mentioned by Heer in the Cretaceous of Atane, Greenland.

INTRODUCTIOK.

Coming now to the Laurineoe, I have to remark somewhat more definitely on the Cretaceous species referred to this family. The relation of some of them to the genera to which they have been referred is generallynbsp;acknowledged, and the presence of the Laurinece in our Cretaceous Floranbsp;receives a kind of historical authority from that of a Sassafras in a Cretaceous formation of Greenland,^ of three species of Daphnophyllum in thatnbsp;of Moletein, and of Laurus cretacea, Baphnogene primigenia, Bafhnitesnbsp;G�pperti, in that of Niedershoena. Of the species which have formerlynbsp;been described in the Flora of the Dakota Group, Laurus Nebrascensis isnbsp;related to Daphnophgllum ellipticum and B. crassinervium of Heer, whilenbsp;Cinnamomum and Oreodaphne cretacea are comparable to Baphnogene prim-igenia of Ettingshausen. Persea Sternbergii is also evidently of the samenbsp;family, and the two leaves, described here below under the name ofnbsp;Laurus proteoefoUa, are, indeed, allied to species of Laurus or of Persea bynbsp;their nervation, especially by the more acute angle of divergence of thenbsp;lower veins, though they show in the grooved middle nerve a characternbsp;often remarked in species of Ficus, especially Ficus protogcea, Heer, of thenbsp;Greenland Cretaceous Flora. Moreover, the fruit described (�Cret. FI.,�nbsp;p. 74) as Laurus macrocarpa satisfactorily completes the evidence affordednbsp;by the leaves of the existence of species of Laurineoe in the vegetable worldnbsp;of the Cretaceous epoch. We have, however, to eliminate from this familynbsp;Lauraphyllum reticulatum, which appears more properly referable to Ficus.nbsp;Its nervation, and especially its areolation, formed of square or irregularlynbsp;polygonal meshes by the interposition of tertiary veins between the secondary ones and parallel to them, and the rectangular subdivision of itsnbsp;branches, are of the same character as in Ficus Gteinitzi, Ett., Ficus protogcea,nbsp;Heer, and as in many species of this genus now growing in Cuba, andnbsp;even Florida, Ficus suffocans, F. lentiginosa, F. pertusa, F. dimidiata, etc.nbsp;Numerous specimens recently found in Kansas represent the fossil speciesnbsp;in characters more precise than formerly, as seen in its more detailednbsp;description under the name of Ficm laurophyllum.

But if the reference of some of the above-mentioned leaves to the Laurinece is not contested, it is not the same in regard to those which, at

^ In Arct. Fl./^ vol. vi, 2(1 part, pp. 75-78, Heer describes as new species Laurus plutonia, L. angusta, L. Holies, L. Odini, with Cinnamomum Sezannense, Wat., from the Upper Strata of Atane,

OEETACEOUS AND TERTIAEY FLOEA.

first appearance, were considered as more positively related to this family, and which have been described under the generic name of Sassafras. Thenbsp;question of the relation of those leaves which, by their number, seem tonbsp;be the essential components of the North American Cretaceous Flora, hasnbsp;been already touched upon (�Cret. FL,� p. 77). But since the publicationnbsp;of that work I have obtained from divers localities a large number of specimens of all the forms described there as species, and I have now somenbsp;more data to offer to the consideration of paleontologists on the subject.

From historical documents the presence of Sassafras species in the Flora of the Dakota Group is as legitimately presumable as that of speciesnbsp;of Lauras or Persea. In his �Flora fossilis arctica,� Heer has described asnbsp;Sassafras arcticwm a leaf which, by its form, is similar to those describednbsp;as Sassafras cretaceum, as remarked by the author, differing merely by itsnbsp;base tapering somewhat less narrowly to the petiole. The nervation isnbsp;of the same character. Saporta considers the Greenland leaf as a truenbsp;representative of Sassafras. He has himself published in the �S�zannenbsp;Flora,� ^ as S. primigenium, two fragmentary leaves whose base, more narrowly tapering, is similar to that of S. Mudgei of the �Gret. FL,� as wellnbsp;as the lobes which, enlarged in the middle, have that ovate-lanceolatenbsp;shape so distinctly marked in the present S. officinale. There is also nonbsp;appreciable difference in the nervation. The lower secondary veins of thenbsp;middle lobe ascend a little higher in the leaves of the S�zanne Flora, andnbsp;unite with those of the lateral lobes somewhat nearer the borders of thenbsp;sinuses. But in some of the specimens of Kansas the same appearancenbsp;is remarked also, and the difference between the greater or less distancenbsp;which separates from the sinuses the branches which unite the uppernbsp;division of the secondary veins is observable upon leaves of S. officinale,nbsp;this division being sometimes marginal, sometimes curving one to threenbsp;millimeters lower than the border of the sinuses. Comparing leaves ofnbsp;Sassafras officinale with those represented by Saporta in the � Flora ofnbsp;S�zanne� and the specimens of S. Mudgei from Kansas, it is impossible fornbsp;me to recognize any character, even any specific difference, by which thesenbsp;leaves could be separated. It is therefore not surprising that Dr. Newberry first, and after him Heer and Schimper, did consider Cretaceous

366, tab. viii, figs. 9 and 10.

INTEODUCTIOK

specimens of this kind as representing species of Sassafras. In the last volume of his superb work on Vegetable Paleontology,^ Prof. W. P. Schim-per, speaking of leaves of Sassafras cretaceum, of which I had sent himnbsp;photographical designs, remarks: �That those leaves, very variable innbsp;size, present such a remarkable likeness to those of S. officinale, now livingnbsp;in North America, that one would be disposed to consider them as belonging to a homologous species.� He rightly adds that the only differencenbsp;seems to be in the thicker substance of the fossil leaves. Even on thisnbsp;point I have from Texas specimens of the present S. officinale, whose leavesnbsp;appear of a consistence nearly as thick as it seems to be in those of thenbsp;Dakota Group.

On the other hand, no species of the Lanrinex family living at our time is known with dentate leaves; and it may be remarked, from thenbsp;figures, that the two leaves described as Sassafras cretaceum (�Gret. FI.,�nbsp;pi. xi, figs. 1 and 2) have the borders of the lobes somewhat dentate, andnbsp;some of the secondary veins running into the point of the teeth, or cras-pedodrome. This character is still more marked in S. mirabile, loc. cit.,nbsp;pi. xii, fig. 1, a form extremely common in Southern Kansas, and represented in very numerous and remarkable varieties. In some of the leavesnbsp;the secondary veins are all camptodrome, and therefore the borders of thenbsp;lobes are entire. In others, as seen, pi. xi, fig. 2, the outside lateral veinsnbsp;are craspedodrome, and thus the borders are dentate, while on the insidenbsp;they curve along the borders, which are entire. In the fine complete leafnbsp;(fig. 1 of the same plate) the middle lobe has the veins all camptodromenbsp;on the left side, while on the right one, a few of them, one or two, reach tonbsp;the border, which has, therefore, one or two short indistinct teeth, and thenbsp;lateral lobes are clearly dentate on the outside only. This evidently showsnbsp;such a disposition, to variations of nervation and border divisions, that Inbsp;formerly considered as unjustifiable a specific, and still more a generic,nbsp;division between the leaves of pi. xi, figs. 1 and 2, and those of pi. xii,nbsp;figs. 2 and 3, of the �Gret. Flora.� When, therefore, we find the samenbsp;difference between the leaves which represent S. mirabile (pi. xii, fig. 1),nbsp;it seems that the same conclusion should follow. But in this case, withnbsp;the more generally predominant character of the indentation of the leaves,

* Trait� de Pal�ontologie v�g�tale, vol. i�, p. 298.

CEETACEOUS AND TEETIAEY ELOEA.

which, in some specimens larger than the one figured, are now deeply cut by divisions like pointed lobes, there is still another character, remarkednbsp;on specimens recently discovered, which seems more forcibly to separatenbsp;these forms from the Laurinece, and indicates a more evident relation tonbsp;the Araliacem. A number of those specimens communicated by M. Chs.nbsp;Sternberg, to whose careful and zealous researches the Flora of the Dakotanbsp;Group is indebted for many important discoveries, represent large leaves,nbsp;which, by the outlines, the nervation, and the dentate borders of the lobes,nbsp;are like 8. mirahile of pi. xii, fig. 1. The leaves, however, which are muchnbsp;larger, the lobes measuring as much as ten centimeters in length fromnbsp;the point of union of the primary nerves, greatly differ by the forkingnbsp;-of the lateral nerves from a point two and one-half centimeters abovenbsp;their base, thus forming, of course, a subdivision of these lobes intonbsp;two equal parts, or a palmately five-lobed leaf. They are described asnbsp;Sassafras {Araliopsis) dissectum. Among the innumerable varieties in thenbsp;shape of the leaves of the living Sassafras officinale we see a constant andnbsp;gradual mode of division, passing from a round or oval and entire shapenbsp;to a bilobed and trilobed one; but, as yet, I have been unable to observenbsp;a single case of subdivision of the lateral lobes, or to find a palmatelynbsp;five-lobed Sassafras leaf. This character is, on the contrary, far morenbsp;generally seen in the Araliacece of our time. Even in a section of thenbsp;Araliacem, the genus Hedera, whose leaves may be compared to some ofnbsp;those under examination, I do not know any species with trilobate leaves.nbsp;Hedera turbascens, H. discolor, H. argentea, H. aurifolia, H. jatropoefolia,nbsp;have leaves five to seven palmately lobed, or when occasionally trifid theirnbsp;segments are narrow and acuminate. From this the relation of the fivenbsp;palmate leaves to the Araliacem becomes more evident.

Going further into this kind of investigation, we are met by a new difiiculty in the appearance of another modification in the character ofnbsp;this peculiar type of leaves. In examining the first specimens of thenbsp;species represented (pi. xii and xiv), I could but consider them as representing either Sassafras {Araliopsis) ohtusum or S. mirahile, the specimensnbsp;being fragmentary, having only the lobes or part of them preserved. Asnbsp;long as the auricled and peltate base was unknown, the reference of thenbsp;specimens could not be different. The nervation, the form of the lobes.

INTRODUCTION.

their size, all are of the same character as in S. mirabile. But in the peltate base of the leaves there is another character which, separately considered, relates the leaves to the Menispermacece. We thus have Sassafras already represented in those leaves by S. Mudgei, and less positively bynbsp;S. acutilohum; Araliopsis, to which are referable S. mirabile, with the dentate S. cretaceum, 8. obtusim, 8. dissectum, 8. platanoides, Platanus recurvata,nbsp;and in a new generic division, under the name of Aspidiophyllum, thenbsp;leaves which, either Arabia or Sassafras, by their upper trilobate part,nbsp;are necessarily separated from these genera by their auricled peltatenbsp;�appendage. Still, the subdivisions in the classification of the peculiar andnbsp;so-called Sassafras leaves have to be pursued further, for by degrees andnbsp;by the gradual obliteration of their lobes they become round or truncate,nbsp;or broadly pointed at the top, preserving more or less the narrowed base,nbsp;tapering to a long petiole, and the trifid craspedodrome nervation from anbsp;�distance above the borders, and thus they become more evidently relatednbsp;to other vegetable orders. One species is a true Hedera, another passesnbsp;to the Hamamelideoe, and a number have their affinity with the Ampelidece.

The characters of the leaves of the Ampelidece, especially those of Cissus, are somewhat obscurely represented in Sassafras Harlcerianumnbsp;(� Cret. FL,� pi. xi, figs. 3 and 4; pi. xxvii, fig. 2) and in S. obtusum (pi. xiii),nbsp;more distinctly in Cissites acuminatus (pi. v, fig. 3) and C. Heerii (pi. v, fig. 2),nbsp;two new species described in this memoir. They appear to constitute annbsp;indivisible group. Some of the leaves formerly described as Populites arenbsp;also referable to this section, or to another less exactly defined Ampelo-phyllum, allied by some of its characters to Hedera, by others to Oredneria,nbsp;thus intermediate between the Ampelidoe and the lilicacece; by the areola-tion this genus is related to Greviopsis, and also more distantly to Chon-draphyllum of Heer, as remarked in the description. From this it is perceivable that this Sassafras type, which at the beginning was regarded asnbsp;simple, well defined, and limited in its character, is, on the contrary, multiple, and representing forms which, from increased researches and discoveries, indicate affinity to a number of different genera or orders of thenbsp;vegetable kingdom.

The same remark is equally applicable to the leaves which have been �described in the �Cret. FL� under the generic name of Protophyllum. The

C F 2

CRETACEOUS AND TERTIARY FLORA.

disagreement in the affinities of its species has been explained in the remarks following the description of the genus. I have now to add stillnbsp;to this division two leaves recently communicated from Kansas, representednbsp;in pi. iii, fig. 1, and pi. viii, fig. 4. They fully confirm the former observations. By the outline of the leaves, their craspedodrome nervation,nbsp;and the presence of two pairs of secondary veins under the primary onesnbsp;and at a right angle to the midrib, they represent a species of Protophyllum;nbsp;but the border base of the leaves is truncate, not subpeltate, and by thisnbsp;difference the leaves are rather referable to Oredneria, from which, however, they differ by the veins as well as their divisions, being all craspedodrome, and by the truncate, not cordate, base of the leaves. I formerlynbsp;published a short description of them under the name of Oredneria^? micro-phylla. It now seems that, by their evident relation to Protophyllum quad-ratum, they have to be admitted in this last generic division, an opinionnbsp;which may be put at naught by the discovery of specimens pointing tonbsp;another reference for these leaves.

We have, also, an addition of three new species to the group of Cretaceous plants described under the generic name of Menispermites. In this case, however, there is no difficulty whatever in conformably unitingnbsp;into a definite group the characters of the leaves which, round, ovate, ornbsp;oval, with borders entire or undulate, have a common generic affinity,nbsp;indicated by their nervation. In order more clearly to bring into view thenbsp;relation of the undulate-lobed forms of leaves described in the �Gret. FI.�nbsp;(pi. XX, figs. 1-4, and pi. xxv, fig. 1), I have represented (pl. xv, fig. 4)nbsp;a finely and wholly preserved leaf of Menispermites obtiisilohus, which,nbsp;though small, is easily identified with the large one of �Cret. FI.� (pl. xxv,nbsp;fig, 1). Now, comparing it to figs. 2 and 3 of the present pl. xv, thenbsp;identity of nervation is defined by the five basilar veins, with a thin pairnbsp;of marginal veinlets underneath; and by the upward direction of thenbsp;internal lateral veins, which in fig. 4 ascend to above the middle, passnbsp;still higher in the short oval leaf, fig. 3, and reach nearly to the obtusenbsp;point in fig. 2. The subdivision of the tertiary veins is in all the leavesnbsp;of the same type, and the shape of the leaves or their outlines are merenbsp;modifications, depending upon the direction of the veins. The leaf, fig. 3,nbsp;is peltate from the point of attachment of the petiole near the middle.

INTEODUCTION.

The character of the nervation remains, however, the same. It is somewhat obscured in the figure from indistinctness of the specimen. In figs. 1 and 2, representing leaves entirely preserved and nearly round, thenbsp;nervation is marked by three pairs of primary nerves on each side of thenbsp;midrib, and under them by one pair of true marginal veinlets curving onnbsp;each side toward the borders. Comparing, therefore, these peltate leavesnbsp;with fig. 4, the position of the petiole is the only notable difference, and thenbsp;transition to fig. 5 by slight modifications of characters is easily remarked.nbsp;The peltate form of these round leaves has suggested the fitness of a slightnbsp;modification in the characters assigned to the genus Pterospermites in thenbsp;� Gret. FI.� (p. 94), the leaves being sometimes rounded or subcordate atnbsp;base. The difference is immaterial, and is remarked even upon leaves ofnbsp;the same species of Menispermum of our epoch. These round peltatenbsp;(leaves, for example, are so much like those of living species of Cissampelos,nbsp;that they rather prove the adaptation of this generic division to all thenbsp;Cretaceous leaves which I have referred to it.

The Magmliacece are more numerously and definitely represented in the North American Cretaceous Flora than they are in that of Europe.nbsp;Magnolia alternans and M. Capellini have been described by Heer in hisnbsp;�Phillites Cr�tac�es du Nebraska;� and since that time these two speciesnbsp;have been recognized throughout the whole explored area of the Dakotanbsp;Group, as also in the lower stage of the Cretaceous of New Jersey, and innbsp;the Upper Cretaceous of Greenland. M. speciosa of Moletein has been discovered in Colorado with a fruiting cone or carpite of this genus. Twonbsp;other species have been described from the Dakota Group: one, M. ohovata,nbsp;by Dr. Newberry, in his ��Ancient Floras,� another, M. tenuifolia, in �Cret.nbsp;FL,� and two new ones, M. ohiusata and M. Ishergiana, by Heer, fromnbsp;Atane. In Europe, M. amplifolia and M. speciosa are described by Heernbsp;in the Flora of Moletin�there represented by leaves and fruit.

To the same order belongs Liriodendron, so easily recognized by the peculiar form of its leaves. Its Cretaceous origin, or rather existence, isnbsp;marked in the Dakota Group by a number of specific representatives locallynbsp;and distantly distributed. The genus is not represented in the Cretaceousnbsp;Flora of Europe; but in the � Cretaceous Flora of Groenland� Heer describesnbsp;six varieties of Liriodendron Meekii from Atane, and no less than eight

CEETACEOUS Al^D TEETIAEY FLOEA.

specific forms have been described from Nebraska and Kansas�some of them extremely well defined. This shows, perhaps, more evidently thannbsp;any other fact remarked on the characters of the plants of the Dakotanbsp;Group the great disposition to variableness by modification of some characters in the first Dicotyledonous plants. These changes have either causednbsp;a multiplication of specific forms preserving traces of the original typesnbsp;in traversing the subsequent geological formations, or have graduallynbsp;destroyed the number of specific representatives of some genera, as innbsp;Liriodendron, or even caused the total disappearance of some of the bestnbsp;defined and more predominant types, like those of Oredneria, Ptercyphyllum,nbsp;etc. Of these, however, the original characters may have been so widelynbsp;varied that the ultimate derived forms have not yet been distinctly recognized on plants living now. The two last-named genera, Oredneria andnbsp;Protophyllwm, may possibly be referable to some subdivisions of the Colum-niferce, the Buttneriaceoe and Pterospermce, for example.

The three species which I have described under the insufiiciently-defined genus of Sterculia are all very uncertain in their relation. As much may be said for the following and last classes of the vegetablenbsp;kingdom:

To the Acerece is referable Negtmdoides acutifolius. The leaf, however, as seen from pi. xxi, fig. 5, and its description, is too fragmentary for anbsp;satisfactory determination of its characters. Acer antiguUm is describednbsp;by Ettingshausen in his �Flora of Niedershcena,� but from the opinionnbsp;of the author the reference is uncertain. The leaf rather resembles anbsp;deformed form of Qmrcm or of Liriodendron. In the same order Heer has,nbsp;from the Upper Cretaceous of Greenland, a Sajdndus jprodromus, represented by one leaf only, which has evidently the character of the genus.nbsp;A beautiful species of Sapindus described here from Colorado is also presentnbsp;at Atane. This genus is therefore Cretaceous. The reference to the Bham-nacem of the leaf described as BTiamnus tenax in �Cret. FI.� is apparentlynbsp;legitimate, for of the same group three other species, B. prunifolius, anbsp;CelastrapTiyllum, and an Ilex, are described here from the same formation.

To the Anacardiacece we have probably to refer, as Bhus Lebeyana, the species described as Populus and as Juglans Lebeyana as seen in �Cret.nbsp;FL, p. 110. I have not obtained from the Dakota Group any new materials

lOTEODUOTIOK

comparable to this form, especially common in Nebraska; but I have seen a very fine specimen of it got out of a deep tunnel in Oregon, presentingnbsp;upon its surface small punctiform protuberances, apparently oily glands,nbsp;like those remarked upon leaves of the living BJius aromatica and othernbsp;species of this genus. The leaves are figured (pi, Ivi, figs. 5,6). A speciesnbsp;of Bhus is described from the Cretaceous of Greenland by Heer, whilenbsp;considering historical authority, we have the same evidence in favor ofnbsp;Juglans by a species of this genus in the Cretaceous Flora of Moletein andnbsp;one in that of Greenland.

Of the Bosifiorem we have from the Dakota Group one leaf and one fruit described as Frunm. I have recently received from M. Towner a fruitnbsp;of the same character upon a specimen bearing leaves of Aralia Towneri.

The Myrtiflorce, as well as the Leguminosoe, present by a number of specimens in the Greenland Cretaceous, have not been thus far positivelynbsp;recognized in Kansas and Nebraska, but seen by one silique only innbsp;Colorado.

The few groups not considered in this review have been remarked upon already in the �Cretaceous Flora,� and the views in regard to thenbsp;leaves referred to them have not been modified either by remarks ofnbsp;European authors or by the discovery of new materials.

The want of positiveness in the characters of some of the Cretaceous plants cannot in any way weaken reliance upon the data derived from thenbsp;exposition of the Flora of the Cretaceous age, nor throw any discredit onnbsp;the conclusions which they dictate. What the Flora of the Dakota Groupnbsp;positively shows is a great predominance of dicotyledonous plants in itsnbsp;composition; and that is all that may be positively known as yet of thenbsp;remarkable change it attests in the vegetation of that period. The causes,nbsp;the mode of proceeding of nature, either by slow, gradual, or by rapidnbsp;modifications, remains as yet inscrutable. But the characters of dicotyledonous leaves cannot be mistaken; the relation of most of them to groupsnbsp;of plants of the present Flora possesses positive evidence. The Cwigulifereoenbsp;with species of Quercus and Fagus; the Salicinem with species of Populus;nbsp;the Platanem with Platanusprimceva, leaves and fruits; the Laurinece,. represented also by leaves and a fruit of Laurus, by leaves of Persea, Oinna-momum, Sassafras; the Araliacece, the Magnoliaceoe, with fruits and leaves;

CEETACEOUS AND TEETIAEY ELOEA.

the numerous forms of leaves of Liriodendron, so peculiar that they cannot be mistaken for those of any other group or plant; even the Menis-permacecb constitute, by their fossil remains, vegetable groups quite as definite as they could be established from living plants.

Since the publication of the �Cretaceous Flora� (vol. vi of the U. S. Geological Reports of Dr. F, V, Hayden) the character of the vegetationnbsp;of the Middle Cretaceous as represented in the Dakota Group has becomenbsp;better defined by the discovery of a large number of specimens of fossilnbsp;plants, which have increased from 130 to 190 the number of vegetablenbsp;forms considered specific, already known from this formation. The wholenbsp;Flora of the Cenomanian epoch, as it is shown in the table of distribution, is composed of 446 species, of which 310 are dicotyledonous and 130nbsp;are cryptogamous and gymnospermous plants. Of the 190 species of thenbsp;Dakota Group, 162 are dicotyledonous and only 28 represent crytogamousnbsp;and gymnospermous plants.

Numerous works on the Jurassic Flora have sufficiently proven that up to its upper member the Wealden, or lower Neocomian, it is entirelynbsp;composed of gymnospermous and cryptogamous plants�especially Ferns,nbsp;Cycadece, and Conifers. The Neocomian, whose vegetation is but littlenbsp;known as yet, shows in its remains the same constituents of its Flora.nbsp;Upon it is superposed in Germany the upper Neocomian, or Urgonian,nbsp;from which a series of fossil plants, 22 in number, have been described bynbsp;Schenk from the Wernsdorf-Schichten of the Carpathian Mountains ofnbsp;Austria; and there also no dicotyledonous plant has been found, andnbsp;nothing indicates the decadence of the reign of the gymnospermous plantsnbsp;or shows any kind of difference which could lead one to presage thenbsp;appearance of the Dicotyledons.

We owe to Heer the most interesting documents on the characters of the vegetation of the Middle Cretaceous�first by the publication of thenbsp;Flora of Kome, and then of that of Atane, both in Greenland.

The Flora of Kome, composed of 85 species, has, says the author, its greatest affinity with that of the Wernsdorf shale or upper Neocomian onnbsp;one side, and with that of the Wealden on the other. With the plants ofnbsp;the higher Cretaceous stages it has only 7 species�Ferns and Conifers�nbsp;in common. Most of the specimens of the group submitted to Heer�s

INTEODUCTION.

examination have been obtained on the peninsula of Noursoak (70� 37' N.), from beds of shale alternating with banks of sandstone, the whole over-lying granite or primitive formation. One of the localities, that of Elkor-fat, is 500 feet above that of Kome, but the plants are of the same kind.nbsp;The vegetable remains belong mostly to cryptogamous and gymnosper-mous plants: 41 Ferns, 1 Marsilia, 1 Lycopod, 3 Eguisetacece, 10 Cycadece^nbsp;21 Conifers, 6 Monocotyledons, and a single Dicotyledonous species.

On the south side of the same peninsula of Noursoak, near Atane, at an elevation of 650 feet above the sea, another lot of plant-remains, collected also by the expeditions of Nordenskj�ld, and submitted to Prof.nbsp;Heer for examination, represents a Flora composed of far different elements.nbsp;It has 170 species: 3 Fungi, 31 Ferns, 1 Marsilia, 1 Selaginella, 1 Equi-setum, 8 Oycadeoe, 27 Conifers, 8 Monocotyledonous, and 97 Dicotyledonousnbsp;plants. These, therefore, constitute more than one-half of the vegetation.^nbsp;The celebrated author remarks, on the geological relation indicated by thenbsp;characters of the plants, that it is not possible to determine it positively,nbsp;as the plants of the Cretaceous are, as yet, too little known. But he admitsnbsp;that the formation of Atane, considering its vegetable remains, is probablynbsp;referable to the lower Cenomanian.

As will be seen in the examination made of the age of the Dakota Group, from data shown in the table of distribution, its Flora seems to benbsp;somewhat more recent than that of Atane, though the relationship is verynbsp;close. The general character of the plants does not greatly differ, but thenbsp;number of the dicotyledonous plants is much greater, amounting in thenbsp;Flora of the Dakota Group to more than five-sixths of the vegetation.

In considering merely what is now known of the vegetation of the Middle Cretaceous (the Cenomanian of d�Orbigny), the first appearance,nbsp;and especially the prodigious development, of the Dicotyledons seems thenbsp;more wonderful that it is not a local phenomenon, but is remarked in thenbsp;formations of the same age over the whole Northern hemisphere. Wenbsp;cannot yet follow it in all the intervening land areas, but it has been tracednbsp;from Greenland to Vancouver Island to Canada, to Kansas, and Colorado,nbsp;and in Europe to Germany, therefore in about 40� N. latitude.

� These data are taken from Hesr's quot; Groenland Flora,� vol. vi, part 2.

CEETAOEOUS AND TEETIAET FLOEA.

With the limited acquaintance we have with the ancient Floras of the world it is not possible to account for the sudden appearance of the Dicotyledons in the Cretaceous time and for their rapid and wide distribution.nbsp;Saporta, justly considered as the botanist who has acquired by his vastnbsp;knowledge the most extensive views on the distribution of the vegetationnbsp;in the ancient epochs, says, on the subject:^ �The organic evolution tonbsp;which the Dicotyledons owe their existence and their distribution mustnbsp;have been produced under the influence of very different conditions. Itnbsp;is possible that the evolution has been originally slow and obscure; possibly also it has been accomplished in a concealed or as yet undiscoverednbsp;locality, in a separate region, and under the influence of peculiar local circumstances. It is probable that the change may have been accomplishednbsp;by the mediation of insects, multiplying at a given time the results ofnbsp;crossing and producing some combinations favorable to the growth of thesenbsp;plants. It is even conceivable that a short time may have been sufficientnbsp;to give origin to plants of this class under the action of causes which arenbsp;still unknown. Whatever hypothesis may be preferred, the fact of thenbsp;rapid multiplication of the Dicotyledons and of their simultaneous occurrence in many localities of the Northern Hemisphere from the beginningnbsp;of the Cretaceous Cenomanian cannot be contested.�

Yes, in this case, as in many others, we may collect facts, but the work of nature in its mode of proceeding for the creation or modification ofnbsp;species remains inscrutable. We may consider the formation of thenbsp;Dakota Group as produced by a very slow, gradual, prolonged depressionnbsp;of the Western slope of the continent, bringing up from the South ornbsp;West the invasion of ocean water charged with muddy materials, periodically heaped farther and farther inland by powerful tides. We may suppose, too, the invading flow as bringing with it seeds or fragments of rootsnbsp;of plants derived from a country now covered by the sea, and distributingnbsp;here and there those germs of vegetable organisms. But all this does notnbsp;account for much in the solution of the problem; it may explain the distribution; but the first appearance, and it seems the simultaneous multiplication, of the dicotyledonous plants remains a fact inconceivable to reason..

* �Le monde des Plantes,� etc., p. 197.

DESCRIPTION AND ENUMERATION OF SPECIES OF THE AMERICAN DAKOTA GROUP FORMATION.

I.

THALLOPHYTES.

ZONABITES, Brgt.

Zonarites digitatus. Gein.

� U. 8. Oeol. Rep.quot; vi, p. 44, pi. i, fig. 1.

The relation of this vegetable organism to that described by Brong-mart and Geinitz is contested on account of the habitat in a different Geological stage, Geinitz having described his plant from the Dyas. Asnbsp;species of Thallophytes of the Devonian are represented by identicalnbsp;forms in more recent formations, even in the Cretaceous of Europe, thenbsp;objection is not imperative.

ACROGENS.

EQUISETACE^.

EftUISETUM, Linn.

Equisetum nodosum, sp. nov.

Stems small, one-half to one centimeter in diameter, obscurely narrowly striate; articulations very inflated, marked with broad round scars of points of attachment ofnbsp;branches above the line of division.

The species is distantly related to E. amissum, Heer, �FI. Arct.,� Ill, p. 60, pi. xiii, figs. 2-8, of the Lower Cretaceous of Kome, essentiallynbsp;differing by the inflated articulations. It is represented by specimensnbsp;Nos. 473,536 of the Museum of Comp. Zool., Cambridge; they are too smallnbsp;and obscure for definite comparison.

Edb.�7 miles N. E. of Glasco, Kansas; collected by CJis. Sternberg.

35

DESCEIPTION OP SPECIES.

FILICES.

SPHENOPTERIS, Brgt.

Sphenopteris corrugata, Newby.

�Later Ext. Fl. of North America/� p. 10; �Illustr.,� pl. ii, fig. 6.

Prond unknown; pinnules ovate or cuneiform, narrowed at the base, obtuse, lobed, and often plicate longitudinally; nerves distinct, dichotomous, branching fromnbsp;the base.�(Newby.)

HYMENOPHYtLUM, Klf.

� nbsp;nbsp;nbsp;Hymenophyllum cretaceum, Lesqx.

�U. S. Geol. Rep./� vi, p. 45, pl. i, figs. 3-46; xxix, fig. 6.

In describing this species I related it to the preceding from the description given by the author, as I had not then seen the figures. Thesenbsp;indicate a degree of relation which cannot he positively ascertained onnbsp;account of the too fragmentary specimens. The fronds were evidentlynbsp;large in the plants of this kind. The divisions are multiple and extremelynbsp;variable. The specimens may, therefore, represent pinnules derived fromnbsp;divers parts of fronds of the same species.

PECOPTERIS, Brgt.

Pecoptoris Nebraskana, Heer.

�U. S. Geol. Rep./� vi, p. 46, pl. xxix, figs. 5, 5a.

GLEICHENIA, Sw.

Gleichenla Kurriana, Heer.

�U. S. Geol. Rep.,� vi, p. 47, pl. i, figs. 5-5c.

Gleiclienia NordenskiOldi, Heer.

Plate I, Figs. 1, la.

Hayden�s �Ann. Rep.,� 1874, p. 334, pl. ii, fig. 5.

Fronds slender, bi-polypinnate; ultimate pinnae alternate, rigid, open, linear, parallel; pinnules subcoriaceous, small, free, oblong-ovate, obtuse, rounded at base onnbsp;both sides, inclined upward; secondary veins few, three or four pairs, the lower forking,nbsp;the upper simple.

Though the American specimens of this species are small they show distinctly the essential characters of the species; the slender rachis of thenbsp;ultimate pinnte rendered flexuous by compression of the basilar border ofnbsp;the pinnules, the very small leaflets free and rounded at base, and thenbsp;disposition of the veins. The specimens which I have for examinationnbsp;are sterile. As seen and figured by Prof. Heer, the fructifications are those

FLORA OF THE DAKOTA GROUP.

of the subgenus Bidymosorus, Deb. and Ett., two sort placed upon the middle of the lower pair of veins, one on each side of the medial nerve.

The rachis of this fern is described by Heer as slender. As it is figured here it appears somewhat broad, though not larger than it is representednbsp;in Heer�s � Fi. Arct.,� iii, pi. ix, fig. 6. The ultimate rachis is, however, verynbsp;slender filiform.

Hah.�Fort Harker, Kansas. Chs. Sternberg.

LYGODIUM, Sw.

Lygrodlum tricliomanoides, Lesqz.

�U. S. Geol. Eep.,� vi, p. 45, pi. i, fig. 2.

PHOE]SroaA.M.^.

GYMNOSPERM^E.

ZAMI^.

PODOZAMITES, Pr. Br.

Fronds pinnate; leaves distant, obliquely or horizontally attached by an attenuated pedicelliform half-twisted flat base articulated upon the rachis and therefore caducous;nbsp;veins equal, longitudinal, converging to both ends of the leaves; borders entire.

This genus of Braun, as amended by Saporta and Schimper, seems adapted for the description of all the leaves of Gycadeae found as yet innbsp;the Dakota Group.

Podozamltes Haydenil, Lesqx.

Pterophyllum, Maydenii, Lesqx., � U. S. Geol. Eep.,� vi, p. 49, pi. i, figs. 6,66; Hayden�s �Ann. Eep.,� 1874, p. 334.

Nothing more definite is known of these vegetable fragments than has been published as quoted above.

Professor Heer, considering the thickness and impressions of the stems, regards these fragments as more probably referable to Conifers ofnbsp;the section of the Araucarites than to Oycadece. No leaves of this section,nbsp;however, can be compared to those which I have figured, and which, bynbsp;their parallel veins and forms, are very much like the leaves of somenbsp;species of Podozamites. Indeed, from the remarks on this genus by Heer,nbsp;the leaves are either narrowed and joined to the stems by decurring to it,nbsp;or produced into a short pedicel attached to the stem by small tuberclesnbsp;or warts. The characters of the genus are thus exactly shown not onlynbsp;by the leaf, but also by the stem whose round small scars indicate pointsnbsp;of attachment like tubercles.

DESCRIPTION OF SPECIES.

The leaves closely resemble those of Podozamites lanceolatus as figured by Nathorst, �FI. of Bjuf.,� pi. xvi, fig. 3. I think, therefore, that the fragments figured in �U. S. Geol. Rep.,� figs. 6 and Qh, should be referred tonbsp;this genus. The relation of the cone, however, which I referred to thenbsp;same species from its likeness to that of Stiehler as PteropJiyllum Prnestinoe^nbsp;is wrong, as it evidently represents a Conifer.

Podozamites oblongiis, Lesqx.

Plate I, Figs. 10, 11.

Leaves oblong, gradually narrowed from below the middle to the flat sessile base,, rounded at the eroded apex; veins thin, parallel, close, equal, distinct with the glass.

These leaves are evidently overturned upon the plate. The apparently truncate lower part seems as an enlarged point of attachment similar to that of species of Cordaites of the coal. But the irregular erosion is accidental or caused by compression of the macerated apex of thicknbsp;coriaceous leaves.

These leaves are of the same character as those of P. lanceolatus^ Schp., in Heer, �FI. Arct.,� iv, pi. vii, figs. 1-7 of the Jurassic Flora ofnbsp;Spitzbergen, differing by the more abruptly rounded apex.

Hah.�Dakota Group, Kansas. Chs. Sternberg.

Podozamites angiistifolius ? Heer.

Leaves long and narrow, somewhat falcate or ensiform, linear-lanceolate, gradually slightly narrowed upward from the middle, obtuse? (point broken) more rapidly downward from below the middle to the point of attachment, distantly veined; primarynbsp;veins obtuse, prominent; surface smooth, minutely lineate.

The preserved part of the leaves is 11 centimeters long, averaging 9 millimeters broad. The point of attachment is flat, 3 millimetersnbsp;broad. As the apex of the leaf is broken it is not possible to see if it isnbsp;acuminate.

The leaves figured by Heer, �FI. Arct.,� iv, pi. vii, figs. 8-11, and pi. viii, fig. 5, are either acuminate or somewhat obtuse and slightly scytheshaped, as in the one described here, but this is broader than any of thosenbsp;of Heer. The nervation seems like that of P. Eichwaldi, Heer, ibid., thenbsp;primary veins being broad, thick, or prominent, so that the surface appearsnbsp;undulate and the intervals marked by irregular or not continuous verynbsp;small veinlets. This leaf is also, from its shape and size, comparable to-

FLOEA OF THE DAKOTA OEOUP.

Podozamites ensis, Nath., �FI. of Bjuf.,� pi. xv, fig. 2. This, however, has the veins narrower and indistinct.

Hab.�South of Fort Harker, 4 miles east of Minneapolis, and 7 miles northeast of Glasco, Kansas. Chs. Sternberg.

Podozamites prselongus, sp. nov.

Leaf large, oblong, linear, narrowed gradually to the point of attachment, obscurely veined; primary nerves parallel and distinct.

The upper part of the leaf broken at 12 centimeters from the base is there 5 centimeters broad. The whole length appears to be about 16nbsp;centimeters. Its size is greater than that of any of the leaves of this genusnbsp;figured by authors, larger than the fragment of P. lanceolatus-latifolius,nbsp;Heer, �FI. Arch,� iv, pi. xxvi, fig. 6.

Though obtuse, the veins appear more distant and broader than in this last species.

Hab.�South of Fort Harker, with the preceding.

Podozamites emarginatus, sp. nov.

Leaves large, linear-oblong, gradually narrowed from below the middle to the flattened base, abruptly rounded and deeply emarginate at the apex; primary nervesnbsp;parallel,distinct or prominent, conjoining at the apex and the base, separated by thinnbsp;�disconnected veinlets.

The leaf is 14 centimeters long, 3i centimeters broad in the middle, the point of attachment 4 millimeters broad. It is abruptly rounded atnbsp;the top to 2 centimeters broad and there deeply obtusely emarginate, thenbsp;borders joining into a small obtuse sinus li millimeters wide.

The emargination of the top may be a casual deformation, but even if the apex was regular and obtuse this species is without marked affinitynbsp;to any other of the genus.

Hab.�Seven miles northeast of Glasco, Kansas.

Podozamites caudatus, sp. nov.

Leaf large, enlarged and oval in the middle, where it is 5 centimeters broad, rapidly narrowed to a point of attachment 7 millimeters broad, and attenuated from above the middle in rounding to a long acumen measuring centimeters broad at the pointnbsp;where it is broken 13 centimeters from the base.

The leaf has a peculiar form, being abruptly enlarged in the middle from above the base and as rapidly narrowed into a long linear acumennbsp;whose upper part is destroyed. The primary veins are flat and broad.

DESGEIPTION OP SPECIES.

distinct, half a millimeter distant, with an indistinct veinlet in the narrow intervals.

The form of this leaf is peculiar, without relation to any of this genus. Hob.�Near Fort Harker, Chs. Sternberg. No. 117 of the Nationalnbsp;Museum.

CONIFEER.

PHYLLOCLADUS, Bich.

Phyllocladus subintegrifolius, Lesqx.

quot; U. S. Geol. Rep.,� vi, p. 54, pi. i, fig. 12; Hayden�s �Ann. Rep.,� 1874, p. 337, pi. 2, fig. 4.

Thinfeldia Lesquereuxiana, Heer, � FI. Arct.,� vi, p. ii, p. 37, pi. xliv, figs. 9, 10.

Figure 4 of �Ann. Rep.� represents the lower half of a leaf of same character as that in �U. S. Geol. Rep.,� 1. c.

ARAUCARIA, Juss.

Araucaria spathulata, Newby.

�Notes on Ext. FI.,� p. 3; �Illustr.,� pi. ii, figs. 5, 5o.

Leaves close, broadly spatUulate, obtuse, narrowed above the enlarged base, carinate; medial nerve distinct at base, effaced from the middle upwards.�(Newby.)

The author remarks that the specimen represents a fragment of a branch nearly half an inch in diameter on which the leaves are thicklynbsp;set, their base slightly decurring scarcely separated from each other.nbsp;From their base the leaves, half an inch in length, radiate in all directions.

The species is closely allied to Abietites curvifolius, Dkr., of the Quader-sandstone of Blankenburg. This has the leaves rounded at thenbsp;apex, a deep medial nerve, and the leaf scars very distinct. This lastnbsp;character is well marked on the figure of A. spathulata. The same figurenbsp;shows the leaves reflexed or spreading at the base, the only part seen.nbsp;In Bunker�s species the leaves are curved up from the middle and arenbsp;longer.

Hob.�Sage Creek, Nebraska. Dr. F. V. Hayden.

TORREYA, Arn.

Torreya oblauceolata, sp. nov.

Plate I, Fig. 2.

Branches slender; leaves long, flat, gradually enlarging upwards from the decur* ring base; medial nerve thin.

The figure represents the best and largest of the fossil fragments.

FLOEA OF THE DAKOTA GEOUP.

None of them has a leaf entirely preserved, and thus the upper end of the leaf is undetermined.

From the decurring base of the leaves the fragment may represent a Sequoia. It has some analogy to S. Smittiana, Heer, �FI. Arct.,� iii, pi.nbsp;xvii, figs. 3, 4, while Torreyaparvifolia and T. DicJcsoniana, Heer, ibid., pi.nbsp;xvii and xviii, have the leaves sessile, and in this last species rounded andnbsp;enlarged above the point of attachment. The leaves of this fragment,nbsp;however, are too long for a species of Sequoia, also flat, not rigid nornbsp;coriaceous, and thus seem referable to Torreya.

Hab.�Cretaceous black shale, near Golden. Rev. A. Lakes.

SEQUOIA, Endl.

Sequoia Beichenbachl, Heer.

Lesqx., �U. S. Geol. Kep.,� yi, p. 51, pi. i, figs. 10, 105.

The supposed relation of the cone referred to this species is contradicted by Professor Heer. Though the cone represents a Sequoia, the specific name is left undetermined.

Sequoia fastigiata? St.

Hayden�s �Ann. Eep.,� 1874, p. 335, pi. iii, figs. 2, 8, 8a.

Branches slender, erect; branchlets filiform; leaves loosely imbricate, short, broadly lanceolate-acuminate, subfalcate or more or less Incurved, costate; strobilesnbsp;ovate-globose, small.

The fragments referred to this species are merely two short branchlets, pi. iii, fig. 8, loo. cit., and some indistinct cones imbedded into the stone.nbsp;The leaves appear to be of the same form as those of this species figurednbsp;by Heer, �M�let. FI.,� pi. i, fig. 10, generally a little broader and shorter,nbsp;and the cones have the same character as that of fig. 12 of the samenbsp;plate. These fragments are also comparable to the species as figured innbsp;Heer, �FI. Arct.,� iii, pi. xxvii, figs. 5 and 6, of the Upper Cretaceousnbsp;of Atane. Professor Heer says, in the first description of this species,nbsp;�Molet. FI.,� 1. c., that the leaves do not seem to have any medial nerves,nbsp;and in �FI. Arct.,� loc. cit., he remarks on the difference of the speciesnbsp;from S. rigida by the absence of a medial nerve. As the costa is distinctnbsp;on the leaves of the Dakota Group the relation is doubtful.

Hab.�Kansas, Clay Centre. H. C. Towner.

DESOEIPTION OF SPECIES.

Sequoia condlta, Lesqx.

Plate I, Fige. 6, 7, 9.'

Hajden�i quot;Ann. Eep.,� 1874, p. 335, pl. ir, figs. 2-7.

Branches rigid, pinnately divided; branchlets slender, filiform, oblique; leaves short, oblong, thick, not pointed, narrowed to the decurring base, appressed to thenbsp;stem, sometimes longer linear-acuminate, curved inward, nerveless; male ament oval,nbsp;scaly, rhomboidal, apiculate.

This species is not rare in the shale of the Dakota Group, but as yet it has been found always imbedded into the shale and in small fragments,nbsp;so that its characters cannot be stated with precision. Generally the leavesnbsp;are lineal-oblong, acute, appressed to the stem, variable in length, sometimes longer, curved inward, resembling those of S. fastigiata, the medialnbsp;nerve being indistinct. The cone, fig. 9, found upon specimens withnbsp;branches of the species, is apparently an unopened fruiting catkin of thisnbsp;species. It has a slender short pedicel covered with very small obtusenbsp;scale-shaped leaves.

Hai.�Kansas; not rare.

GLYPTOSTEOBUS, Endl. �

Glyptostrobus gracillimus, Lesqx.

PUte I, Figs. 6-65.

quot;�. S. Geol. Kep.,� vi, p. 52, pl. ii, figs. 8, ll-ll/; Hayden�s quot;Ann. Eep.,� 1874, p. 337.

I have figured here a mere fragment which I consider referable to the species, though the branch is a little thicker and the leaves ovate, somewhat like those of Sequoia condita, but shorter, as may be seen in comparing both figs. 5a and 6a.

The leaves of this plant and their disposition are remarkably similar to those of Cyparissidium gracile, Heer, �FI. Arct.,� iii, p. 74, pl. 19, fig. i,nbsp;found at Kome and Atane.

THVITES, Stemb.

Thultes crassus, sp. nov.

Pinnately branching; branches comparatively thick, alternate; branchlets short, -obtuse; leaves thick, broadly oblong, equilateral, as broad as long, closely imbricatenbsp;in four rows; medial nerve distinct, inflated on the back.

Species closely allied to Thuites Meriani, Heer, �FI. Arct.,� iii, p. 73, pl. xvi, figs. 17, 18, differing especially by the great thickness of its branches,nbsp;the leaves larger, broader, the facial and lateral of the same size.

Hob.�Seven miles northeast of Glasco. Chs. Sternberg.

PLOEA OP THE DAKOTA GEOXJP.

PmUS, Linn.

Plnus Qiienstedti, Heer.

PLito I. Fi-s. 3, 4.

Hayden�s �Ann. Eep.,� 1874, p. 336, pl. iii, figs. 6, 7.

Leaves iu fassicles of five, very loog and slender, thread-like, deeply nerved, the �base inclosed in long cylindrical sheaths j cone cylindrical, very longj scales withnbsp;broad rhomboidal shields (apophyses), acute on the sides, mammillate in the center.

The specimens representing this species are numerous but all fragmentary. The leaves are generally scattered and imbedded close together, their point of attachment by five is marked by the long sheaths formingnbsp;deep holes into the stone; but none has been thus far found preservednbsp;entire. The species may be, therefore, different from that of Heer.nbsp;described as above, and figured in �Molet. FL,� p. 13, pl. ii, figs. 5-9, Thenbsp;thread-like long leaves, the long cylindrical cone, and the shields of thenbsp;scales are, however, so much alike that I have scarcely any doubt on thenbsp;identity of the Dakota Group species with that of Europe. The length ofnbsp;the leaves as given by Heer, who has had splendid specimens for description, is 20 centimeters. The fragments I have seen are 5 to 8 centimeters.nbsp;The cylindrical cone, 22 millimeters broad, gradually tapering to the base,nbsp;appears to be very long, its impressions perforating large stones, beingnbsp;at least 15 centimeters long. These cones are generally curved as in fig.nbsp;iii. Heer represents them straight but of the same length and width.

In the �Flora of Gelinden� by Saporta and Marion, the authors remark (p. 19) ,that this fossil species does not differ by any importantnbsp;�Character from the living Mexican Pines with quinate leaves which nownbsp;compose the section of the Pseudo-strohus.

Hah.�Near Fort Harker and Clay Centre, Kansas. Chs. Sternberg and H. C. Towner.

fbaome:st8 of conifers of uncertain relation.

Abietltes Ernestinse, Lesqx.

' � U. S. Geol. Eep.,� vi, pl. i, fig. 7.

Sequoia formosa, Lesqz.

�U. S. Geol. Eep.,� vi, pl. i, figs. 9, 9a.

Inolepis? species.

Plate I, Figs. S-Sc.

� Hayden�s � Ann. Eep.,� 1874, p. 337, pl. iv, fig. 8.

Nutlets small, globular, short-mucronate, sessile upon slender branches, c v 3

DESCRIPTIOISr OF SPECIES.

The specimen, fig. 8, shows the impression of three unopened globular, naked nutlets, which, as seen in figs. 8a and 8c, appear to contain smallnbsp;seeds which, in fig. 8c, are obcordate and inflated. These three last figuresnbsp;ar� all spread upon the same specimen with fig. 8.

The relation of this fragment to Inolepis is not certain. The fruits found mixed with a mass of decayed and broken remains of conifers maynbsp;be considered as indeterminable, even in their generic relation, until betternbsp;specimens are obtained.

Hal.�Dakota Group of Kansas.

MONOCOTYLEDONES.

GLUMACE^.

PHBAGMITES, Adans.

Phrag^mites cretaceus, Lesqx.

�U. S. Geol. Kep./� vi, p. 56, pi. i, figs. 13 and 14; pi. xxix, figs. 7, 7o.

Leaves and culms in fragments of various sizes; leaves lanceolate, blunt at the apex, doubly nerved; primary nerves thick or inflated under the thick epidermis, undernbsp;which the intermediate veinlets, three or four, are discernible.

The fragmentary state of the first specimens found afforded reasonable doubt of their reference to this genus. But remains of plants of analogous character have been since discovered in the Upper Cretaceousnbsp;of Greenland and described as Arundo Gr�nlandica, Heer, �FI. Arct.,� iii,nbsp;p. 104, pi. xxviii, figs. 8-11. In this species the primary nerves do notnbsp;appear separated by any intermediate veinlets, at least none could benbsp;observed by Professor Heer. This difference, and also the great size ofnbsp;the leaf, pi. xxix, fig. 7 of the Rep., 1. c., evidently separate the species.

DIOSCOEEiE.

DIOSCOREA, Plum.

Di oscorea? cretacea, Lesqz.

�U. S. Geol. Eep.,quot; vi, p. 66, pi. xxviii, fig. 10.

PALM^.

FLABELLABIA, St.

Flabellaria? minima, Lesqx.

�U. S. Geol. Rep.,� vi, p. 56, pi. xxx, fig. 12.

FLOEA OF THE DAKOTA GROUP. nbsp;nbsp;nbsp;35

DICOTYLEDONES.

MYEICACEiE.

KYRICA, Linn.

Myrica obtusa, Lesqx.

� U. S. Geol. Eep.,quot; vi, p. 63, pi. xxix, fig. 10.

Myrica Dakotensis, Lesqx,

Plate IV, Fig. 9.

M. cretacea,^ Lesqx., Hayden�s �Ann. Eep.,� 1874, p. 339, pi. ill, fig. 4.

Leaves narrowly lanceolate or lineal-oblong, gradually narrowed to a thick short petiole, crenulate on the borders; medial nerve flat and broad; lateral nerves at annbsp;acute angle of divergence, parallel, variable in distance, camptodrome; tertiary veinsnbsp;short, anastomosing with the secondary ones by nervilles at right angles.

The substance of the leaves is thick, coriaceous, the surface polished, the borders slightly reflexed and crenulate; the upper end of both thenbsp;leaves representing the species is destroyed, but on fig. 9 the apex seemsnbsp;rounded or obtuse. They are 7 to 8 centimeters long and about l� centimeters broad in the middle.

The relation of these leaves is more distinctly marked with M. SchenJci-ana, Heer, �Quedl. FI.,� p. 11, pi. iii, fig. 1, and less distinctly with M. cretacea, Heer, ibid.,^. 10, pi. iii, figs. 2a, h, c. They are more lineal, thenbsp;borders less deeply and more closely denticulate, the veins closer, etc.nbsp;By their hard texture and their nervation the leaves are also comparablenbsp;to those of some tropical species of Salix.

Hah.�Fort Harker, Kansas. Chs. Sternberg.

Myrica Sternbergil, sp. nov.

Leaves long, linear-lanceolate; borders distantly obtusely dentate; lateral veins at acute angles of divergence, comparatively thick, flexuous, simple or forking abovenbsp;the middle, the divisions entering the teeth in curving along the borders.

The specimen represents only a fragment of a leaf whose upper and lower parts are destroyed. The fragment is 7 centimeters long and 3nbsp;broad; the secondary veins or their primary divisions enter the teeth by

^ Myrica cretacea. Heer, was, perhaps, published iu the � Flora of Quedlinburg � before I described my species under the same name. But that work of Heer was then unknown to me. I change name, not beingnbsp;certain who has priority for it.

DESCRIPTION OF SPECIES.

their ends or by anastomosing branches, diverging under the teeth and following the borders. By the size of the leaves and the distant obtusenbsp;teeth this species is related to M. Thulensis, Heer, �FI., Arct.,� iii, p. 107,nbsp;pi. xxxi, fig. i; also closely allied by the nervation to M. apiculata, Sap.nbsp;�S�zanne FI.,� p. 342, pi. iv, fig. 5.

Hob.�Two and a half miles north of Glasco, Kansas. Chs. Sternberg.

Myrica? semina, Lesqz.

�U. S. Qeol. Bep./' Ti, p. 63, pi. zxvii, figs. 4, 4a.

BETULACEJE.

BET�LA, ToTim.

Betula Beatrlciana, Lesqz.

�IT. S. Geol. Bep.,� vi, p. 61, pi. v, fig. 5; pi. xxx, fig. 4.

Betulites dentlculatus. Heer.

� Phyll. Cr�t. da Neb.,� p. 15, pi. iv, fig�. 5, 6.

Leaves short, ovate, denticulate, rounded at base; lateral nerves diverging in acute angle, craspedodrome, straight.

The craspedodrome nervation relates these leaves to Betula ov Ahms, though the form of the leaves recalls the type of Populus.

Phyllites betulsefolius, Lesqx.

� U. S. Geol. Bep.,� vi, p. 112, pi. xxviii, figs. 4, 7.

ALNITES, Goepp.

Alnltes grandlfollus, Newby.

�Note* on Ext. FI.,� Catal., p. 9; �IlluBtr.,� pi. iv, fig. 2.

The species is not described by the author. The figure represents a large, round-oval leaf, narrowed to the petiole; the borders deeply regularly undulate, the lateral nerves at an acute angle of divergence, muchnbsp;branched on the lower side, craspedodrome like all the divisions.

The leaf represents the type of Alnus glutinosa by its form and size; but the borders are entire, merely undulate, not denticulate. The base ofnbsp;the leaf is more acutely narrowed to the petiole than it is generally innbsp;the leaves of this genus.

Hob.�Nebraska. Dr. F. V. Hayden.

FLORA OF THE DAKOTA GROUP.

CUPULIFBEJE.

FAGUS, Toxixii.

Fagus polyclada, Lesqx.

�U. 8. Qeol. Eep.,� vi, p. 67, pi. v, fig. 6.

Fagus cretacea, Newby.

Plate II, Figs. 6, 6o.

Newby., �Notes on Ext. FI.,� p. 23; �Illustr.,� pi. ii, fig.3.

Leaf oval, entire, slightly narrowed to the petiole; lateral veins sharply defined, numerous, parallel, craspedodrome, the points of the nerves being prominent and thenbsp;intervals between them forming shallow sinuses.

To offer a point of comparison between this leaf and the one described as Fagm polyclada I have figured it again, distinctly tracing the nervilles,nbsp;which are not visible on the original figure. The type of venation is thatnbsp;of Fagm not of BJiamnus; the marginal veinlets only are more distinctnbsp;than in F, ferruginea, following the border in short curves anastomosing tonbsp;the upper nervilles, nearly as in Castanea.

Hah.�Smoky Hill, Kansas. Dr. F. V. Hayden.

DRYOPHYLLUM, Debey.

Leaves lanceolate or oblong, generally dentate, penninerve; secondary nerves (in denticulate leaves) sub-opposite, straight, simple, entering the teeth directly by theirnbsp;points, or more rarely branching quite near the borders, one of the divisions enteringnbsp;a tooth, the other curving under, following the margins in wavy flexures and joiningnbsp;the next vein above. In the entire leaves the secondary nerves are more or less curved,nbsp;camptodrome, with nervilles transversely decurrent, simple or forking, united by venulesnbsp;at right angles. (Sap.)

Saporta in describing the genus compares the species referred to it to some kinds of Oaks and Chestnuts with coriaceous leaves, now inhabitingnbsp;the mountains of Asia and of Mexico, and which seem to have been thenbsp;ancestors of the Oaks and Chestnuts of the present Flora of North America.

Dryopbyllum (Quercus) primordiale, Lesqx.

�U. 8. Geol. Eep.,� vi, p. 64, pi. v, fig. 7.

Dryophyllum (Quercus) latifolium, Lesqx.

Plate IV, Figs. 1, 2.

Hayden�s �Ann. Eep.,� 1874, p. 340, pi. vi, fig. 1.

Leaf large, broadly ovate, rounded at base, deeply sinuate, obtuse or blunt at the apex; medial nerve thick; secondary nerves distant, straight or slightly curvingnbsp;up to the borders, the lower more or less branching.

DESGEIPTION OF SPECIES.

The fine leaf, fig. 1, is nearly 12 centimeters long and 9 broad in the middle, its widest part, coriaceous, deeply undulate. The nervation isnbsp;thick and coarse, the secondary nerves, 8 pairs, alternate, diverging at annbsp;angle of 50� are parallel, except a pair of basilar, thin, short marginalnbsp;veinlets which, nearly at right angles to the medial nerve, follow close tonbsp;the borders. The lower veins are more or less branching and enter thenbsp;undulations or broad obtuse teeth, somewhat less prominent at the endsnbsp;of the branches than at those of the nerves. The nervilles thin, but verynbsp;distinct, are flexuous, at right angles to the veins, more generally continuous. Fig. 2 is an incomplete fragment which I consider as representingnbsp;the same species. The upper end of the veins and of their branches arenbsp;connected by strong nervilles following close to the borders; but they arenbsp;not subdivisions of the secondary veins. This nervation is like that ofnbsp;Castanea and of some species of Quercus.

Hob.�Fort Barker, Kansas. Chs. Sternberg.

Dryophyllum (Quercus) Holmesil, Lesqx.

Plate rV, Fig. 8.

Dryophyllum {Quercus) saUdfolium} Lesqx., Hayden�s �Ann. Rep./' 1874, p. 340, pi. viii, fig. 2.

Leaf linear-lanceolate, rounded in narrowing to the base, minutely acutely denticulate; lateral nerves numerous, parallel, alternate or opposite, slightly bowed sub-camptodrome.

The fragment represents a slightly falcate somewhat thick leaf, rather membranaceous than coriaceous, with a narrow medial nerve and closenbsp;parallel secondary ones, some of them as far as can be seen ascending tonbsp;the teeth and passing under the sinuses by an upper branch, some othersnbsp;curving along the borders and reaching the teeth by short branchlets.nbsp;This species is related to Dryophyllum lineare, Sap., �S�z. FL,� p. 350, pi.nbsp;iv, fig. 6. The teeth, however, of the American species are more distinct,nbsp;turned outside in the lower part of the leaf, inclined upward in the uppernbsp;part, as in D. subcretaceum of the same author, ibid., p. 348, fig. 10.

Hob.�Near the San Juan River, at a higher Cretaceous stage than that of the Dakota Group; Southwest Colorado. W. H. Holmes.

� Name preoccupied aa Quercus salieifolia, Newby., � Ext. FI.,� p. 24.

FI^OBA OF THE DAKOTA GROUP.

aUEBCUS, loan.

Quercus Dakotensls, sp. nov.

Leaf subcoriaceous, ovate-lanceolate, narrowed in rounding to the base, less abruptly, liowever, to an acute or blunt apex (not distinct), entire on the borders towardnbsp;the base, nearly regularly dentate from below the middle upward, short pedicellate;nbsp;medial nerve straight; secondary nerves thin, slightly bowed, divided into two or threenbsp;branches, each entering a tooth.

The leaf is 9 centimeters long and 4J centimeters broad in the middle; the point not distinct appears blunt; the pedicel is slender, nearly 1 centimeter long as far as it is seen before entering the stone; the secondarynbsp;nerves diverge 55� to 60�.

The species is related to Quercm BeyricUi, Ett,, �Kreidefl. von Nied-ers.,� p. 14, pi. ii, fig. 2, from which it differs by the teeth not being turned upward or serrate, but abruptly acuminate outward; by the texture, whichnbsp;is not distinctly coriaceous; by thin secondary nerves and a narrow straightnbsp;midrib. The upper veins are under the same angle of divergence, craspe-dodrome; the lowest pair, attached a little above the base of the leaf,nbsp;follows the entire border up to the lower teeth. This species has also anbsp;degree of affinity to Castama Hausmanni, Dkr., �Paleont.,� iv, p. 181, pi.nbsp;xxxiv, fig. 1. The teeth are of the same character.

These three species may be referable to the preceding genus, but the tertiary divisions of the veins are not discernible in any of them.

Hah.�South of Fort Harker. Chs. Sternberg. No. 62, Mus. Comp Zool., Cambridge.

Quercus hexagona, Lesqx.

�U. S. Geol. Eep./� vi, p. 64, pi. v, fig. 8.

This leaf, to which I could not indicate any related form when I described it {1. c.), is, in shape especially, allied to the Oligocene Quercusnbsp;Osbornii, pi. xxxviii, fig. 17, which, itself, is comparable to Quercus tephrodes,nbsp;Ung., as figured in Sieber, �Nord-B�hm Braunkohl. FI.,� iii, fig. 17.

Quercus Ellsworthlana, Lesqx.

quot; U. S. Geol. Eep.,quot; vi, p. 65, pi. vi, fig. 7.

Another specimen referable to this species, as yet insufficiently represented and described, is a leaf of the same size and form as that of the �U. S, Geol. Rep,,� 1. c. The nervation is of the same character, at least

DBSCEIPTION OF SPECIES.

for the distance and the ramification of the secondary nerves; the lower ones only are more open and more bowed in passing to the borders, thenbsp;lowest pair being nearly at right angles to the thick medial nerve. Thenbsp;specimen is No. 1175 of the U. S. National Museum.

Quercus poranoides, Lesqz.

� �. S. Geol. Kep.,� vi, p. 66, pi. xxx, fig. 9.

The generic relation of this fragment, like that of the preceding, is not positively ascertained.

Quercus Morrlsoniana, sp. uov.

Plate XVII, Pigs. 1, 2.

Leaves of medium size, coriaceous, petiolate, ovate-lanceolate, acuminate; medial nerve strong; secondary nerves numerous, alternate, curved in passing to the borders,nbsp;camptodrome, simple, or some of them forking near the entire borders.

The species is related by its characters, shape, size, facies of the leaves,, and nervation to the Miocene Quercus neriifolia, A. Br. The midrib isnbsp;strong, prolonged into a petiole 11 centimeters long. The lower veins arenbsp;slightly more open than the upper; all are nearly parallel, variable innbsp;distance, more or less bowed in passing to the borders, which are verynbsp;entire. The leaves average 10 to 12 centimeters long, 3 to 31 centimetersnbsp;broad in the middle where they are the widest, gradually narrowing in anbsp;curve to the base and slightly decurring to the petiole.

The embedding material is a sandstone too coarse for the preservation of the areolation; flexuous nervilles, transversely decurrent, are moi�e ornbsp;less distinct. By this character the leaves are related to Q. nervosa, Sap.,nbsp;��t.,� ii, i, p. 86, pi. iii, fig. 12.

Hah.�Base of the mountains, near Morrison, Colorado. H. C. Beckwith.

Quercus sallclfolia, Newby.

�Kotes on Ext. FI.,� p. 24; � Illustr.,� pi. ii, fig. 1.

Leaves jietiolate, smooth, thick, entire, abruptly pointed at both ends; medial' nerves strong, straight or flexuous; secondary veins unequal in size, strong near theirnbsp;base, becoming finer, flexuons, and branching toward the borders, where some of themnbsp;inosculate by irregular curves while others terminate in the margin.

The facies of the leaf and the alternation remarked by the author of large with smaller secondary veins, a character essentially pertaining to thenbsp;willows, seem to justify the reference of this leaf to Salix. The coriaceous

FLORA OF THE DAKOTA GROUP.

texture of the leaf and its smooth surface do not contradict this reference; for all the species of willows of the Dakota Group are coriaceous, as arenbsp;generally the willows of the tropical or warm regions.'

Hah.�Blackbird Hill, Nebraska. Dr. Hayden.

Quercus cnneata, Newby.

�Notes on Ext. FI./' p. 25.

Leaves short, petiolate, lanceolate, pointed at both ends, acute, entire, or slightly wave-margined; midrib strong; secondary veins remote, nearly straight, with shortnbsp;intermediate ones; surface smooth, texture originally thick and coriaceous. (Ky.)

The author compares this species to Q. imbricaria, Michx., for the

form and consistence of the leaves.

Hah.�Blackbird Hills, Nebraska. Dr. Hayden.

Quercus antlqua, Newby.

�Notes on Ext. FI./� p. 26.

Leaves of medium size, lanceolate in outline, acute, often somewhat flexuous; margins serrate-dentate, with strong obtuse teeth, which are appressed or turnednbsp;upward; midrib strong, percurrent; secondary veins numerous, of unequal strength,nbsp;arched upward, craspedodrome. (Ny.)

Hob.�Lower Cretaceous sandstone, Banks of the Rio Dolores, Utah.

Quercus sinuata, Newby.

�Notes on Ext. F]./'p. 27.

Leaves small, ovate in general outline, narrowed to the petiole or slightly decurrent; margins deeply lobed; lobes rounded, broader than the sinuses that separate them, three, nearly equal on either side; summit broadly rounded or obscurely lobed,nbsp;often oblique; midrib straight or slightly flexed; secondary veins strong and simple,nbsp;running to the margin of each lateral lobe. (Ky.)

The author compares the species to the living Q. obtusiloba, Michx. Hob.�Same as the preceding.

SALICINEJE.

SAIIX, Toum.

Salix nervillosa. Heer.

�PhyU. Cr�t. du Neb./� p. 15, pi. i, fig. 3.

Leaves oblong, lanceolate, very entire; secondary veins in an acute angle of divergence, curved, camptodrome; nervilles curved, at right angles to the midrib.

Hab.�Nebraska. Dr. Capellini.

DESCEIPTION OF SPECIES.

Sallx protesefolia, Lesqz.

Plate I, Figs. 14-16; XVI, Fig. 3.

� U. S. Geol. Eep.,� vi, p. 60, pi. v, figs. 1, 4.

The leaf figured (pi. xvi) is related to this species merely by its form, resembling that of pi. v, fig. 4, of the �Report,� 1. c. The nervationnbsp;is indistinctly preserved, as in fig. 2 of the same plate. The other leavesnbsp;(pi. 1, figs. 14^16) are all much narrower but broader toward the base, andnbsp;gradually tapering to a long point. They have the same kind of venationnbsp;and merely represent modified forms of this extremely variable species.

Hob.�Kansas, near Fort Barker.

Sallx Meekii, Newby.

� Later Ext. FI.,� p. 19; � Illustr.,� pi. i, fig. 1.

Leaves petioled, thin and delicate, lanceolate, acute at both ends, entire; midrib slender; secondary nerves fine, in an acute angle of divergence 35�, gently arched andnbsp;anastomosing near the margins. (Ny.)

This is apparently the same species as the preceding, which, with an apparent difference in the texture of the leaves, the more or less acutelynbsp;narrowed base, the great variety of size of the leaves, includes also thenbsp;two following forms:

Hal).�Blackbird Hills, Nebraska. Dr. Hayden.

Sallx cuneata, Newby.

�Later Ext. FI.,� p. 21; �Illustr.,� pi. i, figs. 2, 3.

Leaves of medium size, sessile or short petiolate, lanceolate, acute at both ends, broadest toward the apex, gradually narrowed below to the base; medial nerve distinct; secondary veins delicate, with an acute angle of divergence (20�), gently archednbsp;above and inosculating near the margin. (�iy.)

The figures show the leaves larger in the middle, not toward the apex; they are more rapidly narrowed to the base and abruptly curve tonbsp;the petiole in reaching it.

Hah.�Mouth of Sioux River, Nebraska. Dr. Hayden.

Sallx flexuosa, Newby.

�Later Ext. FI.,� p. 21 ; �Illustr.,� pi. i. fig. 4.

Leaves narrow, lineair, pointed at each end, sessile or very short petioled; medial nerve strong, generally somewhat flexuous; secondary veins diverging about 40�, somewhat branched and flexuous, curving and inosculating near the margins, (Ny.)

The author considers this as a variety of S. Meekii.

Hah.�Blackbird Hills, Nebraska. Dr. Hayden.

FLOEA OF THE DAKOTA GEOUP.

POPULUS, Liim.

Populus litlgriosa. Heer.

quot;Phyll. Cr�t. du Neb.,quot; p. 13, pi. i, fig. 2; Newby., �Illustr.,� pi. iv, fig. 1.

Leaves round in ontline, very entire at base; the two pairs of lower lateral veins opposite, the other alternate and distant; nervilles curved, simple or forking. (Hr.)

Hah.�^Tekamah, Nebraska. Dr. Capellini.

Populus elllptica, Newby.

quot;Later Ext. FI.,quot; p. 16; quot;Illuetr.,quot; pi. iii, figs. 1, 2.

Leaves long-petioled, suborbicular or transversely elliptical, slightly cuneate at the base and apiculate at the summit; lower half of leaf entire, upper half or morenbsp;very regularly and rather finely obtusely serrate or crenate, the points of the teethnbsp;inclining upward; primary nerves usually fine, sometimes three, radiating from thenbsp;base at equal angles; from them the secondary veins spring at acute angles. (Ky.)

The species is remarkably similar, by the characters of the leaves, to P. cuneata, Newby., loc. cit., p. 64, pi. xiv, figs. 1, 4, a Miocene species ofnbsp;the type of P. arctica, or is, perhaps, one of its numerous varieties.

Hah.�Blackbird Hills, Nebraska. Dr. P. F. Hayden.

Populus micropliylla, Newby.

quot;Later Ext. FI.,� p. 17; �Illustr.,� pi. iU, fig. 5.

Leaves very small, scarcely an inch in length, broadly cuneate at the entire base, rounded and deeply dentate from the middle upward; teeth conical, acute or blunt atnbsp;the apex; nerves finely radiating from the base, branching above, the branches entering the teeth. (Ny.)

Hah.�Same as the preceding. Dr. P. V. Hayden.

Populus? cordifolia, Newby.

quot;Later Ext. FI.,� p. 18; � Illustr.,� pi. iii, fig. 7.

Leaves heart-shaped, slightly decurrent on the petiole; margins entire; nervation fine but distinctly defined; medial nerve straight or slightly curved, running to thenbsp;margin; lateral nerves, six on each side, diverging about 50�, nearly parallel, straightnbsp;or slightly curved near the apex, the lower branching; nervilles at right angles or forking, rarely continuous. (Ky.)

Hob.�Same locality as the preceding. Dr. P. F. Hayden.

DESCEIPTIOJS� OF SPECIES.

POPULITES, Lx.

Populites liancastriensis, Lesqz.

� U. S. Geol. Rep./' vi, p. 58, pl. iii, fig. 1.

Populites elegaus, Lesqz.

� U. S. GeoL Rep.,� vi, p. 59, pl. iii, fig. 3.

Populites cyclopliylla? Heer.

� U. S. Geol. Rep.,� vi, p. 59, pl. iv, fig. 5.

From a remark of Professor Heer, this leaf is not referable to his Populus cyclophylla, as I supposed it. Indeed, from the craspedodromenbsp;nervation, this leaf is rather a Cissites than a Pojpulus. Its relation is asnbsp;yet undefined.

PLATANEN.

PLATANUS, Linn.

Platanus Newberryana, Heer.

�U. S. Geol. Rep.,� vi, p. 72, pl. viii, figs. 2, 3; ix, fig. 3.

Platanus obtusiloba, Iiesqz.

�U. 8. Geol. Rep.,� vi, p. 69, pl. vii, figs. 3, 4.

Platanus primaeva, Lesqz.

� U. S. Geol. Rep.,� vi, p. 69, pl. vii, fig. 2; xxvi, fig. 2.

Platanus Heerii, Lesqx.

Plate III, Fig. 1; VII, Fig. 5.

�U. S. Geol. Rep.,� vi, p. 70, pl. viii, fig. 4; ix, figs. 1, 2.

Fig. 1 of pl. iii represents a fragment of a merely undulate, not lobate, leaf. The nervation has the normal character; the petiole is longer thannbsp;I have seen it in any other specimen of this species. Another*leaf, preserved entire, seen in the Museum of Comp. Zool., Cambridge, No. 225, isnbsp;still smaller than this one, only 6 centimeters long and 5i broad. It hasnbsp;the same nervation, the borders more deeply undulate, and two short,nbsp;rather acute, lateral lobes. Fig. 5 of pl. vii is still a smaller form of thisnbsp;same species.nbsp;nbsp;nbsp;nbsp;quot;quot;

The leaves of this species have been found at two different localities on the Salina River and near Fort Barker, Kansas.

Platanus diminutiva, Lesqz.

quot;U. S. Geol. Rep.,� vi, p. 73, pl. viii, fig. 5.

FLORA OF THE DAKOTA GROUP.

LIQUIDAUBAB, Lion.

liiquidambar integrifollum, Lesqx.

Plate XIV, Fig. 3.

�U. S. Geol. Eep.,� W, p. 56, pi. ii, figs. 1, 3; xxiv, fig. 2.

There is a degree of uncertainty in regard to the relation of the leaves described under this name, as I have remarked it in the �U. S. Geol. Rep.,�nbsp;1. c. If on one side they are related by their forms, especially the entirenbsp;margin, to species of Aralia, or perhaps more of Sterculia, their nervationnbsp;has more analogy to that of Liguidamhar than to any other of the groupsnbsp;to which they have been compared. Two well-preserved specimens of thenbsp;Museum of Comp. Zool., Cambridge, show the secondary veins somewhatnbsp;variable in distance and divergence, moderately curving to quite near thenbsp;borders, where they abruptly bend, following upward to the point wherenbsp;they anastomose in simple festoons. They are separated by short tertiarynbsp;veins parallel to the secondary ones, dividing in the middle of the areasnbsp;in joining the borders at right angles as nervilles. I have not observednbsp;this character in any of the fossil leaves which I have described as Aralia,nbsp;nor do I find it in the few living species which I have for comparison.nbsp;Another point of relation is remarked in the sub-cordate base of the leavesnbsp;of the cretaceous species which, like Liguidamhar Styracifim and the common Miocene species L. Lkiropceum, have the lower lateral lobes eithernbsp;curved back or at right angles to the petiole, so that the base of the leafnbsp;is never cuneate.

M 0 US .

FICUS, Lmn.

Ficus primordlalls. Heer.

� Phyll. Cr�t. du Neb./' p. 16, pi. iii, fig. 1.

Leaves coriaceous, lanceolate, narrowed to the base, very entire, penninerve; lower pair of secondary veins at a very acute angle of divergence from the midrib, thenbsp;others more open, all camptodrome.

I refer to this species two specimens (Nos. 26 and 33, Museum Comp. Zool., Cambridge), representing: the one, the impression of the upper surface of a lanceolate or oblong-lanceolate leaf, same size and shape as that of

DESCRIPTION OF SPECIES.

Heer, with base and top also destroyed. The midrib is narrow, the secondary veins thin, the lower pair at a more acute angle of divergence; but the divergence of those above is gradually more obtuse, not abruptly so, as isnbsp;the leaf of the �Phyllites.� The veins are close, 5 to 6 millimeters distant, not decurring to the medial nerve, slightly arched in passing upnbsp;toward the borders, which they follow in curves, anastomosing by nervillesnbsp;at right angles with the anterior veins.

The other specimen bears the impression of the under surface of a leaf and the upper part of two others, these tapering into a long acumen. Thenbsp;areolation is very distinct, exposing a coarse reticulation composed of large,nbsp;irregularly quadrate areas divided into small polygonal meshes.

These leaves have great affinity to those of the following species; they differ by their shape, oblong in the middle, by the secondary veins beingnbsp;more distinct, especially near and along the borders; the areolationnbsp;appears to be of the same character.

Hah.�South of Fort Harker. CTis. Sternberg.

Ficus Halliana, Ijesqz.

� U. S. Geol. Rep./' tI, p. 68, pi. xxviii, figs. 3, 9.

Ficus Beckwithii, sp. nov.

Plate XVI, Pig. 5; XVII, Figs. 3, 4.

Leaves sub-coriaceous, lanceolate or oblong, very entire, narrowed upward to a long acumen, more rapidly downward from below the middle and slightly decurring tonbsp;the petiole; midrib strong, gradually thicker toward the base; secondary veins numerous, parallel, camptodrome; nervilles close, flexuous, and sub-continuous, at rightnbsp;angles to the veins.

This species is of the same type as F. protogcea, Heer, �FI. Arct.,� iii, p. 108, pi. XXX, figs. 1-8, differing by the form of the leaves, which innbsp;F. protogoea are oblanceolate or largest toward the apex. The veins arenbsp;closer, though at the same angle of divergence, simple, 7 to 8 millimetersnbsp;distant, at an angle of divergence of 36�. The nervilles also are very closenbsp;and distinct. The leaves average 15 centimeters in length and to 4nbsp;centimeters broad below the middle.

Hah.�Near Morrison, Colorado. H. C. Beckwith.

FLOEA OF THE DAKOTA GEOUP.

Ficus? angustata, ep. nov.

Leaves narrowly lanceolate, comparatively long; medial nerve thick; secondary veins very close, at an acute angle of divergence, camptodrome.

The leaves are long, lanceolate, gradually acuminate, and also gradually narrowed to the base, 12 centimeters long, 2 centimeters broad. The veins at an angle of divergence of 20� pass upward slightly curved, andnbsp;abruptly bend close to the borders following them in single bows. Theynbsp;are only 2i millimeters distant.

This species differs from the preceding by narrower leaves, the more acute angle of divergence of the veins and their relative positions. Itnbsp;might be compared to Ehamnus tenax, Lesqx., �U. S. Geol. Rep.,� vi,nbsp;p. 109, pi. xxi, fig. 4; but the leaves, though of the same width, arenbsp;nearly twice as long, and the angle of divergence of the veins is only halfnbsp;as broad; the medial nerve also is much thicker.

Hob.�Bluff Creek, Kansas. Chs. Sternberg.

Ficus magnolimfolia, sp. nov.

Plate XVII, Figs. 5, 6.

Leaves very entire, oval or broadly lanceolate, broader below the middle, rounded in narrowing to the short petiole, and declined downward at the slightly decurringnbsp;base; medial nerve of medium size, strict; secondary veins at an acute angle of divergence, close, very oblique, nearly straight from the midrib to near the borders, simplynbsp;camptodrome.

The leaves, 8 to 10 centimeters long, Si to 5i centimeters broad, with a short not inflated petiole about 1 centimeter long, appear somewhat thicknbsp;but not coriaceous; they are acute or tapering to a short acumen (all thenbsp;points are broken). The veins close, 5 to 7 millimeters distant, under annbsp;angle of divergence of 40�, are thin, parallel, except the lowest pair whichnbsp;is a little more oblique. The areolation is obsolete; only a few nervillesnbsp;are seen at the end of the veins, anastomosing in marginal curves alongnbsp;the borders and close to them.

These leaves have a great likeness to those of Magnolia Capellini, Heer, �Phyll. Gr�t. du Neb.,� p. 21, pi. iii, figs. 5, 6, differing especially bynbsp;the more acute angle of the more numerous and closer secondary nerves.nbsp;By this character, and also by the slightly decurring base of the leaves, they

DESCEIPTIOU OP SPECIES.

are related to M. alternans, Heer, l. c., p. 20, pl. iii, figs. 2,4. They may represent one of these species; but on account of the simple curves of the veins close to the borders, and also of their position close to each other,nbsp;they do not appear to be referable to Magnolia. The petiole is not inflatednbsp;as it is often in Ficus, but the lower pair of veins is more oblique, and, asnbsp;seen in fig. 6, the medial nerve is narrowly split or channeled in the middle.

Hob.�With the preceding. H. C. Beckwith.

Ficus Glascoena, sp. nov.

Leaves large, thick, coriaceous, polished on the surface, oblong-lanceolate, obtusely pointed, narrowing and slightly decurring to the petiole; medial nerve very broad;nbsp;secondary veins thin, at a broad angle of divergence, scarcely curved in passing to thenbsp;borders, joining without curving to it a somewhat thick marginal vein.

The leaves are thick, 15 to 20 centimeters long, 6 to 7 broad; the midrib 2 to 3 millimeters broad at base. The type of venation resemblesnbsp;that of Ficus parasitica, Shott., as figured by self-impression in �Bil. FL,�nbsp;pl. xxiii, fig. 1; the thin lateral veins sometimes branching in the middle,nbsp;abruptly anastomosing to a somewhat thick marginal vein which followsnbsp;close to the borders in successive bows. The secondary veins appearnbsp;separated by parallel thinner shorter tertiary veinlets; but the divisionsnbsp;of the third order and the details of areolation are obscure.

Hah.�^Two and a half miles south of Glasco, Kansas. Cha. Sternberg.

Ficus distorta, Lesqz.

Plate XIV, Kg. 4.

Bayden�a �Ann. Eep.,� 1874, p. 343, pl. v, fig. 6.

Leaf coriaceous, entire, obovate, unequilateral, pointed or acuminate, apparently gradually narrowed to the base; nervation pinnate; secondary nerves thick, parallel,nbsp;equidistant, camptodrome; nervilles strong, at right angles to the veins, anastomozingnbsp;and subdividing into an irregularly quadrate or polygonal areolation.

A mere fragment of a leaf of which the upper and lower parts are destroyed. The characters do not positively indicate its relation to Ficus.nbsp;It is figured and described for future comparison.

Hah.�Near Fort Harker, Kansas.

FLOEA OF THE DAKOTA GEOUP.

Ficus laurophylla, Lesqx.

Plate I, Figs. 12, 13.

Hayden�s �Ann. Eep.,� 1874, p. 342, pi. v, fig. 7.

Laurophyllum reticulatum, Lesqx., �U. S. Geol. Kep.,� vi, p. 76, pi. xr, figs. 4, 5.

Leaves coriaceous, polished on the upper face, entire, narrowly lanceolate, acuminate, gradually taperiug to a short thick petiole; medial nerve thick, grooved on the upper side; secondary veins close, very open.

A large number of specimens of this fine species have been examined. Though generally more or less fragmentary and often erased on the surface, the essential characters may be generally recognized. The leavesnbsp;vary in size from 10 to 20 centimeters long and from l� to 4J centimetersnbsp;broad in the middle. They are lanceolate, gradually narrowed both waysnbsp;from the middle. The secondary nerves are parallel, unequal in distance,nbsp;nearly at right angles to the midrib, and also nearly straight in passing tonbsp;near the borders, where they curve and anastomose in festoons. Theynbsp;are generally separated by one or two tertiary veins attached to them bynbsp;branches either oblique or at right angles, whose subdivisions compose annbsp;irregularly quadrate areolation.

By its nervation this species has a typical relation to F. G-lascoena. The curves of the secondary veins, which follow close to the borders innbsp;successive bows, form a kind of margin, as in the preceding species; butnbsp;the veins distinctly curve to the festoons and compose them. They donbsp;not abruptly anastomose with them by their attenuated ends; for thisnbsp;reason the marginal flexures are thin, rarely distinct in this species, whilenbsp;in F. Glascosna they appear as formed by a truly independent nerve, morenbsp;deeply and distinctly marked than the ends of the secondary veins.

In the collection of the Museum of Comp. Zool. of Cambridge I have found fourteen specimens of leaves same size and form as those describednbsp;here, with the same character of areolation, but with the secondary veinsnbsp;at an acute angle of 30�. All the specimens are from the same locality,nbsp;Elkhorn Creek, and seem to represent a truly different species. But thenbsp;lateral veins and their divisions are not distinct enough to be satisfactorilynbsp;described.

At first I considered the relation of these leaves to be with the Laurinem. But as remarked already in the first description of this species,

C F 4

DESCRIPTION OP SPECIES.

the venation is of the same type as that of some species of Ficus of both the present and the older floras, comparable, for example, to that of Ficusnbsp;Geinitzii, Ett., �FL, Niedersch.,� p. 16, pi. ii, figs. 7, 9-11.

The two leaves figured, pi. i, show the under face, where the veins are more distinct and the medial nerve half-round. On the upper facenbsp;the midrib is deeply channeled, but not inflated at the point of union tonbsp;the short petiole which is rarely longer than 11 centimeters.

Hah.�Commonly found throughout the Dakota Group formation from Minnesota to Southern Kansas.

PROTEACE^.

PROTEOIDES, Heer.

Proteoides daplinogenoides, Heer.

�TJ. S. Geol. Eep.,� vi, p. 85, pi. xv, figs. 1, 2.

Proteoides grevilleseformis, Heer.

�U. S. Geol. Eep.,� tI, p. 86, pi. xxviii, fig. 12.

Proteoides lancifoliiis, Heer.

quot; Quedlinb. FI.,� p. 12, pi. ill, figs. 5, 6.

Leaves narrowly lanceolate, narrowed in the upper part, very entire.

Two specimens. Nos. 63 and 76, of the Museum of Comp. Zool. of Cambridge, seem referable to this species. The first is a fragment of anbsp;linear-lanceolate leaf narrowed upward to an inclined apparently obtusenbsp;point, 8 to 9 centimeters long, 11 millimeters broad in the middle, thenbsp;base destroyed. The medial nerve is narrow, and the thin lateral veins,nbsp;two of which are seen near the base, come out at a very acute angle ofnbsp;divergence and are soon effaced upward.

The other leaf is larger, 16 millimeters broad in the middle, 8 centimeters long, lanceolate, gradually equally narrowed both ways, obtuse at the apex. Its medial nerve is flat, somewhat broader, 1 millimeter nearnbsp;the base, wherefrom two lateral nerves ascend at a very acute angle ofnbsp;divergence (about 10�), and no other veins are distinct up to above thenbsp;middle of the leaf, where a few alternate ones come out at a broad anglenbsp;of divergence, curving up as in fig. 6 of Heer, loc. cit. Except that thisnbsp;leaf is slightly broader the characters are identical.

Hah.�Near Fort Harker, Kansas. Chs. Sternberg.

FLORA OF THE DAKOTA GROUP. nbsp;nbsp;nbsp;51

EMBOTHRITES, Ung.

Embotlirites (?) daplineoides, Lesqx.

� U. S. Geol. Rep./� vi, p. 87, pi. xxx, fig. 10.

From the comparison of a number of well-preserved specimens of Andromeda Parlatorii, Heer, recently received from Kansas, I am disposednbsp;to consider this fragment as referable to this last species.

LOMATIA, R. Brown.

Liomatia? Saportanea, Ijesqx.

Plate III, Fig. 8 (enlarged).

Hayden�s quot;Ann. Rep.,� 1874, p. 346.

Todea Saportarua, Lesqx., quot; U. S. Geol. Rep.,� vi, p. 48, pi. xxix, figs. 1-4.

Leaves compound, linear in outline; ultimate divisions membranaceous or sub-coriaceous, narrowly lanceolate, acute, connate by the decurring base forming a narrow nerved wing to the rachis; medial nerve strong and straight, continuous to the apex; secondary veins simple, close, parallel, diverging at an acute angle in passing upnbsp;close to the borders, which they follow in simple bows; tertiary veins shorter, anastomosing with the secondary ones by oblique diversely inclined veinlets.

The ultimate divisions of the leaves are parallel-oblique or somewhat curved downward, alternate or sub-opposite, a disposition similar to thatnbsp;of the divisions of the pinnae of a number of species of ferns. They arenbsp;gradually decurrent on the rachis, following it downward as a narrow^nbsp;veined or smooth margin. The venation of the leaves is distinctly seennbsp;on the enlarged fragment, fig. 10.

My first impression in regard to these remarkable and fine vegetable remains was that they represented an extinct kind of fern. I even supposed that, considering the peculiar disposition of the leaflets and theirnbsp;venation, which is sometimes mixed with curved lines, we had here vegetable remains of a new type, constituting a link of transition between thenbsp;ferns and the plants of a higher order. The segmentation of the leaves isnbsp;similar to that of some species of fossil ferns, Sphenopteris desmomera,^nbsp;for example, which, according to the remarks of the author, has no relationnbsp;to any living fern; also related to the fragments described by Debey andnbsp;Ettingshausen^ under the generic name of Monheimia. For not only havenbsp;they a similar division of the pinnae, but, as seen in fig. 6, the nervation

* Saporta, quot;Plantes fossiles des lits �. poissous de C�rin, p. 22, pi. xiv.

quot; Urweltlicher Acrobryen,� p. 31, pi. iv, figs. 1-10.

DESOEIPTION OF SPECIES.

is somewhat analogous, the numerous parallel secondary veins curving up along the borders, some of them united by oblique veinlets.

Competent observers in Europe have contradicted these views and referred the fossil fragments to the Proteaceoe, comparing them to somenbsp;species of Lomatia; and later I have received from the Oligocene of Florissant a large number of specimens, partly figured (pi. xliii), whose relationnbsp;both with the Cretaceous species and with living specimens of Proteaceoenbsp;is evident.

liomatia Saportanea, var. long�ifolia.

Leaves larger, divisions longer and broader.

None of the lateral leaflets are preserved entire, but from the fragments they are at least 8 or 9 centimeters long, though comparatively narrow, only i to 1 centimeter broad. The upper leaflets, two pairs ofnbsp;which are preserved, with the terminal upon one of the specimens, are 6nbsp;centimeters long and 7 millimeters broad, the terminal having the samenbsp;.size and characters.

Besides the difference in the size of the leaflets, these appear a little more distinctly coriaceous, and their surface is smooth without any tracenbsp;of venation. Better specimens may prove this to be a different species.

Hab.�The specimens from which the variety is described are from Morrison, Colorado, procured by A. Lakes. The others, first described,nbsp;are from Kansas.

LAUEACEJE.

LAURUS, Linn.

Laurus �febrascensis, Lesqx.

�U. S. Geol. Eep.,quot; vi, pi. 74, p. x, fig. 1; pi. xxviii, fig. 14.

Laurus macrocarpa, Lesqx.

�U. S. Geol. Eep.,� vi, p. 74, pi. x, fig. 2.

Laurus protesefolia, Lesqx.

Plate III, Figs. 9, 10; XVI, Fig. 6.

Hayden�s �Ann. Eep.,� 1874, p. 342, pi. v, figs. 1, 2.

Leaves subcoriaceous, broadly lanceolate, gradually narrowed from below the middle into a long blunt acumen, more rapidly attenuated to the base; medial nervenbsp;straight or slightly curved; lateral nerves slender, camptodrome, parallel, except thenbsp;lower pair slightly more oblique.

The leaves vary in size from 9 to 16 centimeters long and 2i to 31

FLORA OF THE DAKOTA GROUP.

centimeters broad at or below the middle. The secondary veins, distinctly curved in passing from the midrib to the borders, are more or less distant,nbsp;rarely separated by shorter tertiary veins cut at right angles by strongnbsp;nervilles, which are simple or anastomosing in the middle, the upper onesnbsp;ascending to the borders. The areolation is not seen, the surface appearing punctulate or closely dotted by small areoles.

In my first description of this species, 1. g.,1 compared it to Proteoides dapJmogenoides, from the shape of the leaves only. This affinity is distant.nbsp;By the form of the leaves this species rather resembles Ficus Krausiana,nbsp;Heer, and F. BeckwitUi, described above. Its venation is that of Tjaurusnbsp;Nehrascensis, from which it differs by the narrower medial nerve, thenbsp;secondary veins more slender and more curved in passing to the borders,nbsp;the prolonged point of the leaves, etc.

Hal).�Near Fort Harker, Kansas. Clis. Sternberg. Recently found at Morrison, Colorado, by A. Lakes.

Lavirus? modesta, sp. nov.

Plate XVI, Pig. 4.

Leaves small, linear-oblong, cuneate to the petiole; midrib thick; secondary veins irregular in distance, camptodrome, following close to the borders in prolonged curves.

There is only a fragmentary specimen of a small, apparently linear-lanceolate leaf (point broken), whose relation is not positively ascertained. The nervation is like that of Laurusprimigenia, Ung., in Sap. ��t.,� 2,1,nbsp;p. 89, pi. iii, fig. 8, the lateral veins at about the same distance and obliquenbsp;in the same degree, curving high and close to the borders; but no trace ofnbsp;areolation is distinct. This fragment is also related to Myrtophyllumnbsp;pusillum, Heer, �Quedl. FI.,� p. 14, pi. iii, fig. 10, represented by a stillnbsp;smaller fragment of leaf, round at base, with secondary veins curved andnbsp;following high along the borders.

Had).�Near Morrison, Colorado. H. C. Beckwith.

PERSEA, Gsertn.

Persea Leconteana, Lesqx.

� U. S. Geol. Rep.,� vi, p. 75, pi. xxviii, fig. 1.

Persea Sternbergii, Lesqx.

�U. S. Geol. Rep.,� vi, p. 76, pi. vii, fig. 1.

DESOEIPTION OF SPECIES.

CmNAMOMUU, Barm.

Ciunamomum Scheuchzeri? Heer.

�U. S. Geol. Rep./� vi, p. 83, pi. xxx, figs. 2, 3.

Professor Heer considers the reference of these leaves to C. Scheuchzeri as uncertain; for though the form of the leaves is much the same, thenbsp;middle nerve is too thick for that species, especially toward the point.nbsp;Saporta is also of opinion that the presence of C. Scheuchzeri in the Cretaceous is very improbable, as in Europe this species is essentially in thenbsp;upper Miocene. In his paper (�Descriptions of the fossil plants collectednbsp;by Mr. George Gibbs�), Professor Newberry doubtfully refers to Oinnamo-mum Heerii, Lesqx., some leaves whose affinity of nervation is in his opinionnbsp;with C. Scheuchzeri or C. lanceolatum. Following Professor Heer�s opinion,nbsp;I had changed the original name of the �Rep.,� 1. c., to that less definitenbsp;of Daphnogene cretacea (Hayden�s �Ann. Eep.,� 1874, p. 343); but ifnbsp;specific identification is not ascertainable from the fragmentary specimensnbsp;obtained thus far, the close relation is at least indicated by the old name,nbsp;which should, therefore, be preserved. Another reason against the changenbsp;of name is the intimate relation, or perhaps identity, of the Cretaceousnbsp;C. Heerii, with a Tertiary species of the genus.

Ciunamomum Heerii, Lesqx.

�U. S. Geol. Rep.,� vi, p. 84, pi. xxviii, fig. 11.

Leaves thick, coriaceous, very entire, ovate, taper-pointed, rounded at the base to a short petiole; lateral nerves emerging a little above the base, ascending higher thannbsp;the middle of the leaves, branching outside.

There is scarcely any modification to be made to the description of the �Rep.,� 1. c., which I am able now to complete from a recently procured specimen of an entirely preserved leaf. This leaf, 9 centimeters longnbsp;without the petiole (1 centimeter long), is broadest above the base, roundednbsp;to the petiole, joining it in an abruptly and short declining curve, andnbsp;tapering above to a somewhat acute or merely blunt point. The medialnbsp;nerve is broad and deep, enlarged to the base from the point of union ofnbsp;the lateral primary nerves 7 millimeters above the top of the petiole, gradually narrowed upward but distinct or persistent to the apex. The lateralnbsp;nerves though thick are not as strong as the midrib, ascend in slightlynbsp;curving inward up to nearly the second pair of secondary veins, where

FLORA OF THE DAKOTA GROUP.

they are effaced near the borders. The secondary veins,^ two pairs, are alternate, distant, much curved in ascending high toward the borders, thenbsp;lowest joining the medial nerve above the middle of the leaves, while fromnbsp;the base downward to the fork of the primary nerves the area is filled bynbsp;a series of thin nervilles derived at right angles from the midrib. Thenbsp;lateral primary nerves are divided in numerous lateral branches, 5, 6nbsp;curving in passing outside toward the margins, where they become effaced.

This leaf is well enough represented by the figure in �U. S; Geol. Rep.,� vi, made from a specimen whose borders had been ground from thenbsp;middle downward and rounded to the point of union of the lateral nervesnbsp;in such a way that the relative position of the nerves to the base of thenbsp;leaf could not be ascertained, nor the disposition of the borders in joiningnbsp;the petiole. The size of the newly-found leaf is larger and its broadestnbsp;point is close toward the base.

Excepting this last character, and its thinner venation, the Cretaceous leaves are very similar to those described from the Mississippi Eocene asnbsp;C. Mississippiense, lately identified with numerous leaves of C. affine, of thenbsp;Laramie and Carbon Groups. These are of about the same size, butnbsp;all are rather oval-acuminate than ovate, the broadest part being in thenbsp;niiddle. In C. polymorplmm, to which both the Cretaceous and Tertiarynbsp;species have been compared, the leaves are broader above the middle.

The specimen figured in �U. S. Geol. Rep.,� vi, 1. c., came from near Ellsworth, Kansas. That of Nanaimo was, as far as I can recollect, in anbsp;still more imperfect state of preservation, and as I have not preserved anbsp;copy of the plates delivered to Dr. Evans, which have never been published,nbsp;I am unable to see, if, indeed, the leaf of Nanaimo is identical with that ofnbsp;the Dakota Group. This, however, could not force a definite conclusionnbsp;of the age of the flora of Nanaimo, as the Cretaceous type of Cinnamomumnbsp;appears preserved with very little modification in the different Tertiarynbsp;stages of this continent.

OREODAPHNE, Nees.

Oreodaphne cretacea, Lesqx.

�U. S. Geol. Rep./� vi, p. 84, pi. xxx, fig. 5.

A fine leaf of this species recently found in Kansas (No. 215, Coll, of the Museum Comp. Zool., Cambridge) has all the characters of the leaf

DESCEIPTIOK OF SPECIES.

figured. It differs merely by the secondary nerves not being as thick. The areolation is not distinct.

SASSAFEAS, Nees.

Sassafras Mudgei, Lesqx.

quot;U. S. Geol. Rep./� vi, p. 78, pi. xiv, figs. 3, 4; xxx, fig. 7.

Sassafras acutilobnm, Lesqx.

Plate V, Figs. 1, 5.

� U. S. Geol. Rep.,� vi, p. 79, pi. xiv, figs. 1, 2.

The form appears specific, as it is represented by leaves of very different size and always with the same characters. All the lobes are very entire, the lateral either broadly diverging, sometimes nearly at right anglesnbsp;to the midrib or erect; the venation is distinct but not coarse. The leaf,nbsp;fig. 5, is one of the smallest seen of this species. The largest measuresnbsp;12 to 14 centimeters long without the petiole, or still more, for I have seennbsp;from Kansas a fragment, only the middle lobe, 10 centimeters long fromnbsp;the sinuses to the a;pex and 4 centimeters broad. As the lateral lobesnbsp;greatly vary in their divergence, of course the width of the leaves differnbsp;much. The species is especially abundant at Thomson Creek, near Fortnbsp;Harker, with S. cretacewm and other forms of the same type.

SASSAFEAS (Araliopsis), Lesqx.

Sassafras (Araliopsis; cretaceum, Newby.

� Later Ext. FI.,� p. 14; � Illustr,� pi. vi, figs. 1, 4 (fragments of leaves). Lesqx., � U. S. Geol. Rep.,� vi, p. 80, pi. xi, figs. 1, 2; xii, fig. 2.

Sassafras (Araliopsis) obtiisum, Lesqx.

S. cretaceum, var. oltusum, Lesqx., � U. S. Geol. Rep ,� vi, p. 80, pl xii, fig. 3; xiii, fig. 1.

This form should be considered as specific, not merely on account of its shorter, more obtuse lobes, but particularly of the venation, which isnbsp;much coarser than in the preceding species. The primary nerves, especially, are much broader and sharply cut. It is found with S. cretaceum atnbsp;Thomson Creek; but it is also found by itself alone in other localities.

Sassafras (Araliopsis) mirablle, Lesqx.

� U. S. Geol. Rep.,� vi, p. 80, pl. xii, fig. 1.

Platanus latiloba, Newby., � Later Ext. FI.,� p. 23; � Illustr.,� pl. ii, fig. 4.

To the characters indicated in �Rep.,� vi, may be added the thick coriaceous substance of the leaves, which in small specimens appear

FLORA OP THE DAKOTA GROUP.

horny; the great divergence of the lateral lobes nearly at right angles to the medial nerve and also generally curved down; the middle lobe isnbsp;always comparatively short and broad.

Sassafras (Araliopsis) dissectum, sp. nov.

Leaves very large, long and narrowly cuneate to the petiole, palmately flve-lobate by subdivision of the lateral lobes diverging at an acute angle from the medial one.

The leaves of this form are very large, some measuring 22 centimeters from the top of the petiole to the apex, 20 centimeters between the extremities of the lateral lobes. The base is narrowly cuneate, long, decurringnbsp;to the petiole; the three primary divisions are joined in obtuse but narrownbsp;sinuses; the lateral ones at an acute angle of divergence are cut into twonbsp;short obtuse dentate lobes, while the middle one is taper-pointed, notnbsp;lobed, but deeply undulate-dentate. This form might be considered as anbsp;var. of S. mirahile, but if differs greatly in the general facies, the lateralnbsp;lobes oblique erect lobed and unequilateral, the lateral primary nervesnbsp;alternating at base or joined to the medial at a distance from each other,nbsp;the long lanceolate undulate-dentate middle lobe and in the nervation,nbsp;the primary nerves being thick indeed, while the secondary nerves andnbsp;their branches are thin, generally effaced along the borders.

Hah.�This form has not been seen among the numerous specimens of fossil plants examined from the Dakota Group until recently. It isnbsp;represented in the collection of the Museum of Comp. Zool. of Cambridgenbsp;by a number of fine specimens, all obtained from 3 and 7 miles north ofnbsp;Fort Harker by Chs. Sternberg.

Sassafras (Araliopsis) recurvatum, Lesqx.

Plataniis recurvata, Lesqx., � U. S. Geol. Kep.,� vi, p. 71, pi. x, figs. 3-5.

Leaves three to five palmately lobed; lobes nearly equal in length, the medial broader; lateral nerves curving downward, either simple with mere secondary veins ornbsp;forking above the base; lobes undulate or obtusely dentate on the borders.

This form is evidently transient in its characters. By the cuneate and decurrent base of the leaves joining the petiole at a distance belownbsp;the point of union of the three primary veins and by the trilobate division,nbsp;it is a Sassafras. But by the irregularity of the lobes or the subdivisionsnbsp;of the leaves in lobes and teeth, it seems referable to Platanus, while anbsp;tendency to become five-lobate by the forking of the lateral nerves is a

DESCEIPTION OF SPECIES.

character of the Araliacece. This last character is still more marked in the following species.

This form is very rare. Except the specimens figured in the �Rep.,� 1. c., I have not seen any identifiable with it, except a well-preserved leaf.nbsp;No. 148, counterpart 105, of the Museum Comp. Zool., Cambridge, whichnbsp;in all its characters, especially by its peculiar nervation, represents in anbsp;diminutive form fig. 3 of pi. x. The lateral nerves join the-medial only anbsp;little above the base of the leaf, and the lower pair of secondary nervesnbsp;follow upward along the borders and by an inward curve anastomose withnbsp;the outside curved end of the second pair above the middle of the leaf.

Sassafras (Arallopsis) platauoides, sp. nov.

Plate VII, Pig. 1.

Leaves narrowly cuneate from the middle downward, palmately flve-lobate in the upper enlarged part; lobes short, the upper half-round or obtuse, apiculate, the lowernbsp;deltoid-acute; primary nerves tripartite from far above the base of the leaves; lateralnbsp;nerves branching in the middle, primary and secondary divisions passing out to thenbsp;points of the lobes.

The leaf figured is 13 centimeters long from the point where it joins the enlarged medial nerve in gradually decurring to it, and 11 centimetersnbsp;broad between the lower lateral lobes, which, though shorter than thenbsp;upper ones, are turned outside, while those above are directed upward;nbsp;the point of union of the veins is 2� centimeters above the base of thenbsp;leaf, the medial nerve underneath being 3 millimeters thick or three timesnbsp;as broad as the medial nerve above the division. The lobes are of anbsp;peculiar shape, the lower ones deltoid-acute, short, about 1 millimeternbsp;long; the upper ones longer, rounded and narrowed to a blunt apex; thenbsp;terminal is of the same shape but still longer; all are joined in obtusenbsp;sinuses.

The close relation of this leaf to Platanus Heerii, �U. S. Geol. Rep.,� vi, pi. ix, figs. 1, 2, will be easily recognized; but still, the long narrowlynbsp;wedge-form base, the subdivision of the lateral primary nerves, are characters represented in Araliofsis, especially in the preceding species, so thatnbsp;it is extremely difficult to say with which of these generic divisions thisnbsp;kind should be identified.

Hah.�Near Clay Center, Kansas. H. C. Towner, from a figure com-

FLORA OF THE DAKOTA GROUP.

municated. Bnt other leaves of the same characters, only a little smaller (Nos. 694, 672, Museum Comp. Zool., Cambridge), have been found bynbsp;Chs. Sternberg, on Thomson Creek, 7 miles south of Fort Harker.

Sassafras (Arallopsis) subintegrlfolium, Lesqx.

� U. S. Geol. Rep./� vi, p. 82, pi. iii, fig. 5.

From a number of specimens more or less similar to those of the leaf figured �U. S. Geol. Rep.,� vi, L c., I believe it represents only a deformation of S. cretaceum, especially of its variety obtusum. I have, however,nbsp;received quite recently, from North Kansas, leaves of Sassafras perfectlynbsp;entire or lobate on one side only, identical in shape and size with thenbsp;leaves of Sassafras officinale commonly found also entire, bilobate or trilobate. They were sent by Mr. L. C. Mason, of Delphos.

ARISTOLOCHIACE^.

ARISTOLOCHIA, Tourn.

Aristoloclila deutata, Heer.

�Phyll. Cr�t. du Neb./� p. 18, pi. ii, figs. 1, 2; Lesqx., � U. S. Geol. Rep.,� vi, p. 87, pi. xxx, fig. 6.

DIOSPyRIJSTE^.

SAPOTACITES, Ett.

Sapotacites Haydenii, Newby.

� Later Ext. FI.,� Catal., p. 8; � Illustr.,� pi. v, fig. 1.

No description is given of this species. The leaf, of medium size, is obovate, slightly emarginate at the obtuse apex; secondary nerves at annbsp;acute angle of divergence, close, curved in passing up toward the borders,nbsp;divided by short oblique veins detached from both sides of the lateral nerves.

Hob.�Nebraska. Dr. F. V. Hayden.

DIOSPYROS, Linn.

Diospyros primseva. Heer.

quot;Phyll. Gi�t. du Neb.,� p. 19, pi. i, figs. 6, 7; Neivby., �Later Ext. FI.,� Catal., p. 8; �Illustr.,� pi. iii, fig. 8.

Leaves oblong-oval, very entire, rather obtuse at the apex; secondary veins flex-uous, branching, camptodrome.

The author compares it to his D. anceps of the European Miocene, and to H. Alaskana of the same formation of Alaska. The species is notnbsp;rare in Kansas.

DESCEIPTION OF SPECIES.

Diospyros amblguap Lesqx.

2). ancepi, Leeqx., �U. S. Geol. Rep.,� vi, p. 89, pl. vi, fig. 6.

Diospyros rotundifolla, Lesqx.

�U. S. Geol. Rep.,� vi, p. 89, pl. xxx, fig. 1.

ERICACEJE.

ANDROMEDA, Linn.

Andromeda Parlatorii, Heer.

�Phyll. Cr�tl du Neb.,� p. 18, pl. i, fig. 5; Lesqx., � �. S. Geol. Rep ,� vi, p. 88, pl. xxiii, figs. 6,7; xxviii, fig. 15.

Andromeda affinis, Lesqx.

Plate II, Fig. 5.

Hayden�s �Ann. Rep.,� 1874, p. 348, pl. iii, fig. 5.

Leaf thick, narrowly lanceolate, acuminate, entire j medial nerve comparatively thick; lateral veins close, parallel, at an acute angle of divergence, camptodrome.

The leaf, 5i centimeters long, 11 millimeters broad in the middle, is gradually narrowed downward to the petiole and upward to a somewhatnbsp;long acumen; the angle of the lateral nerves is 30�; the areolation isnbsp;composed of round or quadrate polygonal minute areoles.

This species is closely allied to the preceding; the veins are less oblique and more curved.

Hab.�Spring Canon, with fragmentary leaves of A. Parlatorii. Dr. F. V. Hayden.nbsp;nbsp;nbsp;nbsp;,

AEALIACE^.

ARALIA, Linn.

Aralia formosa. Heer.

Plate XI, Figs. 3, 4.

Heer, �Moletein FI.,� p. 18, pi. viii, fig. 3.

Leaves petioled, triple-nerved, trilobate; lobes dentate, blunt at the apex.

This species, as represented by American specimens, though positively identified, presents a few unimportant points of difference. In Heer�snbsp;figures the base of the leaves is wedge-form and the divisions oblique;nbsp;in those which I have for examination the middle lobe is oval or lance-

* The name of this species is changed as preoccupied by Heer.

FLOEA OF THE DAKOTA GEOUP.

olate, the lateral linear lanceolate, not enlarged in the middle, as far as seen from the one partly preserved, and the borders are obtusely serratenbsp;from near the base. In Heer�s figures the medial lobe is shorter and narrower, and it is, like the other, denticulate only in the upper part. Thenbsp;secondary veins are not very distinct; a few, of which the base only isnbsp;seen, are parallel, close, at an open angle of divergence. The leaves arenbsp;thick; the petiole is not preserved, but as seen in Heer�s specimen it isnbsp;short and thick.

Heer compares this species for the shape of the lobes to A. Japonica, which, however, has the leaves five-lobed, and indicates its relation to A.nbsp;primigenia of Mount Bolca and of Alumbay.

Hab.�Near Morrison, Colorado. H. C. Beckwith.

Aralia Saportanea, Lesqx.

Plate VIII, Figs. 1, 2; IX, Figs. 1, 2.

Hayden's �Ann. Rep,,� 1874, p. 350, pi. 1, fig. 2.

Leaves large, sub-coriaceous, triple-nerved and ffve-lobate by division of the lateral nerves, fan-shaped in outline, narrowed in a curve or broadly cuneate, andnbsp;decurring to a long slender petiole; lobes narrowly lanceolate or linear-lanceolate,nbsp;acute or blunt at the apex, equally diverging, distantly dentate from below the middlenbsp;upward; secondary nerves sub-camptodrome.

This beautiful species is known by numerous finely preserved specimens. The leaves, 9 to 20 centimeters long from the top of the petiole to the summit of the middle lobe, are of the same width between the pointsnbsp;of the lower lateral lobes; the petiole is long and comparatively slender,nbsp;though appearing thick upon one of the specimens, probably enlarged andnbsp;flattened by compression. The preserved broken part on one of the leavesnbsp;measures 5 centimeters. The lobes cut down to about two-thirds of thenbsp;leaves are narrowly lanceolate, slightly narrower near the obtuse sinuses,nbsp;equally diverging, the lower lateral ones much shorter, curved down, andnbsp;decurring to the base of the leaves. The leaves, triple-nerved from thenbsp;division of the primary nerves a little above the base, become five-nervednbsp;from the forking of the lateral nerves at a short distance from their base.nbsp;The secondary veins emerge at an acute angle of 30�, curve in ascendingnbsp;to the borders, and sometimes enter the teeth by their ends; the uppernbsp;more generally follows close to the borders in festoons, emitting under the

DESCEIPTIOK OF SPECIES.

teeth short branches which enter them. There are not any intermediate tertiary veins, but the nervilles are strong, often continuous, anastomosingnbsp;in the middle of the areas and forming by subdivisions a small quadrangular areolation (pi. viii, fig. 1). The typical relation of these Aralianbsp;leaves is marked with Sassafras f AraliopsisJ cretaceum and S. mirabile,nbsp;though the generic and specific characters are far different.

Hah.�South of Fort Barker. Chs. Sternberg. A number of splendid specimens have been found all at the same locality near Brookville, Kansas.

Aralia quinquepartita, Lesqx.

�U. S. Geol. Eep.,� vi, p. 90, pi. xv, fig. 6.

Of this species, described, 1. c., from two fragmentary specimens, I have now seen some better leaves. One, the largest, is 16 centimetersnbsp;long from the top of the petiole to that of a lateral lobe preserved entire.nbsp;It is deeply divided into six narrow oblanceolate lobes, obscurely dentatenbsp;toward the apex, the lower lateral nearly entire. The medial lobe, 2 centimeters broad above the middle, is only 1 centimeter broad near the sinus.nbsp;Though somewhat thick, the leaves are rather membranaceous than coriaceous, the upper face smooth. The lateral veins are obsolete, appearingnbsp;very thinly distributed, about like those of A. Sa^ortanea. The divisionnbsp;of this leaf in six is abnormal; the primary lateral nerves on one sidenbsp;fork twice and therefore form three lobes, while on the other side the lateral nerves fork once only and have thus two divisions only.

Hab.�The best specimens seen of this form are from south of Fort Barker. Chs. Sternberg.

Aralia Towneri, Lesqx.

Plate VI, Fig. 4.

Hayden�s �Ann. Rep.,� 1874, p. 349, pi. iv, fig. 1.

Leaf large, coriaceous, polished on the upper face, irregularly flve-lobed to below the middle; lobes entire, oblong, obtusely pointed; primary nerves in three, from nearnbsp;the top of the petiole, the lateral ones forked at a distance from the base; secondarynbsp;veins open, variable in distance, very curved in passing toward the borders, campto-drome, separated by short tertiary veins parallel to them or at right angles to the midrib..

The leaves of this fine species are, as seen from another better preserved specimen, 15 centimeters long from the top of the petiole and 22 to 24 centimeters broad between the points of the lobes, which, descending

FLOEA OP THE DAKOTA GEOUP.

much lower.than the middle, are 7 to 10 centimeters long and 3 to 3i centimeters broad. The primary nerves are comparatively narrow; the form of the lobes is oblong, the point somewhat obtuse, the sinuses broad and alsonbsp;obtuse. The secondary nerves distant, nearly simple, at an open angle ofnbsp;divergence, pass toward the borders in curves and follow them in festoons,nbsp;anastomosing by nervilles with those above. They are generally separated by short tertiary veins forming by ramifications in more or lessnbsp;oblique directions, square or polygonal, large meshes.

Hub.�Clay Centre, Kansas. H. C. Toioner.

Aralia subemarginata, sp. nov.

Leaves of medium size, thick, coriaceous, five-palmate, cuneate to the base; lobes cut to the middle of the leaves, entire, obovate, rounded or emarginate at the apex;nbsp;primarj^ nerves in three, the lateral forking near the base; venation camptodrome.

The lobes of this leaf are nearly equal in length, about 5 centimeters long from the narrow obtuse sinuses, 5 to 6 centimeters broad in thenbsp;upper part; lateral veins few, distant, 3 or 4 pairs, some of them forkingnbsp;on the lower side, much curved in passing to the borders. This speciesnbsp;is closely allied to the preceding, differing by the short, obovate, roundednbsp;or emarginate lobes and the nervation. The only specimen seen is No.nbsp;810 of the Museum Comp. Zool., Cambridge.

Ifab.�Three miles southeast of Fort Marker, Kansas. Obs. Sternberg.

Aralia teniiinervis, sp. nov. nbsp;nbsp;nbsp;,

Plate VII, Fig. 4.

Leaf small, truncate at base, palmately five-lobed; lobes much diverging, lanceolate or linear-lanceolate, acute; sinuses broad and obtuse; primary nerves thin, flex-uous, apparently diverging from the same point near the base of the leaf; lateral veins close, parallel, camptodrome.

The base of the leaf is destroyed and the point of union of the lateral nerves is not seen. It appears to be about like that of fig. 3 of the samenbsp;plate, a leaf related by its shape. The thin primary nerves, the closenbsp;lateral thin veins, separate this species from all the others described above.nbsp;Its type is that of Aralia angustiloba, Lesqx., of the Chalk Bluffs of thenbsp;Gold-gravel formation of California.

Hob.�Clay County, Kansas. H. C. Toioner.

DESOEIPTION OF SPECIES.

Aralia radlata, sp. nov.

Plate VU, Figs. 2, 3.

Leaves small, palmately live-lobed; base truncate and abruptly declined to the petiole; lobes equally diverging, lanceolate-acuminate, the lower at right angles to thenbsp;medial nerve; primary nerves in three or five united near the basilar border of the leaves.

This description and the figures of this species are made from sketches communicated by Mr. H. G. Towner, the discoverer. As I have seen anbsp;poorly preserved specimen only, apparently representing the species, I amnbsp;unable to give more details on the characters. In fig. 2 the lateral nervesnbsp;are branching a little above the base. This division is observed in mostnbsp;of the Cretaceous leaves I have described of this genus, and it is especiallynbsp;from this kind of nervation that I have considered them as referable tonbsp;Aralia. But in fig. 3 the primary veins are in five from the base, and thisnbsp;is a character of Sterculia. The great similarity of the leaves cut to two-thirds of their length into lanceolate, gradually cuneate lobes, the habitatnbsp;at the same locality, seem to prove that they represent the same species.

Hob.�Clay Centre, Kansas. H. C. Towner.

Aralia concreta, Lesqx.

Plate IX, Figs. 3, 4, 5.

Hayden�s �Ann. Rep.,� 1874, p. 349, pi. iv, figs. 2, 3, 4.

Leaves small, very thick, coriaceous, palmately five-lobed to below the middle, broadly cuneate and curving to the petiole; lobes linear or narrowly lanceolate, verynbsp;entire; primary nerves three, from a little above the border base of the leaves, thenbsp;lateral forking, all thick, flat, and deep by impression, preserving nearly the same sizenbsp;to the top of the obtusely-pointed leaves.

The leaves vary in diameter from 5i to 8 centimeters between the points of the lateral lobes, being shorter than broad. The secondarynbsp;nervation and areolation are totally obsolete. Fig. 4 is a remarkable form.nbsp;On account of the rounded base of the leaf the lobes are not as widelynbsp;div�rging and the sinuses narrower. The essential characters, great thickness of leaves, broad percurrent primary nerves, the size also being thenbsp;same, the difference cannot be considered as specific.

Hal).�Clay Centre. E. C. Towner. Bluff Greek, Ellsworth County, Kansas. Chs. Sternberg.

ELOKA OE THE DAKOTA GEOUP. nbsp;nbsp;nbsp;6�

HEDERA, Linn.

He de ra ovalis, Lesqx.

� U. 3. Oeol. Rep,,� vi, p. 91, pl. xxv, fig. 3; pl. xxvi, fig. 4.

Hedera Scliimperi, Lesqx.

Plate lY, Fig. 7.

ir.ayilen�s �Ann. Rep.,� 1874, p. .351, pl. vil, fig. 5.

Leaf siib-reniform, broader than long, rounded at the top, abruptly narrowed or obliquely sub truncate to the petiole, three nerved from a little above the base; lateralnbsp;nerves curving and more or less oblique toward the borders, anastomosing by thicknbsp;branches and veinlets with the divisions of short distant secondary veins curvingnbsp;along the borders and entering by short veinlets the distant slightly marked denticula-tions of the margins.

The leaf is coriaceous, 6i centimeters broad and 6 centimeters long without the petiole, which is only 7 millimeters long. As seen on thenbsp;specimen it appears enlarged to a point of attachment, not very distinct,nbsp;however. The lateral veins are inclined on one side toward the medialnbsp;nerves; on the other they rather tend down or toward the borders; thenbsp;veinlets all nearly at right angles, anastomosing with the divisions ofnbsp;the secondary veins, form an irregular areolation of angular, square, ornbsp;polygonal meshes. The areolation is of the same character as in thenbsp;preceding species, and is analogous to that of Greviopsis tremuloefolia andnbsp;of Cissus ampelopsidea, Sap., and recognizable also in the following species.

Hah.�South of Fort Barker. Clis. Sternberg.

Hedera platanoidea, Lesqx.

Plate nr. Figs. 5, 6.

Hayden�s �Ann. Rep.,� 1874, p. 351, pl. ill, fig. 3.

Leaf small, broadly ovate, rounded at the top, truncate at base, short petioled, entire, triple-nerved at a short distance above the basal borders of the leaves; primarynbsp;nerves craspedodrome.

The leaves, five to six centimeters in diameter, are about as broad as long; the borders are entire, though somewhat forced outside over thenbsp;points of the primary nerves and thus very obscurely and obtusely trilobate. The lowest branches of the primary lateral nerves follow the borders in festoons along the base of the leaves as in the preceding species,nbsp;and there is also under the primary nerves a pair of marginal veinlets at

DESCRIPTION OF SPECIES.

right angles to the midrih. The secondary veins and their divisions all reach to very near the borders, even seem to reach them, anastomosing atnbsp;their ends with a veinlet which follows close to the margins in successivenbsp;short curves like a marginal vein. The nervilles are strong, more or lessnbsp;at right angles to the nerves, not continuous, anastomosing in the middlenbsp;of the areas, composing a net of large irregular quadrangular or polygonalnbsp;meshes. The surface of these leaves is rough, the venation deep and distinct, the substance thick, nearly coriaceous; the short petiole (7 millimeters long) is enlarged at the base.

Hob.�Near Fort Barker. Chs. Sternberg.

AMPELIDEiE.

CISSITES, Heer.

Leaves more or less deeply trilobate by the esteusion of the lateral primary nerves always in three, rounded and broadly cnneate to the base; lobes deltoid ornbsp;round, entire or dentate, sometimes lobed; secondary nerves mostly camptodrome.

Under the name of Oissites insignis, and without definition of the genus. Professor Beer has described a fragment of leaf which has apparently anbsp;degree of affinity to those which I place under this generic division. Thenbsp;leaves are closely allied to Araliopsis by the primary nervation alwaysnbsp;being trifid generally from a distance above the basal borders, and by thenbsp;areolation and the more or less distinctly trilobate division. The secondary veins are generally camptodrome.

Cissltes insigni.s. Heer.

�Phyll. Cr�t. du Neb.,� p. 19, pi. ii, figs. 3 (4 restored).

Leaves coriaceous, palmitely deeply trilobate; lateral lobes very unequal, lobes crenate at the apex.

This leaf is very coriaceous, triple-nerved, deeply palmately trilobate. The lower part of the lower lobe is larger than the upper, which is entirenbsp;and bears three obtuse teeth toward the base; the secondary veins arenbsp;thin, anastomosing in curves at a distance from the borders.

Cissites salisburiaefolius, sp. nov.

Sassafras ohtusum, Lesqx., �U. S. Geol. Rep.,� vi,p. 81, pi xiii, figs. 2, 4.

Populites salisburiafoUus, Lesqx., �Am. Jour, of Sci..und Arts,� xlvi, 1868,p. 94.

These leaves, first described as Populites, then as Sassafras or Arali-

FLORA OF THE DAKOTA GROUP.

opsis, and now as Oissites, have indeed some characters which relate them to Araliopsis. They are palmately trilobate, have about the same form asnbsp;Araliopsis cretaceus var. obtusus, and an analogous distribution of thenbsp;nerves and secondary veins. They differ much by the thin texture of thenbsp;leaves and the disposition of the lobes to become more or less obtuselynbsp;and distinctly dentate at the apex, as seen by figs. 2 and 4. The rapidlynbsp;narrowed base and the very long petiole give to them a peculiar fan-likenbsp;shape. Their relation to this group seems indicated by their affinity tonbsp;Oissites insignis.

Cissites Harkerianus, Lesqx.

Plate III, Figs. 3, 4.

Hayden�s �Ann. Rep.,� 1874, p. 352, pi. vii, figs. 1, 2.

Scbssafras (^Araliopsis) HarJeerianum, Lesqx., �U. S. Geol. Rep.,� vi, p. 81, pi. xi, fig. 4.

Leaves coriaceous, broadly rhomboidal in outline, and cuneate to the petiole, palmately sub-trilobed; lateral primary veins joined at a short distance above the base;nbsp;secondary veins and their divisions camptodrome.

The leaves figured here are smaller than fig. 4, pi. xi, of the �U. S. Geol. Rep.,� vi; but this is the only difference, and a number of specimensnbsp;have been found of leaves of intermediate size. The nervation is, ofnbsp;course, more or less pronounced, according to the face exposed upon thenbsp;stone. The relation of this and the preceding species to Araliopsis isnbsp;easily remarked.

Cissites affinis, Lesqx.

Platanus affinis, Lesqx., �U. S. Geol. Rep.,� vi, p. 71, pi. iv, fig. 4; xi, fig. 3.

Leaves coriaceous or sub-coriaceous, triple-nerved from near the base, sub-trilobate, rounded in narrowing to the petiole, broadly deltoid to the apex; borders marked bynbsp;short distant teeth at the points of the excurreut nerves and their branches.

Cissites aciiminatus, Lesqx.

Plate V, Figs. 3, 4.

Hayden�s �Ann. Rep.,� 1874, p. 353, pi. viii, fig. 1.

Leaves deltoid from the middle to the acute point, rounded from the middle downward to the petiole, triple-nerved from the base.

These leaves, 7 to 8 centimeters long and nearly as broad, much resemble those of the preceding species; they differ merely by the bordersnbsp;being entire, the secondary nerves more numerous and camptodrome. Innbsp;fig. 4 the points of the lower pair of these lateral nerves reach to the borders

DESCRIPTION OP SPECIES.

and force them outside, forming short teeth. The difference between this and the preceding form becomes, therefore, less marked and may not benbsp;considered of specific value. But the same remarks can be made on thenbsp;numerous transitional forms of this peculiar flora, as it has been remarkednbsp;already.

Hah.�Near Fort Barker. Chs. Sternberg.

Cissites, Heerli, Lesqx.

Plate V, Fig. 2.

Hayden�s �Ann. Rep1874, p. 353, pi. v, fig. 3.

Leaf fan-shaped in outline, broadly cuneate to the base from above the middle, divided at the upper border into live nearly equal acute lobes separated by broadnbsp;sinuses; primary nerves trifld from above the basal border of the leaf, ascending withnbsp;the lower pair of secondary nerves to the points of the teeth; upper lateral veius andnbsp;all the subdivisions camjitodrome.

Though the base of the leaf is destroyed its outline is clearly defined by tbe preserved part of the borders and the direction of the lateral primary veins. Except that the two lower secondary nerves ascend to thenbsp;points of two lobes, the nervation of the leaf is of the same type as thatnbsp;of the two preceding species. Though the close relation of these leavesnbsp;is evident, this one cannot be compared to Araliopsis. It, therefore,nbsp;authorizes a separation of this group, which by its characters is related tonbsp;the Amjpelidece, especially to Oissus.

Hah.�Near Fort Barker, Kansas. Chs. Sternberg.

Ampelopliyllum, Lesqx.

Hayden�s �Ann. Rep.,� 1874, p. 354.

Leaves ovate or obovate, obtuse, entire, narrowed to a long petiole, or sub-cordate, palmately three-nerved from above the base; nerves flexuous, branching on both sides,nbsp;ascending to the borders.

Anipelophyllum attenuatiim, Lesqx.

Plate III, Fig. 2.

Hayden�s �Ann. Rep.,� 1874, p. 354, pi. ii, fig. 3.

Leaf sub-coriaceous, cuueiform in outline, enlarged and rounded at the top; borders entire, wavy; lateral primary nerves joining the middle at a distance from thenbsp;base, flexuous, branching out and inside, ascending to the borders.

The leaf, 6� centimeters long without the petiole and about the same width between the points of the primary lateral nerves, is rounded at the

FLORA OF THE DAKOTA GROUP.

top and undulate by the out-running of the veins. It is triple-nerved from a distance above the base, and has above the point of connectionnbsp;of the primary nerves two or three pairs of alternate secondary veins,'nbsp;variable in distance, straight or curved, unequally parallel, reaching thenbsp;borders either directly or by their bi�anches, which by oblique branchletsnbsp;or by connections of nervilles at right angles form irregular quadrate largenbsp;meshes. There are under the primary nerves two pairs of marginal veinletsnbsp;with the same degree of divergence as the primary ones (40�-50�). Thenbsp;form of this fine leaf and its mode of nervation are peculiar, and of anbsp;character analogous to that of leaves described under the generic namenbsp;of G-reviopsis in the �S�zanne Flora� by Saporta. There is, however, anbsp;marked difference in the primary temate nervation and in the entirenbsp;borders of the leaves. The two lower pairs of tertiary veins show also fornbsp;this leaf a relation to Credneria, and especially to the small leaf of Platanusnbsp;lleerii, pi. iii, fig. i. The secondary and tertiary nerves are of a differentnbsp;character.

Hob.�South of Fort Barker. Clis. Sternberg.

Ampelopliylliim ovatiim, Lesqx.

Hayden�s �Ann. Rep./� 1874, p. 355.

Celtic? ovata, Lesqx., �U. S. Geol. Rep.,� p. 66, pi. iv, ligs. 2, 3.

Leaves ovate, obtuse or undulate, truncate or obtusely pointed, enlarged toward tbe base and abruptly rounded and sub-truncate or cordifonn at base; nervation trifidnbsp;from the base, craspedodrome.

Though the relation of these leaves to the preceding species is mot very distinct, it is, however, more marked than to the leaves of Celtis.nbsp;But for the craspedodrome, and especially the ternate primary nervation,nbsp;they might be referable to Populus or Fopulites, having indeed some degreenbsp;of affinity to P. elegans, Lesqx., � U. S. Geol. Rep.,� vi, pi. iii, fig. 3.

HAMAMELIDEtE.

HAMAMELITES, Sap.

Leaves membranaceous, glabrous, petiolate, oblong-lanceolate or ovate; nervation pinnate; secondary nerves at an acute angle of divergence, craspedodrome, branching on the lower side; branches and subdivisions generally camptodrome.

The leaves described in this generic division have the essential characters of the leaves of both Hamamelis and Alnus.

DESCEIPTION OP SPECIES.

Haniamelltes tenuinervis, sp. nov.

Leaf broadly ovate, rounded at both ends, entire from the middle downward, regularly deeply undulate upward, pinnately nerved; lower lateral nerves alternate,nbsp;curving along the borders, camptodrome, mostly simple, the upper more oblique, simplenbsp;or branching, reaching the borders at the outer end of the undulations, or broad roundnbsp;teeth.

The base of the lateral medial nerves is somewhat decurrent in joining the midrib at an acute angle of divergence, while the lower ones, more open, join it in a broad curve nearly at right angles, all more or lessnbsp;curving in passing to the borders. The leaf is 5 centimeters long withoutnbsp;the short petiole (about 1 centimeter long), and nearly as broad. The onlynbsp;leaf known to me, to which this might be compared, is Parrotia pristina,nbsp;Heer, �FI. Arct.,� vol. iv, p. 83, pi. xxi, fig. 5 {Quercus fagifolia, Goepp.),nbsp;from which it differs not only by the leaf being shorter and broader, butnbsp;by the distribution of the lateral nerves, the two lower pairs being alternate and at a short distance from each other, as in Alnus serrulata, Linn.,nbsp;while the upper, sub-opposite, parallel, and more distant, are branchednbsp;and reach the borders at a more acute angle of divergence and a lessnbsp;pronounced curve.

Hob.�Four miles northeast of Minneapolis, Kansas. CM. Sternberg.

Hamamelites quadrangularis, Lesqx.

Hajden�s �Ann. Rep.,� 1874, p. 355.

Alniies qiiadrangularis, Lesqx., �U. S. Geol. Rep.,� vi, p. 62, pi. iv, fig. 1.

The leaf is small, slightly more coriaceous than the one described above; the borders are less distinctly undulate, and the secondary nervesnbsp;thick, closely parallel, less divided; the two lower pairs of nerves are thinner and closer, following the borders like marginal nerves.

Hamamelites Kansaseanus, Lesqx.

Plate IV, Fig. 5.

Hayden�s �Ann. Rep.,� 1874, p. 355.

Alnus Kanscuteanus, Lesqx., �XT. 8. Geol. Rep.,� vi, p. 62, pi. xxx, fig. 8.

From the specimen figured here, which is better preserved than that copied in the �U. S. Geol. Rep.,� vi, the description is somewhat modified.nbsp;The leaves are small, obovate in outline, cordate or obtuse at the gradually

FLORA OF THE DAKOTA GROUP.

narrowed base; the borders are deeply regularly undulate from below the middle; the two lower pairs of lateral nerves thinner than those abovenbsp;and more open are camptodrome, the other craspedodrome. The basilarnbsp;border seems to pass over the top of the petiole as in Menisfermites.

Hob.�This species is not rare in Kansas. The specimen figured was communicated by Prof. B. F. Mudge. No. 698 of the National Museum.

Hamamelites quercifoliiis, sp. nov.

Leaf oblong, coriaceous, lanceolate, rounded to the base, blunt at the apex, undulate on the borders; nervation pinnate, deep; lateral veins close, oblique, craspedodrome, branching on the lower side.

The leaf has great likeness to Bryopyllum fQuercusJ latifolmm, pi. iv, fig. i. It is about the same length but narrower, only 5� centimeters broadnbsp;in the middle, as in the preceding species; the two lower pairs of secondarynbsp;nerves are thinner, less oblique, more open than the eight others above.nbsp;These slightly curve in passing to the borders and enter, like the divisions,nbsp;the outside curve of the undulations.

Hah.�Bluff Creek, Ellsworth County, Kansas. Chs. Sternberg. There is only one specimen (No. 62a of the Museum Comp. ZooL, Cambridge).

Hamamelites? cordatus, sp. nov.

Plate IV, Fig. 3.

Leaves large, thickish, broadly oval-oblong, deeply narrowly cordate at base, obtusely dentate; nervation pinnate; lateral nerves oblique, slightlj�' curved in passingnbsp;toward the borders, much branching on the lower side, craspedodrome.

This fragment represents a leaf about 12 centimeters long, 7 to 8 centimeters broad. It is undulate-dentate all around, pinnately nerved, with the secondary nerves at equal distance, and parallel, except two pairs ofnbsp;smaller ones attached to the base of the lower lateral nerves. Of these,nbsp;the upper curves downward, branching and entering the borders by itsnbsp;apex and by its divisions, the lowest, simple and marginal, follows thenbsp;nearly auricled basal borders. Nothing is seen of the areolation. Somenbsp;simple parallel nervilles continuous and at right angles to the veins arenbsp;seen in the upper part of the leaf, which by its facies and some of itsnbsp;characters resembles a Viburnum.

Hab.�Near Fort Barker, Kansas. Chs. Sternberg.

72 nbsp;nbsp;nbsp;DESCRIPTION OF SPECIES.

MAGNOLIACE^.

MAGNOLIA, Linn.

Magnolia alteriians, Heer.

�TJ. S. Geol. Rep.,� vi, p. 92, pi. xviii, fig. 4.

Better specimens of this species, though not many, have recently been found in Kansas.

Magnolia Capellini, Heer.

�Phyll. Cr�t. du Neb.,� p. 21, pi. iii, figs. 5, 6.

Leaves coriaceous, broadly oval, very entire; secondary veins at an acute angle of divergence, curving to the borders, camptodrome.

The leaves of this species are similar in size and shape to those described as Ficus magmlicefoUa, pi. xvii, figs. 5 and 6. This last figure,nbsp;especially, does not differ from those published by Heer, except by thenbsp;closer secondary veins and by the base, which is slightly decurrent in thenbsp;leaves of Ficus, while in fig. 5 of Heer it is abruptly rounded and subcord-ate or subauricled. This appearance, however, may be merely casual,nbsp;resulting from the breaking of the base, as seen in all the leaves of thisnbsp;species described by Heer in �FI. Arct.,� vol. iv, pi. xxxiii. Two specimens of this species found in Colorado have the base decurrent upon anbsp;short petiole, and the nervation of the species.

Hab.�The two specimens mentioned above (Nos. 12 and 12h of the collection of the Museum of Comp. Zool., of Cambridge) are from Morrison, Colorado, found by A. Lakes. I have received a number of othersnbsp;more or less fragmentary from Kansas.

Magnolia speclosa, Heer.

�ilolet. FI.,� p. 20, pi. vi, fig. 1; ix, fig. 2; x; xi, fig. 1.

Leaves large, coriaceous, eUiptical-ovate, narrowed upward into a long acumen and downward to a thick petiole; medial nerve thick; secondary nerves curved, camptodrome. (Heer.)

The leaves of this species are enlarged in the middle and more rapidly attenuated to a long acumen and to the petiole than in the preceding. The medial nerve is much thicker. The specimens which I refer to itnbsp;differ in nothing from Heer's figure except, perhaps, by the lateral nerves,

FLORA OF THE DAKOTA GROUP.

which appear somewhat closer. As the veins are very indistinct the reference is somewhat uncertain.

//aj�Near Morrison, Colorado. A. Lakes. Specimen Nos. 13 and 13a of the Museum Comp. Zool., of Cambridge.

Magnolia tennifolia, Lesqx.

�U. S. Geol. Eep.,� vi, p. 92, pi. xxi, fig. 1.

Magnolia obovata, Newby.

�Later Ext. FI./� p. 15; �Illustr.,� pi. ii, fig. 2; iv, fig. 4.

Leaves large, obovate, entire, tMck and smooth, pointed and slightly decurrent on the petiole; nervation strong; midrib straight and extending to the summit; lateralnbsp;nerves pinnate, set at somewhat unequal distances, straight and parallel below, forkednbsp;and inosculating above, forming a festoon parallel with the margin; terminal nervesnbsp;forming an irregular network of polygonal and relatively large areoles. (Newby.)

Eah.�Blackbird Hills, Nebraska. Dr. Hayden.

Magnolia species.

Plate XI, Fig. 6.

A flattened immature receptacle or carpite of a Magnolia. The short-pediceled cone is oblong-obtuse, covered with short obtuse carpels.

Hal.�Near Morrison, Colorado. //. C. Beckwith.

LIRIODENDRON, Linn.

Liriodendron Meekii, Heer.

�Phyll. Cr�t. du Neb.,�� p. 21, pi. iv,figs. 3,4.

Leaves panduriform, emarginate at the top, bilobate; lobes obtuse; secondary veins branching. (Heer.)

Hal.�Tekamah, Nebraska. Professor Capellini.

Lirlodendron primfevum, Newby.

�Later Ext. FI.,�� p. 12; �Illustr.,�� pi. vi, figs. 6, 7.

Leaves three-lobed, upper lobe emarginate, all the lobes rounded; nervation delicate, medial nerve straight or slightly curved, terminating in the sinus of the superior lobe; secondary nerves gently arching upward, simple or forked near the extremities,nbsp;a few more delicate ones alternating with the stronger. (Ny.)

From comparison of specimens received from Greenland, Professor Heer considers this species, as also the leaves described as Leguniinositesnbsp;Marcouanus, Heer, and Pliyllites olcordatus, Heer (Newby., �Illustr.,� pi.nbsp;V, figs. 2, 3), as identical with the preceding species.

DBSCKIPTION OP SPECIES.

liiriodendron intermedium, Lesqx.

� U. S. Geol. Rep.,� vi, p. 93, pl. xx, fig. 5.

No other specimen has been found as yet than the fragmentary one described in the �Report.�

Liriodendron g-ig-antenm, Lesqx.

� U. S. Geol. Rep.,� vi, p. 9.3, pl. xxil, fig. 2.

A number of well-preserved specimens, recently obtained in Kansas, distinctly display the characters of this species originally described fromnbsp;a fragment, the upper lobe of a leaf only. The leaves are very large, 20nbsp;centimeters broad between the lower lobes, which are broad (6 centimeters),nbsp;oblong, rounded or obtuse, at right angles to the medial nerve; upper lobesnbsp;more oblique, shorter, narrowed and rounded to an obtuse point, joiningnbsp;the lower in a narrow deep sinus at a short distance (2 centimeters) fromnbsp;the thick medial nerves; lateral nerves parallel, nearly at equal distances,nbsp;slightly oblique, curved down in joining the medial nerve.

By the form of the leaves this species is more than any other related to the living L. Tulipifera. As far as can be seen from the fragment ofnbsp;L. intermedium, this last species differs much from L. giganteum, especiallynbsp;by the deeply emarginate leaf, the very oblique upper lobes at a great distance from the lower ones. The facies of the leaves of these two speciesnbsp;is far different.

Hab.�Two miles from Glasco, Kansas. The specimens. Nos. 206, 513, 535, found by Chs. Sternberg, like those of the four following species,nbsp;belong to the collection of the Museum Comp. Zool., Cambridge.

Liiriodendroii acuminatiini, Lesqx.

�Bull. Mus. Comp. Zool., Cambridge,� vol. vii. No. 6, p. 227.

Leaves small, about half as large as those of the preceding species, cut into two pairs of narrow linear accnminate lobes all arched upward, about 10 to 12 centimetersnbsp;long.

A remarkable species; the lobes, 1 centimeter broad, have only a medial nerve.

Hah.�Same location as the preceding. Specimens Nos. 476,504,504�.

Liriodendron cruciforme, Lesqx.

Ibid., p. 227.

Leaves large j upper lobes broad, square or equilateral, at right angles to the broad midrib; lower lobes narrow, linear, acuminate, much longer and arched upward.

FLORA OP THE DAKOTA GROUP.

The shape of the leaves is like that of an anchor, except that the medial nerve, or axis, does not pass above the upper border of the leaf,nbsp;which is cut flat, not, or scarcely, emarginate.

Hob.�Elkhorn Creek. Nos. 197,198, and some fragmentary ones.

liiriodendron semi-alatum, Lesqx.

�Bull Mug. Comp. Zool., Cambridge,� vol. vii. No. 6, p. 227.

Leaves divided at tbe base in two opposite short round lobes, obliquely cut in curving up to near the medial nerve and then diverging and enlarging upward into annbsp;obovate or spatulate entire lamina.

This form is somewhat like fig. 7 of pi. vi, Newby., � Illustr.,� the lower lobes longer obtuse and more defined, the upper part gradually enlarged,nbsp;spatulate, obtuse. It may be a distant form of L. Meekii.

Hah.�Seven miles from Glasco, Kansas. Specimens Nos. 472, 425.

Llrlodendron plnnatifidnm, Lesqx.

im., p. 227.

A simple leaf, with the general facies and the nervation of Lirioden-dron, but narrow linear in outline, subalternately trilobed on each side. The top and base of the leaf are broken, the lobes, separated by broad flatnbsp;sinuses, are half round, entire or irregularly undulate. The fragment isnbsp;9 centimeters long and 5 broad between the outside curves of the medialnbsp;lobes, which are a little larger than the upper and lower ones; the lateralnbsp;veins are close, oblique, parallel, distinct only at and near their point ofnbsp;union to the midrib. The fragment may represent a leaf of a diff�erentnbsp;genus, though its affinity is evidently with Liriodendron.

Hah.�Two miles from Glasco, Kansas. Specimen No. 531 (5261 fragment) .

LIRIOPHYLLUM, Lesqx.

Leaves subcoriaceous, square or broadly rhomboidal in outline, abruptly narrowed to a comparatively short petiole, split from the top to the middle along the line of thenbsp;medial nerve into two primary lobes much enlarged in the lower part, entire or sub-lobate or distinctly bilobate; nervation pinnate.

By the facies and the nervation these leaves have a great affinity to those of Liriodendron. Instead of being merely emarginate at the topnbsp;they are deeply cut down, nearly to the middle, in two lobes joined by anbsp;narrow more or less obtuse sinus. This is indeed the more markednbsp;difference.

DESOEIPTION OP SPECIES.

Eiriophyllum Beckwitliii, Lesqx.

Plate X, Fig. 1. nbsp;nbsp;nbsp;�

Hayden�s �Ann. Rep./� 1876, p. 482, mentioned.

Leaves large, square in outline, cut to near the base into two large diverging lobes; lobes bilobate, obtuse; primary nerve very thick, continuous to a short petiole,nbsp;bilid at a short distance above the base, the divisions ascending to the obtuse point ofnbsp;the upper lobes; secondary veins two, parallel, curved into the lower lobe, all withnbsp;few branches.

The abnormal form of the leaves of this genus renders their description difficult. In this species, which may be a variety or deformation of the following, the leaves are large, about 28 centimeters between the pointsnbsp;of the lower lobes, and nearly 20 centimeters from the base to the apex ofnbsp;the upper. They are divided into two halves from the top to 4 centimetersnbsp;above the base by the splitting of the medial nerve under an angle of 40�,nbsp;and each division is cut at the side in two short obtuse lobes separated bynbsp;a broad sinus. The lower lobe, nearly at right angles to the midrib, is traversed in its whole length by two parallel, strong, secondary nerves, apparently vanishing below the top (broken). Except very few oblique curvednbsp;tertiary veins, no other trace of nervation or areolation is distinct. Thenbsp;medial nerve from under the sinus downward is 3 millimeters broad�asnbsp;broad as the short pedicel broken 2 centimeters below the slightly decurrent base of the leaf.

Ifa�.�Near Morrison, Colorado, If. C. Beckwith. Found only in one good specimen.

Liriopliyllitm populoides, Lesqx.

Plate XI, Figs. 1 and 2.

Leaves smaller, broadly ovate, cuneiform at base, divided nearly vertically from the top to above the middle into two obtuse lobes, enlarged on the rounded sides abovenbsp;the base and there sometimes prolonged into a short obtuse lobe; medial nerve straight;nbsp;lateral nerves strong, parallel, equidistant, four pairs, effaced near the borders, rarelynbsp;branching; nervilles at right angles.

-In comparing these leaves with the preceding the essential characters are seen to be identical, though the appearance is far different. The largenbsp;size of the leaf and the subdivision of the two primary lobes in L. Beck-witliii are the more marked differences. But in fig. 1 of this species thenbsp;lower side is continued into a short lobe, indicating a subdivision like thatnbsp;of the leaf pi. x, fig. 1, and the nervation is of the same type as in the leaf

FLORA OF THE DAKOTA GROUP.

pi. xi, fig. 1; the two lower lateral nerves turn outside toward the short lobes, while the upper is evidently tending upward.

JIah.�With the preceding in numerous specimens. A. C. Beckwith, A. LakCiS. One specimen also has been found in Kansas.

Liriopliyllum obcordatum, sp. nov.

Leaf small, obovate, entire, narrowly deeply emarginate at tbe top, gradually narrowing to the petiole (broken); medial nerve narrow; lateral nerves at an acutenbsp;angle of divergence, alternate, camptodrome.

This leaf, 6 centimeters long and 3 broad above the middle, is cut from the top to one-third of its length into two obtuse entire slightlynbsp;diverging lobes by the splitting of the medial nerve, as in the two preceding species. It is perfectly entire, gradually narrowed from above thenbsp;middle, or cuneiform to the base, with two pairs of alternate distant secondary nerves at an acute angle of divergence and curving in passingnbsp;toward the borders. The tertiary nervation and the areolation are totallynbsp;obsolete.

Hab.�With the preceding. Rev. A. Lakes.

Carpites liriopliylli ? sp. nov.

Plate XI, Fig. 5.

An oblong seed 3 centimeters long, 7 millimeters broad in the middle, narrowed and blunt at one end, acute at the other; irregularly obscurelynbsp;lineate on the surface.

The reference of this fruit to Liriophjllum is hypothetical. The seed was found on one of the specimens of M. Beckwith, with leaves of L.nbsp;�populoides.

ANONACBiE.

ANONA, Linn.

Anona cretacea, sp. nov.

Leaf lanceolate or oblong-lanceolate, gradually narrowed to a short flattened petiole; medial nerve thick; secondary nerves open, nearly at right angles toward thenbsp;base, branching, camptodrome.

A fragment of leaf of which the lower half only is well preserved. It is similar in its size, form, and venation to A. lignitum. Ung., �Syllog.,�

DESCRIPTION OF SPECIES.

p. 25, pi. X, figs. 1-6. The relation of this leaf to this genus is as evident as it can be indicated by a single specimen representing only part of a leafnbsp;and no fruit.

Ilab.�Near Glasco, Kansas. CJis. Sternberg. No. 414 of the collection of the Museum of Comp. Zool., Cambridge.

MENISPERMACE^.

MENISPERMITES, Lesqx.

�U. S. Geol. Kep.,� vi, p. 94.

The definition of this genus has to be somewhat modified in this: the leaves are not only broadly deltoid and more or less distinctly trilobate, but also round or ovate, entire, with a camptodrome nervation.nbsp;From this, the group is subdivided in two sections, represented one bynbsp;lobate, the other by entire leaves.

Menlspermltes ol)tusilol)US, Lesqx.

Plate XV, Fig. 4.

�U. S. Geol. Rep.,� vi, p. 94, pi. xxv, figs. 1,9; xxvi,fig. 3.

M. ohtusilohus var., ibid., p. 95, pi. xxii, fig. 1.

Menlspermltes Sallnensls, Lesqx.

� U. S. Geol. Kep.,� vi, p. 95, pi. xx, figs. 2,3. nbsp;nbsp;nbsp;�

)

Menlspermltes acutllobus, sp. nov.

Plate XIV, Fig. 2.

Leaf large, triangular in outline, broadly rounded or nearly truncate at base, deltoid, dentate-lobate, amp;ve-nerved from near tbe base, coriaceous; nerves more or lessnbsp;branching on the lower side, craspedodrome, with their divisions; nervilles at rightnbsp;angles to the nerves, anastomosing in the middle of the areas.

The specimen figured is the only one seen. Comparison of the figures representing this species and M. obtusilobus, pi. xv, fig. 4, shows the closenbsp;affinity of the leaves�M. acutilobus merely differing by the large acutenbsp;distant teeth of the borders. The primary nervation is the same as thatnbsp;in pi. XV, fig. 1; the secondary veins are distant, equally oblique, andnbsp;curving toward the borders, scarcely branching, all craspedodrome, andnbsp;entering the teeth of the borders, a character already cemarked in all thenbsp;specimens of M. obtusilobus, whose secondary veins are more generally

FLORA OF THE DAKOTA GROUP.

craspedodrome even when the borders are not undulate-dentate, and always so when the leaves are undulate.

Hah.�Clay County, Kansas. H. C. Toimier.

Meuispermltes populifolivis, Lesqx.

Plate IV, Fig. 4.

Hayden�s �Ann. Rep.,� 1874, p. 357.

Leaf broadly ovate, obtuse, subcordate or truncate at base, palmately five-nerved from near the basal borders; primary lateral nerves at a more .acute angle of divergence, branching on the lower side; secondary nerves equidistant, parallel, all campto-drome.

The leaf is coriaceous, smooth on the surface, perfectly entire, bh centimeters long and as broad in its largest diameter below the middle.nbsp;The primary lateral veins diverge about equally from each other at annbsp;angle of about 30�; the lower is nearly simple and has still a thin marginalnbsp;veinlet underneath; they branch from the lower part, and the secondarynbsp;nerves at a distance above fotk only at their ends toward the borders.nbsp;The areas are crossed by very strong nervilles at right angles to the nerves,nbsp;anastomosing in the middle. The areolation is obsolete.

Hob.�South of Fort Harker. Chs. Sternberg.

Meuispermltes cyciopliyllus, Lesqx.

Plate XV, Fig. 3.

Hayden�s �Ann. Rep.,� 1874, p. 358, pi. vi, tig. 4.

Leaf thick, subcoriaceous, very entire, nearly round and centrally peltate, deeply concave, palmately five-nerved; inner lateral nerves curving inside, the outer open,nbsp;nearly at right angles to the medial nerve, all dividing by open straight branches anastomosing at a distance from the borders in double rows of arches; basilar veins 3 to 5,nbsp;diverging star-like from the central point.

The leaf is 7 centimeters long and 6 broad in its widest diameter; the middle is rounded downward and a little more narrowed upward to thenbsp;round subtruncate apex. The point of attachment of the petiole is nearlynbsp;central, and though surrounded by a series of nerves diverging star-like,nbsp;it has, like the other species of this genus, five primary nerves turningnbsp;upward, the lower ones representing marginal veins. The leaf is concavenbsp;from the point of attachment of the petiole, which passes down into thenbsp;stone, leaving an opening like the pipe of a funnel.

Hob.�Near Fort Harker, Kansas. CJis. Sternberg.

DESCEIPTIOX OF SPECIES.

Menispermites grandis, sp. nov.

Plate XV, Figs, 1, 2.

Leaves subcoriaceous, large, flat, nearly round, broader than long, peltate; borders entire or undulate; nerves radiating from the point of attachment, camjitodrome; primary nerves five.

This species differs from the preceding not only by the large size of the leaves but especially by the nervation Avhich is simply camptodrome,nbsp;the veins and their divisions curving along close to the borders and anastomosing in a single row of festoons. Even the medial nerve has thenbsp;same character and does not ascend to the borders, but is forked near thenbsp;apex in camptodrome divisions.

Hah.�Near Clay Centre, Kansas. IT. C. Towner.

Menispermites ovalis, Lesqx.

Plate XV, Fig. .'5.

Hayden�s �Ann. Rep.,� 1874, p. 357, pi. v, fig. 4.

Leaf narrowly oval or oblong, obtusely pointed, rounded at base, palmately five-nerved ; lateral nerves at an acute angle of divergence, the inner ones ascending to near tlie top, branching outside; branches numerous, parallel, curving along the borders in festoons.

This fine leaf, preserved nearly entire, is 7 to 8 centimeters long, 3i centimeters broad, nearly exactly oval-oblong, perfectly entire. It is lessnbsp;distinctly palmately five-nerved than the leaves of the other species of thisnbsp;genus; the two internal primary nerves are as strong as the medial one,nbsp;curve gradually nearly parallel to the borders, and near the top join thenbsp;branches of the midrib with which they anastomose in curves; the outsidenbsp;lateral nerves are thinner and shorter; they ascend also nearly parallel tonbsp;the borders, disappearing in the middle of the leaf in anastomosing withnbsp;branches of the lateral primary nerves. This is a mere deviation fromnbsp;the type.

Under the name of JJaphnogene Kanii, Heer has published (� FI. Arct.,� i, p. 112, pi. xiv), from the Miocene of Greenland, leaves related by theirnbsp;form to this Cretaceous species. The same kind of leaves are described bynbsp;Saporta and Marion in the �Flora of Gelinden,�' p. 63, pi. x, as Cocculusnbsp;Kanii. In these leaves the primary nervation is in three from the base;nbsp;in the Cretaceous leaf it is positively in five and therefore different,nbsp;appearing intermediate between that of the leaves described above asnbsp;Menispermites and that of Haphnogene, or Cocculus Kanii.

Hah.�Near Clay Centre, Kansas.

FLORA OF THE DAKOTA GROUP.

MALVACE^.

STEECUIIA, linn.

Leaves alternate, petiolate, palmately deeply trilobate; triple-nerved from the top �of the petiole.

This definition represents the characters of the coriaceous leaves which I refer to this genus, and which I separate from Aralia merely onnbsp;.account of the primary divisions. Most of these leaves have only thenbsp;primary nerves distinct and rarely any trace of the secondary veins. By anbsp;lower division of the lateral primary nerves, species referable, perhaps, tonbsp;this genus are described above as Aralia. If, as Schimper says, Sterculianbsp;Majoliana, Massal., �FI. Foss. Senig., p. 319,� is referable to the group ofnbsp;Stercxdia Labrusca, most of the species that are described as Aralia, if notnbsp;all, should be placed also with Sterculia. I do not admit this conclusion.

Sterculia lugubris, sp. nov.

Plate VI, Figs. 1-3.

Leaves coriaceous, large, divided near the cuneate base into three very long sub-linear acuminate lobes; primary nerves thick, distinct to the apex.

The leaves, narrowly cuneate, somewhat decurrent at base to the thick petiole, which they reach a little below the point of union of the primarynbsp;nerves, vary in length from 12 to 24 centimeters from the base to the apexnbsp;of the lobes, which are united by obtuse comparatively narrow sinusesnbsp;at a short distance�3 to 6 centimeters�from the base. The lobes, 1 to 2nbsp;centimeters broad in the middle, are slightly narrowed to their base, andnbsp;gradually tapering from the middle upward to an acuminate point. Thenbsp;lateral are curved downward, or scythe-shaped. No trace of secondarynbsp;nervation is visible.

There is in the collection of the National Museum a set of specimens representing an analogous form, though perhaps specifically different. Thenbsp;lobes, descending nearer to the base, are shorter (7-14 centimeters long),nbsp;straight, not recurved, linear-oblong, slightly narrowed from the middlenbsp;downward to the broad obtuse sinuses and gradually to the -apex. All thenbsp;points of the lobes are destroyed. Their divergence is about 25�.

Hah.�Colorado, near Golden. A. Lakes. The variety is from Kansas. Chs. Steryiberg.

C F 6

DESCEIPTION OF SPECIES,

Sterculia obtusiloba, Lesqx.

Plate VIII, Fig. 3.

Aralia tripartita, Lesqx., Hayden�s �Ann. Eep.,� 1874, p. 348, pl. i, fig. 1.

Leaves coriaceous, small, palmately three-lobed; lobes equal, linear, obtuse, very entire; secondary nerves obsolete.

The only leaf I have seen of this species is figured. It is 7 centimeters long, 6 centimeters hroad between the points of the lateral lobes, which diverge at an angle of 25� and are cut down to about two-thirds ofnbsp;the leaf. The medial lobe is a little narrower than the lateral (1 centimeter broad); the leaf is cuneate to the base and apparently a little decurrent to the petiole (broken); its surface is smooth. This leaf, followingnbsp;the definition of the genus, represents a Sterculia. Its name was changednbsp;accordingly.

Hai.�Near Fort Harker, Kansas. Chs. Sternberg.

Sterculia aperta, sp. nov.

Plate X, Figs. 2, 3.

Leaves subcoriaceous, palmately three-lobed, and triple-nerved from near the base; lobes lanceolate, blunt at the apex; angle of divergence broad.

This species is different from the preceding by the form of the broader lanceolate obtusely pointed lobes, the leaves not as thick and larger.nbsp;Fig. 3 shows traces of secondary nerves equidistant and curving to thenbsp;borders, the lower ones on the medtah nerve being at right angles to it.nbsp;These leaves are related to Sterculia lahrusca, Ung., a species which,nbsp;already present in the Eocene of France, is found also in all the stages ofnbsp;the Tertiary, including the Pliocene, in very variable forms. A numbernbsp;of specimens in the Museum of Comp. Zool. of Cambridge represent anbsp;form which seems intermediate between this and the preceding. Thenbsp;leaves are 8 to 10 centimeters long, somewhat thick but not coriaceous,nbsp;with lobes more or less diverging, linear-lanceolate, gradually narrowednbsp;above to a blunt point, nearly equal in length, 4 to centimeters long,nbsp;12 to 14 millimeters broad.

JIab.�Kansas. Found at divers localities. Chs. Sternberg.

FLORA OP THE DAKOTA GROUP.

TILIACE^.

GREVIOPSIS, Sap.

The remark made on the definition of this genus, �U. S. Geol. Rep.,� vii, p. 257, is applicable also to the Cretaceous leaves which I havenbsp;described under this generic name. The character of the nervationnbsp;especially relates them to those figured by the celebrated author in thenbsp;�Sezanne Flora.�

Grevlopsis Haydenii, Lesqx.

�U. S. Geol. Eep.,� vi, p. 97, pi. ill, figs. 2, 4; xxiv, fig. 3.

The leaf represented in this last figure was described first in � Amer. Jour. Sci. and Arts,� July, 1868, as Populites Jlahellata.

ACERACE^.

ACERITES, Newby.

Acerites prlstinus, Newby.

�Later Ext. FI.,� p. 15; �lllustr.,� pi. v, fig. 4.

Leaves petiolate, cordate at the base, flve-lobed; lobes entire, acute;? five strong and nearly equal veins radiate from the base into the lobes. The small nerves arenbsp;distributed over the surface in a flue net-work of which the meshes are sub-rectangular. (Ny.)

The figure represents a fragmentary leaf of the same character as those described and figured in �U. S. Geol. Rep.,� vi, p. 56, pi. ii, figs. 1,nbsp;3, under the name of Liquidamhar integrifolmm. The relationship of thesenbsp;leaves seems to be with the Araliacece, but it is as yet unascertained.

Negundoides acutifolivis, Lesqx.

�U. S. Geol. Rep.,� vi, p. 97, pi. xxi, fig. 5.

SAPmDACE.Jl.

SAPINDRS, Linn.

Sapindus Morrisoni, sp. nov.

Plate XVI, Figs. 1,2.

Leaflets subcoriaceous, short petioled, lanceolate-acuminate, unequal at the rounded narrowed slightly decurring base; lateral nerves alternate, parallel, curvingnbsp;in passing to the borders, camptodrome.

The fragment represents apparently the base of a large pinnately

DESCRIPTION OF SPECIES.

divided leaf, with leaflets alternate, short petioled, more enlarged on one side near the base. The fragments of leaflets distributed on the samenbsp;piece of coarse shaly sandstone indicate their original connection with anbsp;pinnate leaf. The lower part of the stem does not bear any fragments ofnbsp;the base of other leaflets attached to it. The stone is coarse, the nervation is obscure and has no trace of subdivisions of the secondary veins.nbsp;The leaflets average 12 to 14 centimeters in length, 2i to 3 centimeters innbsp;width in the broadest part below the middle.

Hah.�Near Morrison, Colorado. H. C. Beckwith.

Fragments of what I consider a variety of this species have been sent by Chs. Sternberg to the Museum of Comp. Zool., Cambridge, from Ellsworth County, Kansas (Nos. 24, 37). These represent two leaflets only,nbsp;both unequal at base, one about the same size as the specimens fromnbsp;Morrison, merely differing by the lateral veins being a little more oblique;nbsp;another leaflet is shorter and has the veins open proximate. It has beennbsp;found also at Atane with S. frodromus, Heer, ��Fi. Arct.,�� iii, p. 117, pi.nbsp;xxxiv, which it resembles.

FRANGULACEJS.

CELASTROPHYLLUM, Ett.

Celastrophyllum eusifolium, Lesqx.

quot;U. S. Geo). Kep.,� vi, p. 108, pi. xxi, figs. 2, 3.

ILEX, Linn.

Ilex strangrulata, Lesqx.

Plate III, Fig. 7.

Hayden�s � Ann. Rep.,� 1874, p. 359, pi. vii, fig. 8.

Leaf coriaceous, narrow, panduriform or strangled in the middle to a small angular lobe, rounded at base in narrowing to the petiole, entire in the lower part, little enlarged and irregularly distinctly obtusely dentate in the upper; secondary \reinsnbsp;proximate, in a very open angle of divergence, irregularly camptodrome or mixed.

This leaf is about centimeters long (point broken) without the li �centimeter long petiole. The general outline of the leaf is lanceolate, butnbsp;it is narrowed in the middle, as by erosion, nearly to the medial nerve,nbsp;and gradually enlarged upward by undulations or successive large obtusenbsp;irregular teeth. The surface is rugose; the lateral nerves, mostly camp-

FLOEA OF THE DAKOTA GEOUP.

todrome, follow close to the borders, the lower pair at a more acute angle of divergence as marginal veins, and those of the middle abruptly curved,nbsp;following also close to the borders with the same appearance as that ofnbsp;the basilar nerves. This nervation is related to that of some species ofnbsp;Mijrica, and still more of Ilex, like L Abichi, 1. herberidifolia, Heer, of thenbsp;Miocene. The areolation, distinct only on a small area where the epidermisnbsp;is destroyed, is in small, angular or irregularly square areoles. The narrowing of the leaf in the middle appears as produced by the gnawing ofnbsp;insects. But if the vein which follows the border is not a deceptive representation caused by the thickness of the leaf, this peculiar deformation is 'nbsp;natural. Leaves of Ilex are often variously and abnormally cut.

Hab.�Same as Dryofhyllum fQuercusJ Holmesii, in connection with coal strata of Southwest Colorado at a higher stage of the Cretaceous.nbsp;H. Holmes.

FKANGULACE^.

PALIURUS, Tonm.

Paliurus membranaceus, Lesqz.

�U. S. Geol. Rep.,� vi, p. 108, pi. xx, fig. 6.

RHAMinJS, Juss.

Brhamnus tenax, nesqx.

� U. S. Geol. Rep.,� vi, p. 109, pi. xxi, fig. 4.

Rbamnus prunlfolius, sp. nov.

Leaf coriaceous, ovate-lanceolate, rounded in narrowing to the base; medial nerve deep, straight; lateral nerves at short distance, parallel, open, arched in passingnbsp;toward the borders and curving along and close to them; nervilles close, numerous,nbsp;oblique to the nerves.

This leaf, 4 to 5 centimeters long (point broken), nearly 3 centimeters in the middle, resembles what Heer describes as Salix nermllosa, �Phyll.nbsp;Cr�t. du Neb., pi. i, tig. 3; but the lateral nerves are open, joining thenbsp;medial nerve nearly at right angles, parallel from the base of the leaf, whichnbsp;is not cuneiform but more rounded; the nervilles are oblique to the veins.nbsp;The nervation is that of a Bhamnus.

iTaJ.�Near Glasco, Kansas. Chs. Sternberg. No. 479 of the Museum Comp. Zool., Cambridge.

86 nbsp;nbsp;nbsp;DESCEIPTION OF SPECIES.

JUGLANDE^.

JUGLANS, Linn.

Juglans? Debeyana, Heer.

� U. S. Geol. Eep.,� vi, p. 110, pl. xx�i, figs. 1-5.

Populus Debeyana, Heer, �Phyll. Cr�t. dti !Veb.,� p. 14, pi. i, fig. 1; Newby., �Notes on Ext. FI.,� p. 17; �Illustr.,� pl. iv, fig. 3.

' nbsp;nbsp;nbsp;ANACARDIACE^.

Phyllites rlioifolius, Leeqx.

�U. S. Geol. Eep.,� vi, p. 111, pl. xxii, figs. 5, 6.

POMACEA^.

PYRUS, Lindl.

Pyriis? cretacea, Newby.

�Notes on Ext. FI.,� p. 12; �Illustr.,� pl. ii, fig. 7.

Leaves petioled, small, roiindisli, oval or elliptical, often slightly emarginate, entire or finely serrate; medial nerve strong below, rapidly diminishing toward thenbsp;summit; lateral nerves four or five pairs, with intermediate smaller ones, divergiug fromnbsp;the midrib at unequal angles, curved toward the summit, where they anastomose in anbsp;series of arches parallel with the margins; tertiary nerves forming a net-work of whichnbsp;the areoles are somewhat elongated. (ISTy.)

This leaf seems to be a small, lateral leaflet of Juglansquot;? Deheyana.

Hah.�Smoky Hills, Kansas. Dr. Hayden.

AMYGDALE^.

PRUNUS, Tourn.

Prunus cretacea, Lesqx.

�U. S. Geol. Eep.,� vi, p. Ill, pl. xxiii, figs. 8, 9.

LEG�MINOSAE.

LEGUMINOSITES, Auct.

Leguiniiiosites cultriformis, sp. nov.

Plate X, Fig. 4.

Fruit (legume) stipitate, rounded to the point of support, enlarged above it and gradually tapering up to an obtuse point; stipe enlarged at base.

The legume is7i centimeters long without its stipe (a little more than 2 centimeters), 13 millimeters broad above the base, the widest part, andnbsp;gradually narrowed, by the inclination of one of its sides only, to a blunt

FLORA OF THE DAKOTA GROUP.

point. The whole surface is smooth with only some fragments of longitudinal lines.

No remains of Leguminosoe have been discovered in the Dakota Group except the one figured as above. It appears to be a stipitate legume withnbsp;analogy of form and size to those of Loncliocarpus, H. B. amp; Kunth., a genusnbsp;mostly represented in the West Indian Islands, the equatorial America.

OENEBA AND SPECIES OF �NOEBTA1N BELATION.

ASPIDIOPHYLLUM, Lesqx.

Hayden's �Ann. Rep.,� 1874, p. 361.

Leaves large, triangular in outline, palmately trilobate, truncate or rounded to a peltate base; nervation coarse; primary nerves triM, from a short distance abovenbsp;the peltate base of the leaves, the lateral, at an open angle of divergence, sometimesnbsp;curved downward; secondary nerves generally close, parallel, camptodrome, generallynbsp;simple, joined by strong nervilles at right angles.

This group has a great affinity by the form of the leaves and the nervation to that of the Sassafras (Araliopsis). Indeed at first sight itnbsp;appears to differ from it only by the addition of a basilar shield. Thenbsp;nervation, however, differs in some characters, the primary nerves beingnbsp;at a more open angle of divergence, as are also the secondary ones, whichnbsp;are also more curved in passing to the borders. The rounded more or lessnbsp;enlarged shield of the base is nerved by the secondary nerves graduallynbsp;declining downward, one pair generally attached under the point of unionnbsp;of the primary nerves, the others derived from the base of the medialnbsp;nerve and passing downward, the lowest nearly perpendicular in direction, and all abruptly curving and following the borders in continuousnbsp;flexures. The disposition of the lower lateral nerves has an analogy tonbsp;that of Oredneria, with the difference that in Credneria the lower secondarynbsp;nerves are all at right angles to the midrib. The same kind and degree ofnbsp;analogy is marked between these leaves and those of Protophyllum andnbsp;Pterospermites, and also those of Platanus.

Aspldiophylliim trllobatiim, Lesqx.

Plate XII, Fig. I; XIII, Figs. 1-5; XIV, Fig. 1.

Leaves generally large, coriaceous, triangular or rhomboidal in outline, deeply obtusely trilobate, broadly cuneate to the base, enlarged into a half-round entire auricle.

The leaves vary in size from 10 to 24 centimeters long and from 10

DESCRIPTION OF SPECIES.

to 30 centimeters broad between the lateral lobes. Some of them, apparently constituting a variety of the normal form, are not half as large, their nervation is still coarser and the surface rugose, as in pi. xiii, fig. 1, andnbsp;especially pi. xiv, fig. 1. All have been found at the same locality, mostlynbsp;alone. There is also a marked difference in the expansion of the peltatenbsp;base, which is generally half-round, as in pi. xiii, figs. 1, 3, but whichnbsp;sometimes is regularly dentate lobate around, as in pi. xiii, fig. 6. Butnbsp;this fragment may be referable to the following species.

Hah.�Found in numerous specimens 3J miles south of Fort Harker. Clis. Sternberg.

Aspldiophyllum dentatum, sp. nov.

Leaves smaller, palmately three-lobate, peltate at the base; lateral lobes trilobate, the medial long, all dentate in the upper part; secondary nerves camptodrome; base of the leaves contracted into a fan-like five-lobed basilar shield.

The leaves have the same general facies as those of A. trilohatum^ differing by their texture not being as thick, the nervation not as coarse,,nbsp;and by the base of the leaves being contracted under the point of divisionnbsp;of the primary nerves into a narrow neck half a centimeter broad ouljvnbsp;and then abruptly enlarged into a fan-like five-lobed or deeply dentatenbsp;shield or stipule 4 centimeters broad between the summits of the lateralnbsp;teeth and 2 centimeters vertically from the base of the medial nerve to thenbsp;end of the lower lobes. This forni or species with the dentate borders ofnbsp;the middle lobes and the subdivisions of the lateral lobes has its affinitynbsp;to Sassafras [AraliopsisJ cretaceum, while the preceding species has it tonbsp;S. f Araliopsis J mirabile. Another specimen of the same group shows thenbsp;basilar shield transversely oval and entire, stipuliform, also separated fromnbsp;the leaf by a narrow neck. But of this I have seen only a mere fragment,,nbsp;the base of a leaf. It possibly represents still.another species.

Hob.�Eight miles northeast of Minneapolis. Chs. Sternberg. Specimens 607 and 614 of the Museum Comp. Zool. Cambridge.

Aspldiophyllum platanifolium, sp. nov.

Plate II, Fig. 4.

Leaves of various sizes, thinner or not coriaceous, rhomboidal in outline, irregularly short trilobate, triple-nerved high above the base; secondary nerves distant and irregular in position and direction, craspedodrome, with camptodrome divisions.

The few leaves I have seen of this species are about of the same size,

FLOEA OF THE DAKOTA GKO�P.

15 centimeters long, 13 centimeters broad between tbe lateral short broadly obtuse lobes. The substance of the leaves is not coriaceous, rather thinnbsp;or membranaceous; the nervation not as coarse; the primary veins onlynbsp;half as thick as in the preceding species. The medial nerve descends tonbsp;near the basilar margin before passing under it, and thus the, tertiary ornbsp;marginal veins join the lower part of the medial nerve at right angles asnbsp;in Credneria; the upper secondary nerves, only three pairs, are very distant and oblique, not parallel nor equal in distance, and reach the marginsnbsp;by their ends as craspedodrome while all their divisions are camptodrome.nbsp;The relation of this leaf to Platanus is quite distinct, as will be seen innbsp;comparing it to P. Heerii, �U. S. Geol. Rep.,� vi, pi. ix, fig. 1.

Hah.�Clay County, Kansas. H. C. Towner.

Protophyllum, Lesqx.

�U. s. Geol. Eep./� vi, p. 100.

Protopliyllum Sternbergii, Lesqx.

Ihid., p. 101, pis. xvi, xviii, fig. 2.

Protophyllum Leconteanum, Lesqx.

Ihid., p. 103, pi. xvii, fig. 4; xxvi, fig. 1.

Protophyllum Kebrascense, Lesqx.

Ihid., p. 103, pi. xxvii, fig. 3.

Protophyllum quadratum, Lesqx.

Ihid., p. 104, pi. xix, fig. 1.

Protophyllum minus, Lesqx.

Plate IV, Fig. 6.

Ihid., p. 104, pi. xix, fig. 2; xxvii, fig. 1.

This species sometimes has the leaves very rugose and thus resembles P. rugosum, which is, however, very different in the nervation, the largenbsp;size of the leaves, etc.

Protophyllum multinerve, Lesqx.

Ihid., p. 105, pi. xviii, fig. 1.

From numerous specimens less fragmentary than the one figured the leaves are seen to be round or transversely oval with borders entire. Thenbsp;sizes vary from 7 to 14 centimeters long and 9 to 18 centimeters broad.nbsp;The nerves are very close and numerous around the peltate base of thenbsp;leaves; above it they count 8 to 10 pairs, the lower forking generally once,nbsp;the upper simple.

DESCRIPTION OP SPECIES.

Protophyllum rugosum, Lesqx.

�U. S. Geol. Eep.,� vi, p. 105, pi. xvii, lig�. 1, 2; pi. xix, fig. 3.

Among other leaves of this species there is one entirely preserved, No. 747, in the Museum Comp. Zool. Cambridge. It measures 17 c�nti-meters long, 10 broad, is undulate on the borders or somewhat dentatenbsp;by the projection of the lateral veins, and agrees in every point by formnbsp;and nervation with the figure and description of the species (loc. cit.).

Protophyllum Haydenii, Lesqx.

Ihid., p. 106, pi. xvii, fig. 3.

Protopliyllum crednerioides, Lesqx.

Plate II, Figs. 1-3.

Hayden�s �Ann. Eep.,� 1874, p. 363, pi. iii, fig. 1; viii, fig. 4.

Leaves small, nearly round, broadly cuneate or subtruncate at base, long petioled; borders entire or more generally undulate; nervation obscurely trifld; secondary veiusnbsp;parallel, equidistant, at various angles of divergence, more or less branching.

The leaves vary in size from 6 to 8 centimeters both ways. The borders are either deeply undulate or nearly entire, though all the nervesnbsp;and their divisions are craspedodrome; the secondary nerves are open, atnbsp;right angles toward the base. The areolation is formed by anastomosingnbsp;of continuous nervilles at right angles to the veins and by their subdivisions in the areas, also at right angles, forming very small quadratenbsp;meshes, as seen in fig. 3. As in the other species of the genus, the nervation is more or less obscurely trifid. The lower primary lateral nervesnbsp;being at a distance above the borders have under them, as in Oredneria,nbsp;two pairs of thinner secondary or marginal nerves at right angles. Butnbsp;as the lower veins often branch like the upper ones and have the samenbsp;direction as those above, the nervation sometimes appears pinnate, as innbsp;fig. 1. The ternate disposition is, however, distinct in fig. 3.

Hah.�Kansas. Not rare. Chs. Sternberg, H. C. Towner.

Protophyllum ? Mudgel, Lesqx.

�U. S. Geol. Eep.,� vi, p. 106, pi. xviii, fig. 3.

91

FLOEA OF THE DAKOTA GEOUP. ANISOPHYLIUM, Lesqx.

�U. S. Geol. Rep.,� vi, p. 98.

Anisopliyllum semi-alatum, Lesqx.

Ibid., p. 98, pi. vi, figs. 1-5.

No other specimens have been seen of this species since it was first examined.

EREMOPHYLLUM, Lesqx.

Ibid., p. 107.

E r e m o p li y 11 u m li m b r i a t u m, Lesqx.

Ibid., p. 107, pi. viii, fig. 1.

The specimen figured is the only one seen of this kind.

VEGETABLE REMAINS OF UNCERTAIN AFFINITY.

' nbsp;nbsp;nbsp;PHYLLITES, Auct.

Pliyllites Vanonse, Heer.

Ibid., p. 113, pi. XX, fig. 7; xxvlii, fig. 8.

Pliyllites rliomboideus, Lesqx.

Ibid., p. 112, pi. vi, fig. 8.

Pliyllites cotiiius, Lesqx.

Hayden's �Ann. Rep.,� 1874, p. 364.

Bumella Marcouana, lieer, � U. S. Geol. Rep.,� vi, p. 90, pi. xxviii, fig. 2.

Pliyllites iimbonatus, Lesqx.

�U. S. Geol. Rep.,� vi, p. 113, pi. xix, fig. 4.

Pliyllites amorphus, Lesqx.

Ibid., p. 113, pi. xxii, figs. 3, 4.

Ptenostrobus, Lesqx.

Ibid., p. 114.

Ptenostrobus Nebrascensis, Lesqx.

Ibid., p. 114, pi. xxiv, fig. 1.

CARPOLITHES, Auct.

Carpolithes speeies?

Ibid., p. 114, pi. xxvii, fig. 5; xxx, fig. 11.

CAUDEX.

Caudex spiiiosus, Lesqx.

Cmdiniia spinosus, Lesqx., Ibid., p. 115, pi. xxvii, fig. 4.

DESOEIPTIOF OF SPECIES.

COXCIiUDING REMARKS.

The Flora of the Dakota Group, as already remarked,, is considered as relating the formation which it represents to the Cenomanian or Middlenbsp;Cretaceous. In order to ascertain the validity of the relationship, andnbsp;also to have a clear exposition of the general characters of the vegetationnbsp;of the time, I have prepared the following table of the species of fossilnbsp;plants which have been described by authors as referable to that stage ofnbsp;the Cretaceous.

1st. Those from Atane, Greenland; described by Heer in the �FI. Arct.,� including part ii of vol. vi, recently published.

2d. The species known from Moletein and Quedlinburg, described by the same author.

3d. The plants found in the Quader sandstone of the Hartz and of Bohemia, described or mentioned in different memoirs by Hampe, Stiehler,nbsp;Dunker, Goeppert, Feistmantel, Corda, etc.

4th. The species described from Niedershoena in Saxony, by d�Ettings-hausen.

FLOEA OF THE DAKOTA GEOT7P.

TABLE OF DISTRIBUTION OF THE PLANTS OF THE CRETACEOUS CENOMANIAN

FORMATION.

94 nbsp;nbsp;nbsp;DISTEIBUTIOK OP SPECIES.

TMe of Distribution of the Plants of the Cretaceous Cenomanian Formation�Continued.

PLOEA OF THE DAKOTA GEO�P.

ToMe of Duirihution of the Plants of the Cretaceous Cenomanian Formation�Continued.

DISTEIB�TION OF SPECIES.