Office copy LIBRARY OF THE NEW YORK BOTANICAL GARDEN MEMOIRS OF THE NEW YORK BOTANICAL GARDEN Vor. III STUDIES OF CRETACEOUS CONIFEROUS REMAINS FROM KREISCHERVILLE, NEW YORK | BY ARTHUR HOLLICK, Pu.D. NEW YORK BOTANICAL GARDEN AND ` EDWARD CHARLES JEFFREY, Ph.D. HARVARD UNIVERSITY Issued May 20, 1909. MEMOIRS NEW YORK BOTANICAL GARDEN Vor. Ш STUDIES OF CRETACEOUS CONIFEROUS REMAINS FROM KREISCHERVILLE, NEW YORK BY ARTHUR HOLLICK, PH.D. NEW YORK BOTANICAL GARDEN AND EDWARD CHARLES JEFFREY, Ph.D. HARVARD UNIVERSITY LIBRARY CEW YORK ТА NEC SL, Us KUEN., Issued Мау 20, 1909. Б HOT STUDIES OF CRETACEOUS CONTFEROUS REMAINS FROM KREISCHERVILLE, NEW YORK BY ARTHUR HOLLICK, Ph.D. NEW YORK BOTANICAL GARDEN AND EDWARD CHARLES JEFFREY, Рн.р. HARVARD UNIVERSITY PUBLISHED BY THE AID OF THE Davin Гут Funp BEQUEATHED BY CHARLES P. DALY NEW YORK 1909 PRESS OF THE NEW ERA PRINTING COMPANY LANCASTER. PA JUN 3- 1909 PREFACE This Memoir is designed to present the results derived from crit- ical studies of coniferous remains from certain of the Cretaceous deposits at Kreischerville, Staten Island, New York, especially from two exposures, locally known as the Androvette pit and the Drum- mond pit—excavations made for the purpose of obtaining clay and sand for the manufacture of fire-brick, terra-cotta, etc. The material upon which the studies were based was secured largely through the cordial coöperation of Ше Messrs. Androvette, owners of the property, by whom permission was given to collect specimens at all times, and also to make special excavations in search of desirable material, and who provided, at their own expense, the labor to prosecute the work of digging and rough handling of the material. In connection with the preparation, study and illustration of the material the following individual work of the authors 15 indicated. The senior author is responsible for the selection and preparation of the larger specimens and for identifications based upon external characters, this part of the work having been done at the New York Botanical Garden; the junior author for the selection, preparation and microscopical examination of the smaller specimens and lignites, and for identifications based upon internal structure, these features of the work having been accomplished in the Phanerogamic Labora- tories of Harvard University. Plates 1 and 2 are reproductions of photographs taken by the senior author; Plates 3-6 of drawings of selected specimens made by the senior author; Plates 7-10 of drawings made by the senior author from enlarged photographs of selected specimens made by the junior author; Plates 11-29 of photo- micrographs of sections made by the junior author. In all other respects this Memoir is a joint contribution. ARTHUR HOLLICK, New York Botanical Garden. EDWARD С. JEFFREY, Harvard University. CONTENTS PAGE INTRODUCTION I Previous References to de Piledbodiny of "T Eod I Description of the Deposits 5 Stratigraphic Relations . 5 Occurrence and General Car of the Plant Remains 6 Methods Pursued in the Examination of the Material 8 Impressions in the Clay Layers 8 Lignitic Material . 8 Amber . II DESCRIPTIONS OF SPECIMENS ; | . : 12 Abietineae . я . А | қ 12 Pinus . . | | А ; : : & 12 Prepinus А . . : | : А . 19 Pityoxylon | à я 2 . 20 Araucarineae К А ; | Е қ a 22 Androvettia ; ; | : : e 22 Raritania | А А : ; А . 26 IH iddringtonites : i : : . 29 T huites . : А : қ 3І ula ; : қ я : = 33 Geinitzia . . : : . 38 Eugeinitzia . : я 43 Pseudogeinitzia . | 45 Protodammara | А 5 . 46 Anomaspis . ‚ 49 Sphenaspis . | ; я А 51 Dactyolepis . : А j Е Е 52 Pityoidolepis : қ : 53 Araucariopitys А ; : . S54 Brachyoxylon А А ; А : 54 Araucarioxylon . 58 Coniferous Remains of Uds ehr . 61 Sequoia heterophylla . | : р | | 61 Juniperus hypnoides А | ; : ; . 61 Czekanowskia capillaris . : | 3 ; | + 63 viii MEMOIRS OF THE NEW YORK BOTANICAL GARDEN Unidentified Twig Cupressinoxylon Strobilites CONCLUSIONS SUMMARY PLATES INDEX . STUDIES OF CRETACEOUS CONIFEROUS RE- MAINS FROM KREISCHERVILLE, NEW YORK INTRODUCTION PREVIOUS REFERENCES TO THE PALEOBOTANY OF THE LOCALITY In the early reports of the New York State Geological Survey may be found several references to the geology of the vicinity of Kreischerville, but they are very brief, and the fossil plant remains in connection with the Cretaceous deposits receive only a few meager lines of description. Dr. Lewis С. Beck refers to the lignite, and says: “ On Staten Island, near Rossville, is a stratum of this mineral, from three to six inches in width, apparently below high-water mark. Sometimes it has the colour and appearance of wood; at others it is quite compact, and has a dark brown or black colour, and resembles jet . . . Prob- ably, also, amber will be found here, as it accompanies a similar formation at South Amboy, in New Jersey, a few miles from the above locality." Subsequently? he refers to the Rossville lignite as having “ the appearance of wood charred by the action of an acid,” and suggests that it probably owes its origin to the action of the excess of acid which the clay contains “upon drift wood from time to time deposited on this shore of the island." From these descrip- tions it may be inferred that the ideas of those days in regard to the nature, origin and antiquity of this lignite were exceedingly vague. *Investigations prosecuted with the aid of a grant from the Botanical Society of America. ? First Ann. Rept. N. Y. State Geol. Surv. Assembly Document No. 161, Feb. тт, 1837, р. 59. *'Third Ann. Rept. N. Y. State Geol. Surv. Assembly Document No. 275, Feb. 27, 1839, p. 16. 2 I 2 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN Mr. William W. Mather also mentions the abundance of lignite in the clay exposure in this part of Staten Island, and says* that the clay is “ similar in its general characters to that of Cheesequake and Matavan Point, on the Jersey shore, and it appears to have a similar geological position.” Practically the same notes may also be found included in the final reports by these authors,? but after the issue of these there followed a period of some forty years during which nothing of a scientific character appears to have been recorded, or at least published, in regard to the deposits or their accompanying organic remains. In 1881 Dr. Nathaniel L. Britton read a paper before the New York Academy of Sciences, entitled “On the Geology of Richmond County, N. Y.,”* in which the Cretaceous age of the deposits in question is recognized as settled, and he says (p. 172): “ Lignite and pyrites are frequently found in the clay excavations. . . . As the lignite dries, it cracks up into little pieces, thus destroying the texture of the fossil wood composing it, and making it very diffi- cult to retain good specimens. No fossil leaves or shells have been taken from the clays . . . but it is not improbable that they will be found at some future time, when the excavations are more advanced than at present." During this same year the Natural Science Association of Staten Island was organized and the investigation of local geological fea- tures was begun on a systematic basis. In 1885 the prediction made by Dr. Britton was verified, and in the account of the meeting of the Association held on December 12, 1885, may be found the fol- lowing record: “ Mr. Hollick stated that an important find of fossil vegetable remains was discovered in the Kreischerville fire clay on Nov. 15; but that it was deemed best to merely place the fact upon record, and to leave the full account of the same until the material collected had been studied more carefully." Subsequently, at the meeting held on February 13, 1886, these remains were reported upon as follows:* “ All of our specimens were + Second Ann. Rept. N. Y. State Geol. Surv. Assembly Document No. 200, Feb. 20, 1838, P. 137. * Nat. Hist. N. Y., Pt. III, Mineralogy, pp. d 191, 192. 1842.—Ibid., Pt. IV, Geology. Geol. First Geol. Dist., pp. 266, 267. 18 ° Annals N. Y. Acad. Sci. 2: a pls. 15, P 1882. "Proc. Nat. Sci. Assoc. Staten Island 1: 27. 1885. ° Proc. Nat. Sci. Assoc. Staten Island 1: 31. 1886. CRETACEOUS CONIFERALES 3 found in a narrow stratum, nowhere more than a foot in thickness, near the surface of the bed. The stratum was conspicuous from its dark color, due to the mass of lignified vegetable matter which it con- tained. Much of this was broken twigs and branches, some pieces being quite large and showing the woody texture very beautifully; they, however, fell in pieces upon exposure to the air." A list of the angiosperm genera, determined from the leaves, is given, and the final paragraph on the gymnosperms is to the effect that “ Pine needles are distributed plentifully throughout. . . Another con- Чег which has left its mark, 15 so close to Sequoia that it has been referred to that genus. . . . There are also little masses of a yellow substance here and there which I take to be a fossil gum or amber. This could, however, only be determined by chemical analysis. Fruits and seeds should be sought carefully, as they are generally quite satisfactory to determine, being less liable to destruction than the leaves." The excavation from which these remains were obtained was shortly afterwards partly filled in, and for several years no further collections were made; but in 1892 new excavations again exposed the plant-bearing layers, and specimens a were reported upon at the March meeting of that year, уі2. “ Mr. Davis presented unusually fine specimens of lignite, apparently coniferous, from the clay beds at Kreischerville. “The specimens were of the appearance and consistency of Jet and contained considerable amber. . . . Mr. Hollick stated that a record should be made of the re-discovery, since the last meeting, of plant remains in the clay pit at Kreischer- ville, where they were first found and noted six years ago. . . . Аз on the former occasion, however, the specimens found were too frag- mentary for accurate determination. They occur thickly massed together in confusion, in a stratum averaging, in the recent exposure, about two feet in thickness. It was ци found possible to so ‚separate them as to obtain perfect specimens.” These discoveries, meager as they were, nevertheless served to definitely correlate the Staten Island clays with those of New Jersey and to determine their Cretaceous age beyond further question. Occasional fragmentary plant remains were brought to light from time to time subsequently, as new excavations were made, and some ? Proc. Nat. Sci. Assoc. Staten Island 3: 12. 1892. 4 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN of these discoveries also may be found recorded, as follows 19 “A bed of fossil leaves, similar to those found at Kreischerville, has recently been discovered at the bottom of a clay pit at Green Ridge, but the specimens were too fragmentary for determination. The discovery was made by Mr. Heinrich Ries, while engaged in examining the clays of the Island for the New York Geological Survey. . . . Mr. Davis presented a finely preserved specimen of a conifer, from one of the Kreischerville clay pits, identified provisionally as a Juniperus.” It was about this time that what is known as the Androvette pit was opened up, which has since yielded the bulk of the material upon which our work is based. The first specimens found in this pit were collected by Mr. Wm. T. Davis," but it was not until some years afterward that the richest part of the plant-bearing layers were uncovered. During the autumn of 1904 specimens of amber, leaves and other remains were found in abundance and a systematic study of the deposits was begun by the senior author. Preliminary reports were made to the Natural Science Association of Staten Island,” and a brief note was published, with illustrations, in the Journal of the New York Botanical Garden for March, 1905. Particular attention was given to the amber and this feature of the deposits was made the basis of a paper which was read at the Philadelphia meeting of the Botanical Society of America, in 1904, under the title “ The Occurrence and Origin of Amber in the Eastern United States. 13 It was this paper which led to communication between the authors and to the joint work subsequently undertaken, the first result of which was a preliminary paper, entitled “ Affinities of Cer- tain Cretaceous Plant Remains Commonly Referred to the Genera Dammara and Brachyphyllum,”** designed to indicate the value of critical examination of the Kreischerville lignitic material. The investigations there outlined have since been systematically prose- cuted, both in the field and in the laboratory, and the results thus far attained, in so far as the coniferous remains are concerned, are set forth in this Memoir. Other contributions, also based either wholly or in part on the Kreischerville material, are as follows: ” Proc, Nat. Sci. Assoc. Staten Island 3: 20. 1892. “Proc. Nat. Sci. Assoc. Staten Island 3: 23. 1892. "Proc. Nat. Sci. Assoc. Staten Island 9: 31. 1904.—Ibid. 35. 1906. * Amer. Nat. 39: 137-145. pls. 1-3. 1905. * Amer. Nat. 40: 189-215. pls. 1-5. 1906. CRETACEOUS CONIFERALES 3 Hollick, A. “Origin of the nn Found on Staten Island.” Journ. N. Y. Bot. Gard. 7: 11, 12. 1906.—“ A Fossil Forest b " Proc. Staten Island 1. Arts & Sci. 1: 21-22. 1006. “ Insect Borings in Cretaceous Lignite from Kreischerville.” Ibid. 23, 24. Jeffrey, E. C., and Chrysler, M. A. “On Cretaceous Pityoxyla.” Bot. Gaz. 42: 1-15. pls. 1, 2. 1906. Jeffrey, Е.С. “The Wound Reactions of Brachyphyllum.” Annals Bot. 20: 383-394. pls. 27, 28. 1906.—" Araucariopitys, а New Genus of Araucarians." Bot. Gaz. 44: 435—444. pls. 28-30. 1907.— Оп the Structure of the Leaf in Cretaceous Pines.” Annals Bot. 22: 207-220. pls. 13, 14. 1908. DESCRIPTION OF THE DEPOSITS Stratigraphic Relations.—A complete series of (һе Kreischerville deposits shows Cretaceous clays and sands at the base, apparently including both the Raritan and the Cliftwood formation, above which are sands and gravels of Tertiary or early Pleistocene age, probably referable to the Pensauken formation, with Quaternary boulder till at the surface; but all of these deposits are seldom represented in any one section and one or more of the formations may be locallv want- ing, or their exact stratigraphic relations in certain of the sections may be more or less obscurely defined. It is often difficult, for example, to differentiate satisfactorily between the two Cretaceous formations, on account of their general lithologic similarity and the fact that certain wellknown floral elements are common to both, while in certain sections it 15 not easy even to determine whether only one of the formations is represented or whether both are present, for the reason that the line of contact between them 15 generally not well defined. Cross-bedding, and the occurrence of lenticular beds or pockets of lignitic debris between the sand and clay layers, are con- spicuous original features of the deposits at certain levels, indicating running water conditions of deposition, while elsewhere there are pronounced secondary deformations, due to ice .pressure, repre- sented by faulting and folding of the strata. The stratigraphic relations are therefore liable to be misinter- preted unless they are subjected to careful study. Аз a rule, how- ever, the lower white and gray clays and sands are regarded as 6 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN referable to the Raritan and the upper red and Бий sands to the Cliffwood. The phenomena of cross- and lenticular-bedding is best exemplified in the Androvette pit, and some of the sections there exposed also afford the best examples of the stratigraphic relations between the several deposits represented. The most extensive vertical section in this pit is shown in Plate 1. The contact between the Raritan and Cliffwood formations is comparatively well defined and the eroded surface of the latter, together with the overlying unconformable Pensauken sands and gravels, are finely exposed. Boulder till is not present in this section, which underlies a limited unglaciated area. In the Drummond pit the Cretaceous deposits are almost exposed at the surface of the ground. The Pensauken sands and gravels are lacking, having been entirely eroded by glacial action, but they are represented, to a certain extent, as more or less conspicuous con- stituents of the till, which rests in a thin layer directly on the Cre- taceous deposits. These latter here consist of relatively thick clay strata, in which sandy layers are not conspicuous features, and cross- bedding or other indications of the action of running water are wanting; but well defined faulting and slight arching or folding, evidently caused by the weight and thrust of ice, are quite apparent. A view of a portion of this pit is shown in Plate 2, where the arching of the strata, in a gentle anticline, may be noted. These clays are located at a considerable elevation above and to the south of those in the Androvette pit, and as the normal dip of the strata is towards the southeast they should, theoretically, represent a higher and more recent geological horizon than the latter; but the deformation and disturbance to which they have been subjected renders it somewhat hazardous to venture any positive opinion on this point. Occurrence and General Characters of the Plant Remains.—The only plant-bearing beds belong to the Raritan Cretaceous deposits, and plant remains of one kind or another are present in most of the layers. They are sometimes represented merely by finely comminuted charcoal or lignite, but larger fragments and branches and logs of lignite, often charred on the exterior and impregnated or coated with pyrite, are more or less abundant. In many of the latter amber occurs in the interstices. In addition to the charcoal and lignites the remains also include irregular fragments and drops or “tears” of CRETACEOUS CONIFERALES 7 amber, twigs, bark, cones and cone scales, seeds, leaves and leaf Impressions. Most of the logs and larger lignitic fragments were found as isolated specimens scattered through the clay strata in the Drummond pit, while the finer material and nearly all of the leaves and leaf impressions were obtained either from a bed of closely packed lig- nitic and charcoal debris, or from the accompanying clay layers, in the Androvette pit. Many of the leaves, when first uncovered, were found to be beautifully preserved, in the form of thin, brown or black carbonaceous films, but this largely disintegrated and disap- peared on exposure to the air, leaving only faint impressions to rep- resent the specimens in most instances. A large majority were of angiosperms, and as these were the first remains to attract attention, an erroneous impression was originally formed in regard to the rela- tive proportions of the angiosperms and gymnosperms in the flora. The latter were thought to be a comparatively insignificant element, and it was not until subsequently, when the lignites and lignitic debris had been examined microscopically and the amber had been analyzed chemically, that a true conception of the actual abundance of the gymnosperms was obtained; as apparently all of the amber, nearly all of the lignites and the larger part of the lignitic debris were found to belong to this latter class of vegetation. In other words, if the characters of the leaf remains alone should be consid- ered the flora would appear to have been largely angiospermous, while judged solely from the lignitic remains the gymnosperms would seem to have been the dominant class. It is probable, how- ever, that the relative proportions of each class among the leaf impressions affords the more accurate basis for conclusions in this connection, inasmuch as all leaves deposited with the fine clay would have an equal chance of preservation in the form of impressions, irrespective of texture or power of resistance to decay, whereas the wood of angiosperms as a rule yields more rapidly to decay than that of gymnosperms, and it would therefore be less likely to be pre- served in the form of lignite; and we can state as a matter of fact that of the former the few specimens found were badly decayed and unsatisfactory for study, while of the latter the specimens were not only abundantly represented by many of them were unusually well preserved and were capable of being sectioned and the internal structure determined with accuracy. 8 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN If these facts and conclusions are considered as a whole, it is prob- ably fair to assume, from the evidence which they present, that the relative proportions of the two classes of vegetation in the flora were about the same as at the present time, but that the gymnosperms included a larger number of species than now obtain. NIETHODS PURSUED IN THE EXAMINATION OF THE MATERIAL Impressions in the Clay Layers.—Nothing special need be described in regard to the methods pursued in the examination of the remains and impressions obtained from the clay layers, as this material was not different from that with which the paleobotanist ordinarily has to deal. All of the larger specimens with well defined external characters were merely drawn natural size, in order to show the general appearance which they present to the unaided eye, and to enable comparisons to be made with Cretaceous species elsewhere similarly figured. These are included on Plates 3-6. Lignitic Material—The material composed of lignitic remains, especially such as formed the mass of the finer debris, was found to require special preliminary treatment in order to prepare it for crit- ical examination, and some account of the methods pursued in its treatment and preparation has seemed to us advisable, especially as the results attained are almost entirely due to improved technique. This material, while it occurs in the form of a concentrated lignitic layer, contains a considerable admixture of clay and very fine sand. Further, even where the massed debris is almost pure lignite, it is so compacted together by pressure and chemical action that it is impossible to isolate the fragments satisfactorily by ordinary mechan- ical methods. It was ascertained, however, that maceration in 1-3 per cent. caustic alkali had the double effect of separating the frag- ments from the matrix and from each other. | such material was therefore subjected to this preliminary treatment. After maceration for two or three days in the weakest solution of alkali which would effect isolation of the fragments, the disin- tegrated material was washed carefully with runnning water, on trays made of mosquito-screen wire, during which process, in order that the weight of the mass might not break the more delicate isolated fragments, the trays were kept immersed almost to their tops. The latter precaution was found to be a highly important factor in securing CRETACEOUS CONITERALES 9 good specimens. It was also found advantageous to first screen off the coarser fragments by means of a quarter-inch mesh, thus facili- tating the subsequent examination of the finer material for valuable fragments, as the larger pieces are seldom well preserved and there- for merely serve to obscure those of greater value. The critical examination of the material was carried on by means of the dissection microscope and a lens of two- or three-inch focus. Such a lens has a large field and magnifies sufficiently to enable all the fragments of value to be readily picked out. The material thus selected has, of course, to be sectioned for the study of any internal details of diagnostic value, and, as a preliminary step, mineral matter is removed by treatment with a rather strong solution of hydro- fuoric acid. This serves to eliminate everything but iron pyrites, which has to be accepted as a necessary evil, since any reagent which will effect its removal also destroys the material. However, it is possible in most cases, on account of the relative abundance of the specimens yielded by the process of maceration described above, to select those fragments which are free from pyrites for subsequent examination. After treatment with hydrofluoric acid for two or three days, the material, after careful washing and dehydration, is transferred to successively thicker solutions of celloidin in synthol- ether, equal parts. The celloidin containing the objects is finally thickened up by the addition of chips of celloidin, until it will no longer run in the cold. The entire process of embedding is carried on in a paraffine bath, at a temperature of about 60° Centigrade, in wired bottles. When the celloidin has been sufficiently thickened the pieces of lignite are removed with forceps, care being taken to cool the bottle beforehand and to have the objects covered with a thick layer of celloidin. They are then plunged into chloroform, where they remain several hours until the celloidin has become thoroughly hardened. Finally they are transferred to a mixture of equal parts of 90 per cent. alcohol and glycerine, where they may be kept indefinitely. When the objects are to be cut, ios are attached to the object- carrier of the sliding microtome (Jung-Thoma), having previously been cemented to blocks of wood by means of thin celloidin, which is allowed to dry hard. Material prepared by the above method furnishes thin and satis- factory sections, which are well suited for photomicrographic pur- IO MEMOIRS OF THE NEW YORK BOTANICAL GARDEN poses, since the natural brown hue of the lignite serves as an admir- able natural stain. Often, even in thin sections 5 and том in thickness, which are readily secured by the above method, the hue is too intense for the resolution of fine details. In such cases the sections may be very readily bleached by the use of chlorine water, which acts very quickly and without injury to the material. Per- oxide of hydrogen was also employed for this purpose but 15 not at all satisfactory as a bleaching agent for lignites. When the sections are of the desired shade they are dehydrated by means of absolute alcohol, to which a few drops of chloroform have been added to pre- vent the softening of the celloidin matrix which holds the sections together. Thence they are transferred to benzol or xylol for clearing and are mounted in Canada balsam. It is necessary on account of the use of photographic methods of reproduction to have the sections perfectly flat. This is effected by allowing them to set under weighted cover-glasses in a warm place. Subsequently they are subjected to still further pressure by means of photographic clips, which are prevented from breaking the thin cover-glasses by slices of cork or wood, interposed between the cover and the clip. During this process they are kept in the paraffine oven to promote the further hardening of the balsam. The methods described above are somewhat laborious, but the results obtained appear to fully justify the expenditure of effort, as photomicrographs are readily obtained showing every detail of structure with great fidelity, with a small fraction of the time required for the making of drawings. In the case of material from which it was desirable to obtain an enlarged facsimile of the external features, the following method was pursued: The objects were placed in a watch-glass, the bottom of which had previously been covered with a smooth coating of black wax, flooded with alcohol and illuminated with the concentrated rays of a powerful electric lamp. Magnifications of ten or more are readily obtained in this way by the use of suitable modern lenses. The powerful light allows a stopping down of the aperture of the lenses to a sufficient degree to bring about sharpness at all depths of focus, without any material loss of brilliancy, in spite of the almost black color of the lignites. After the negative is secured, the black background, which is necessary for throwing up the sur- face details, is readily painted out by one of the numerous spotting CRETACEOUS CONIFERALES II varnishes used by photographers. A number of the smaller frag- mentary remains were photographed in this way, and from the photographs drawings were made in order to emphasize certain critical features not otherwise к These are reproduced on Plates 7-10, and photo-micrographs of the sections on Plates 11-29. The latter may be advantageously examined with a hand lens. Я mber.—Under this name we include all of the resinous material found in connection with the deposits, of which a considerable amount was collected. The larger fragments were secured during the progress of excavation, without any special search being made for them, while most of the smaller pieces were obtained as an inci- dental result of the subsequent maceration of the lignitic debris. In numerous instances, also, it was found in situ in the interstices of the lignites, as will be described further on in the discussions of certain of these remains. It is manifest that its exact origin could be determined only when found as an actual constituent of some lignite with structure sufficiently well preserved for identification. The isolated fragments may have been derived from several dif- ferent sources. They probably include both bark and wood amber and represent, so far as their origin is concerned, several different genera and species of trees, and hence ап еНоте was made, by means of chemical analysis, to determine approximately the origin of some of this material. Specimens analyzed by Dr. William J. Gies, of Columbia University, gave the results set forth in the following - report: COLUMBIA UNIVERSITY DEPARTMENT OF PHYSIOLOGICAL CHEMISTRY March 22, 1907. DR. ARTHUR HOLLICK, N. Y. Botanical Garden. Dear Dr. Hollick: The samples of American fossil resin [ from Kreischerville] submitted to me by you for chemical analysis were ound to consist of typical amber. This conclusion is justified not only by the fact that the samples closely resembled well known varıeties of amber in all of the grosser aspects, such as hardness, color, specific gravity, etc., but, also, by the observation that the differential chemical qualities of your samples and those of Succinite were essentially the same. 12 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN Of the chemical results that Dr. Meyer and I have obtained in our study of your samples, the following, taken collectively, warrant the deduction that we were dealing with true amber (Succinite) : 1. On destructive distillation, amber oil was obtained, which con- tained succinic acid. 2. When heated in a narrow tube to incipient decomposition, the fumes contained sulfur in the form of sulfid. | 3. The powdered material was more or less soluble in epichlor- hydrin, ether and alcohol. 4. Numerous determinations of elementary composition yielded the following average data: Volatile matter с Н О 5 Ash 28 77.87 9.11 12.57 0.35 0.10 0.4 Nitrogen and phosphorus were absent. These results accord with the specific data that have been accu- mulated by various observers for Succinite. Some of our quanti- tative results indicate that your samples of amber are in some respects somewhat different, though not in essential characters, from most ambers already described. This is a result of no special significance, however, for ambers are variable mixtures of resinous products and chemical differences of minor import are recognizable whenever the data for composition of different samples are compared. Very truly yours, WILLIAM J. GIES. DESCRIPTIONS OF SPECIMENS ABIETINEAE Genus PINUS Linnaeus Remains of the genus Pinus are abundantly represented, in mate- rial from both the Androvette and the Drummond pit, by wood, bark, amber, dismembered cones, cone scales, fragments of needles, and short shoots which include the basal portions of needles and the reduced axes to which they belonged. The short and fragmentary needles are especially abundant in the fine lignitic debris of the Androvette pit and are also found, to some extent, in the clay layers, but in no instance was a perfectly preserved fascicle of leaves obtained, although they clearly show the presence CRETACEOUS CONIFERALES I3 of a number of different species, some with two and others with three, four and five needles in each fascicle, and some of the frag- ments of the needles are two inches long, indicating a greater length when entire. In all cases where the basal portions of the fascicles are preserved a persistent sheath is also present, generally forming a membranous vagina, such as 1s usually found in living species belonging to the sections Pseudostrobus, Pinea and Pinaster. One, possibly two species, show a type of sheath unusual among living hard pines, consisting of loose, short bracts, like those occurring in the section Strobus, and in Pinus edulis Engelm. and related species of soft pines, but not deciduous as in these and other pines of the same group. These remains, when sufficiently well preserved, possess one fea- ture which, in general, serves to distinguish all of the Cretaceous pines thus far examined by us from those now living, and that is the very wide zone of transfusion tissue surrounding the leaf bundles. 'This transfusion tissue differs from that found in living pines in that there are few or no parenchyma cells mixed up with the transfusion elements proper. The endodermal sheath separating the transfu- sion tissue from the mesophyll is also less clearly marked than in living pines, or it may be entirely absent Cones and cone scales which could be definitely identified by their internal structure as belonging to the genus are neither numerous nor well preserved; but other specimens, from the appearance of their external characters alone, might be provisionally so referred and thus add to the number of this class of remains. Inasmuch, however, as we have found one type of scale, superficially resem- bling Pinus very closely, in which the internal structure proves it to belong to the Araucarineae, we naturally regard other similar remains as of doubtful generic relationship unless their internal structure 1s also known. Well-preserved specimens of wood or lignite were found, both in the Androvette and in the Drummond pit, the best ones in the latter. In some of these amber was found in situ, thus demonstrating the source of at least some of this substance occurring in isolated frag- ments and drops in the lignitic debris. In this connection, however, we may again remark that under the name “ amber " we include all of the resinous material found either isolated or in situ in any of the lignites, and that it was not exclusively a product of the genus Pinus. 14 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN Fragments of bark, in all probability belonging to Pinus, are among the most striking of the coniferous remains thus far brought to light. Some of the specimens from the Drummond pit, as will be more fully described further on, present the morphological char- acters of rhytidome, and superficially resemble macerated cycada- ceous cones. Following are more detailed descriptions of some of the best defined remains referable to the genus. Pinus triphylla sp. nov. Plate 3, figs. 6, 7[?]; Pl. 22, fig. 1 “ Pinus, sp. ?" Hollick, Trans. М. Y. Acad. Sci. 12: 31. pl. г. f. 20, 22. 1892. [?] “Leaf of Pinus spc.’ Jeffrey, Annals Bot. 22: 215 [descr.], 220 [reference]. pl. 14. f. 22, 24. 1908. 1 : Remains consisting of leaves and leaf fascicles. Leaves normally in fascicles of three, the lower part of each fascicle enclosed in a thin membranous sheath. It appears to be probable that there are at least two triphyllous species represented in our collections, one of which is of much larger dimensions than the other. Figures 6, 7, Pl. 3, probably belong to the larger form and correspond with the section shown in fig. 24, in the Annals of Botany, above cited. Figure 1, Pl. 22, belongs to the smaller form and is to be regarded as the type section of the species, with which fig. 22 in the Annals of Botany may also belong. Figure 6, Pl. 3, shows the general appearance, natural size, of one of the sheaths, enclosing the basal portion of a triphyllous fascicle, and fig. 7 represents a specimen consisting of a number of leaf frag- ments, also depicted natural size. It is possible that some or all of these fragments may not belong to the triphyllous species, but they were found closely associated with the specimen represented by fig. 6. These two figures are reproductions of those in the Trans- actions of the New York Academy of Sciences, above cited, repre- senting the first specimens found at Kreischerville many years ago and described as follows 15 “ Pine needles are distributed plentifully throughout, and in one specimen there is a sheath or bundle, con- taining three needles. Whether this was the normal number or not * Hollick, A. Proc. Nat. Sci. Assoc. Staten Island 1: 31. 1886. CRETACEOUS CONIFERALES 15 we, of course, cannot say, but it determines one point, namely: that the species was characterized by at least three needles in a sheath— possibly more, but not less.” Figure 1, Pl. 22, shows a transverse section of one of the fascicles, X 40, from the Androvette pit. The needles may be seen to be three in number, enclosed in the remains of the thin membranous sheath, with each needle containing a single bundle. At a higher plane of section each bundle becomes double. This type of short triphyllous fascicle is so constant in its characters, and is so abun- dantly represented in the lignitic debris of the Androvette pit, that it seems worthy of specific recognition under the name here given to it. Ву reason of the fragmentary nature of the remains, however, the fact should be borne in mind that in any specimen exhibiting the external characters of three needles enclosed in a sheath, these char- acters alone, without the corroborative evidence of internal structure, can only be regarded as indicating and not as determining its specific relationship. The positive identification of the species must always depend, primarily, upon the preservation of the specimen being such that the internal structure 15 clearly defined. Locality: Old excavation southwest of Killmeyer’s hotel, Pl. 3, figs. 6, 7. Collected by Arthur Hollick. Specimens in Mus. Staten Island Assoc. Arts and Sci. Androvette pit, Pl. 22, fig. 1. Collected by E. C. Jeffrey and Arthur Hollick. Specimen in Jeffrey collection, Cambridge, Mass. PINUS TETRAPHYLLA Jeffrey Plate 22, fig. 4 Pinus tetraphylla Jeffrey, Annals Bot. 22: 214 [descr.], 220 [name]. pl. та. f. 17. 1908. The remains of this species consist of leaves and leaf fascicles, with the leaves normally in fascicles of four. It is the least common type of short pine shoots thus far found in any of the deposits at Kreischerville. The general characters of the needles are very simi- lar to those of most others of the species of Pinus associated with it. The bundle sheath in this species is comparable to that described by Miss M. C. Stopes in the case of certain Cordaites,** and constitutes one of the numerous pieces of evidence which vouch for the primitive 1* New Phytologist 2: 91-98. pl. 9. 1903. 16 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN nature of the genus Pinus. A similar sheath is also present in P. triphylla, and in the species next described, but in the former it is much less well marked and is scarcely obvious in the low scale of magnification used in the illustration of the characters of that species, shown in fig. 1, Pl. 22. A transverse section, X 40, through a fascicle of P. tetraphylla, is shown in fig. 4, Pl. 22, in which the inner sheath of the transfusion tissue may clearly be seen to pass as a radial stripe through the four foliar traces, separating each of them into two bundles such as are universally found in the hard pines. Locality: Androvette pit. Collected by E. C. Jeffrey and Arthur Hollick. Specimen in Jeffrey collection, Cambridge, Mass. Pinus quinquefolia sp. nov. Plate 22, fig. 2 Remains consisting of leaves and leaf fascicles. Leaves normally in fascicles of five, enclosed below in a membranous sheath and with two fibrovascular bundles in each leaf. Transfusion tissue present- ing the same characters as in P. tetraphylla. The brachyblasts of this species are much stouter than those of P. triphylla. “They present the same fragmentary condition of pres- ervation, but the remains are less abundantly represented in the material from the Androvette pit. Figure 2, Pl. 22, shows the characters of P. quinquefolia as they appear in a transverse section of а fascicle, X 40. The sheath is only partially present. Locality: Androvette pit. Collected by E. C. Jeffrey and Arthur Hollick. Specimen in Jeffrey collection, Cambridge, Mass. Cone Scales of PINUS sp. Plate 9, figs. 11, 12; Pl. 23, fig. 6 A few cone scales of Pinus have been found in the macerated material from the Androvette pit. Most of these are in a very bad condition of preservation, which does not make easy a clear inter- pretation of either their internal or external features of structure. In figs. 1 1, 12, Pl. 9, are shown the upper and lower surfaces respect- ively of one typical cone scale, X 10. The inferior view of the scale shows the presence of a well marked apophysis and a median umbo. The superior aspect of the scale is so well preserved by par- CRETACEOUS CONIFERALES | 17 tial charring that the impress of the wings of the two seeds may still be made out. The superficial features of isolated cone scales of Pinus, in view of the great similarity of cone scales of not very closely related species among living pines, can scarcely have more than a generic significance, so we have refrained from giving any specific name to the type of scale depicted in our figures. It is prob- able, in view of the presence of several diverse types of short shoots in the Kreischerville deposits, that a somewhat similar type of cone scale may have characterized several different species of these pines. Figure 6, Pl. 23, shows the general features of internal structure present in a transverse section of such a cone scale, X 40. There is a single series of bundles present, with the xylem lowermost, as is the case in living pines. One of the bundles appears in the center of the figure. The fibrovascular strand further resembles those found in the ovuliferous scales of living pines, in the very sparing development of transfusion tissue. In this respect, both in living. and in Cretaceous pines, there is a marked contrast between the repro- ductive and vegetative leaves, since in both extant and extinct pines transfusion tissue is very strikingly present in the foliage, although absent or confined to the flanks of the bundles in the case of the sporophylls. Locality: Androvette pit. Collected by E. C. Jeffrey and Arthur Hollick. Specimens in Jeffrey collection, Cambridge, Mass. Bark of PINUS sp.? Plate 3, fig. 8; Pl. 22, fig. 5 Feistmantelia Ward, Nineteenth Ann. Rept. U. S. Geol. Surv., 1897- 98, Pt. П: 693. 1899. Quite common among the coarser plant remains found in the Drummond pit are pieces of coniferous bark, characterized by the scaly structure which is found in the case of rhytidomes; and as we have found similar bark in actual connection with Cretaceous pity- oxyla elsewhere we are disposed to consider it as probably belonging to some species of Pinus. This material is unquestionably identical with that to which Dr. Lester F. Ward gave the generic name of Feistmantelia, the prob- able nature and affinities of which he discusses at some length in a 3 18 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN note on pp. 694-696 (1. c.), regarding it as substantially identical with Feistmantel’s “ Portion of a stem of a coniferous plant,” and comparing it with the genera Clathraria, Omphalomela, Cycadeo- myelon, etc., of other authors. Dr. Ward's discussion should be read by all ho may be interested in learning the wide range of opinion which has been expressed in relation to this and apparently similar material. "Whatever may be thought, however, of the generic identity of our specimens with the remains included in the above mentioned genera by foreign authors, there can scarcely be any doubt that the two species of Feistmantelia from the Cretaceous of the United States!? belong in the same category with our material from Kreischerville and definitely relegate all to the Coniferales and prob- ably to the genus Pinus. Figure 8, Pl. 3, shows a fragment of one of our pieces, natural size, which superficially may be seen to somewhat resemble a dismem- bered cycadaceous cone. Figure 5, Pl. 22, gives a sufficient illustration of the internal struc- ture of such remains in transverse section, X 15. They consist of a ground substance which appears to be phloem tissues, although its state of preservation does not make it possible to speak with entire certainty on this point, interlaced with a network composed of sheets of periderm. The periderm tissues may be seen with par- ticular clearness in this figure as meandering, almost homogeneous ribbands, which, with attentive examination, may be seen to present a transverse striation corresponding to the rows of phellem cells. In longitudinal section the same sort of peridermic meshwork is present, but the meshes are somewhat more elongated. The larger frag- ments very readily break up along the lines of the zones of periderm and fall into small oval pieces, which in an isolated condition have very much the appearance of seeds, thus presenting in mass th^ resemblance to a cycad cone, as previously noted. Locality: Drummond ри. Collected by E. C. Jeffrey and Arthur Hollick. Plate 3, fig. 8, specimen in Mus. Staten Island Assoc. Arts and Sci. Plate 22, fig. 5, specimen in Jeffrey collection, Cambridge, Mass. “Мет. Geol. Surv. India, Ser. XI’ (Jurassic (Oólitic) Fl. Kach) : 61. pl. zo. f. 2. 1876. % Р. oblonga Ward, туз Ann. Rept. U. $. Geol. Surv., 1897-98, Pt. II: 693. pl. 760. f. 10. 1899. Е. virginica Font.; Ward, Monog. U. S. Geol. Surv. 48' (Status Mesoz. Fl. U. 8., second paper): 484; ibid. 48°: pl. 107. f. 3 CRETACEOUS CONIFERALES 19 Genus PREPINUS Jeffrey PREPINUS STATENENSIS Jeffrey Plate 9, figs. 9, 10; Pl. 22, fig. 3; Pl. 23, fig. 5; Pl. 24, fig. 1 Prepinus statenensis Jeffrey, Annals Bot. 22: 207-214 [descr.], 209 [name]. pl. 13. f. 1-15. 1908. The remains which represent this interesting gymnosperm consist of short shoots bearing numerous needles enclosed in a sheath com- posed of comparatively short scales or bracts. These shoots were found rather sparingly in the material from the Androvette pit. Figures 9, 10, Pl. 9, show the well-marked external features, X 10. Above the sheaths are inserted the needles, which are present in an indefinite, large number, and are not verticillate. In this feature they present a contrast to the brachyblasts of all living pines. In cross-section the needles vary from triangular to somewhat polygonal, the latter condition being present in the more internal ones.’ The appearance of these short shoots is such that at first sight one 15 tempted to consider them the ends of leafy twigs, bearing primary leaves. Their internal organization, however, clearly shows that they are really short shoots of an apparently primitive type. They may be compared with the shoots described and figured by Fontaine from the Potomac formation and referred by him to Heer’s Jurassic genus Leptostrobus.” Our specimens are much better defined, how- ever, than those indicated in Fontaine's figures, although they are somewhat more fragmentary on account of the condition of charring which is present in so much of our material. They may also be compared with Seward's Pinites Solmsi from the Wealden of Ecclesbourne.? In fig. 10 may be seen Ше rhombic scars left by the fall of certain of the basal leaves of the sheath. Usually the scales composing the sheath are all present. We have, unfortunately, not found the branches by which these interesting short shoots were borne. There can be no doubt, however, as to the close relationship of these remains with the genus Pinus. In fig. 3, Pl. 22, is shown a transverse section through a part of a 2 Monog. U. S. Geol. Surv. 15 (Potomac or Younger Mesoz. Fl.): 228; ibid. 167: pl. тог. f. 2, 3; pl 102. f. 1-4, pl. 103. f. 6-12; pl. 104. Е. 6 (=Œ L. longifolius Font.). —Ibid. 230. pl. 101. f. 4; pl. 103. f. 5; pl. 104. f. I (=L. foliosus Font.). ? Cat. Mesoz. Plants, Dept. Geol, Brit. Mus. (The Wealden Flora, Pt. II, Gymno- spermz), 196. pl. 18. f. 2. 20 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN fascicle of needles, X 10. Along the upper border of the figure appears the axis of the short shoot, showing that a considerable number of the leaves are absent. There must have been twenty or more of these in the fascicle of this species. The details of structure are not well shown in this figure, since the preservation is not entirely satisfactory. Figure 1, Pl. 24, shows a transverse section through the base of a short shoot, X 18. The wood and phloem are both well preserved in this region and surround an unusually large pith, in which are islands of sclerenchyma such as are found in certain living pines. Resin canals may be seen on the inner side of the xylem. Figure 5, Pl. 23, shows a portion of fig. 1, Pl. 24, magnified x 180. On the lower side of the figure appears a resin canal filled with tyloses, which supplies good evidence that we have here to do with the axis of a deciduous short shoot and not with a relatively main branch, since tyloses could scarcely be present in an ordinary branch showing a single annual ring. Similar tyloses have been observed in the resin canals of the bracts constituting the sheaths and also in the needles attached to the ends of the brachyblasts. Such conditions could only be normally present in short shoots which had already fallen or were about to separate from the axis. ‚ Locality: Androvette pit. Collected by E. C. Jeffrey and Arthur Hollick. Specimens in Jeftrey collection, Cambridge, Mass. Genus PITYOXYLON Kraus PITYOXYLON STATENENSE Jeftrey and Chrysler Plate 23, figs. 1-4 Pityoxylon statenense Jeffrey and Chrysler, Bot. Gaz. 42: 8. pl. г. f. 1-6; pl. 7. f. 7. 1906. The nature and character of the Pityoxyla or fossil pine woods found at Kreischerville were described in the paper above cited, and nothing further needs to be added to the general discussion of the subject at the present time. One special feature, however, calls for attention here, viz., the relation of the Pityoxyla to the amber found so abundantly in the Androvette pit. If the view of the provenance of amber taken by Conwentz?! be * Monographie der Baltischen Bernsteinbaume, p. 5. CRETACEOUS CONIFERALES 21 adopted, it follows that only the actual occurrence of the substance in organic connection with coniferous remains can constitute good evi- dence as to the origin of such amber. A great deal of this material from Kreischerville is in all probability bark amber, and constitutes the indurated remains of the resin, which once flowed in the cortical tissues of extinct conifers; but it is not generally possible to definitely connect such amber with its coniferous source. Fortunately, how- ever, in some of the Kreischerville material amber was actually found in situ in fragments of lignites, and, in all such cases which we have had the opportunity to examine, the correlated wood was found to belong to Pityoxylon statenense. It 15 not improbable, however, as has been suggested by Conwentz in the case of the Baltic amber, that the same type of Pityoxylon may have represented the wood of sev- eral different species of pines. In fig. 1, Pl. 23, is shown a succiniferous fragment in transverse section, X 10, from the Drummond pit. The amber cavity appears along the upper margin of the figure and is clothed with the same abnormal lining of parenchyma as has been described by various authors who have written on the Tertiary amber of the Baltic region. The numerous light spots over the surface of the figure indicate the presence of more or less normal resin canals, similar to those found in the wood of living species of pines. In fig. 2, Pl.-23, is shown another view of the same type of succiniferous Pityoxylon, likewise in transverse section, X 10. Theamber cavity appears in this instance in the upper left-hand region of the figure. Figure 3, Pl. 23, repre- sents the lining of a part of the amber cavity, X 20. It may be noted that there is present a thick Jacket of traumatic parenchyma, quite similar to that described by Goeppert, Conwentz and others in the case of the succiniferous woods of the Baltic deposits. Similar cavities, as is well known, occur as a pathological feature of the wood . of living pines. Figure 4, Pl. 23, shows another transverse section, X 40, of the structure of a well preserved fragment of apparently the same species of Pityoxylon. Normal resin canals of the vertical type are present in numbers, while a single horizontal canal may be distinguished. As has been pointed out in the paper on Cretaceous Pityoxyla above cited, the fossil pine woods of the Kreischerville deposits differ from living pines in the absence of the marginal tracheids from the medullary rays. Since this marked feature occurs also in the cones 22 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN of the pines of the present day, in company with other presumably archaic features, it may be safely assumed that in the case of the Pityoxyla producing the amber of the Kreischerville deposits, in so far as it is wood amber, we have to do with an ancestral feature of structure. Locality: Drummond ри. Collected by Е. С. Jeffrey and Arthur Hollick. Specimens in Jeffrey collection, Cambridge, Mass. ARAUCARINEAE Genus ANDROVETTIA gen. nov. Androvettia statenensis sp. nov. Plate 3, figs. 1-5; Pl. y, figs. 1-8; Pl. 8, figs. 1, 2; Pl. 28, figs. 5-8; Pl. 29, figs. 1-6 Remains superficially resembling fragments of fern fronds; mar- gins irregularly lobed or incised; nervation pinnate, craspedodrome, consisting of a well-defined main vein or midrib and simple or forked secondaries, which leave the midrib at acute angles, ascending or gradually 2. and for the most part bent upward and outward toward the mar Frond-like 5 28 ср of relatively thick tissues and bearing numerous stomata and minute scale-like leaves attached to the surfaces and по. male aments (?) axillary in the leaves of short flattened lateral shoots or branches which are axillary to lateral leaves of the main shoot and are covered with decussately opposite leaf pairs. The general appearance of some of the larger fragments from the clay layers may be seen in figs. 1-5, Pl. 3. These are repro- duced natural size and the striking resemblance to a fern is apparent; in fact they probably would have been so regarded had the internal structure not been determined. Interesting comparisons may also be made with figures of certain remains of more or less problematic relationship from the Potomac formation. Taking everything into consideration these remains constitute the most remarkable that we have obtained from the Kreischerville deposits, presenting as they do a strong superficial resemblance to ferns and to the modern podocarpous genus Phyllocladus, and at the same time to the Cre- taceous genera, Thinnfeldia, Moriconia and Thuites. Their sub- о insignis Font.?, Ward, Monog. U. $. Geol. Surv. 48? (Status Mesoz. Fl. U. $): 12. ; Zamiopsis insignis Font., ibid. pl. 773. f. 4, 5; Thinnfeldia mary- landica E я "d 114. f. 8,9; 2. marylandicum Font., ibid. pl. 776. hr Кален marylandica Font., ibid. pl. 777. f. 7. CRETACEOUS CONIFERALES 23 stance was evidently quite thick and apparently coriaceous in texture, as it is almost invariably well preserved and may be removed from the clay or separated from the other debris in the form of dark brown or yellowish carbonaceous films. Figure 1, Pl. 7, shows one of the phylloclads of this species, X 7, which has been bleached by prolonged sojourn in chlorine water, so as to show at the same time its external appearance and internal venation. The margins present dentate projections, which in some cases are leaves and in others very much reduced leafy branches. A central stouter vein runs through the phylloclad from end to end, giving off vascular strands to the right and left. In two instances leaves may be seen attached to the flat surface, a feature which clearly distinguishes our species from Thinnfeldia on one hand and from most of the living species of Phyllocladus on the other. Figure 2, Pl. 7, shows the base of a phylloclad, X то, and fig. 3, Pl. 7, shows a somewhat larger specimen, X 6, in which one of the leaves on the flat surface of the phylloclad is very well developed. Another leaf, which is less obvious, appears on the left at a slightly higher level. It is a notable fact that by turning over these phylloclads one can observe on the opposite flat surface leaves corresponding in position, 7. e., opposite to those appearing in the figures. A rather less marked opposition of the leaves may be made out on the margins. Figure 4, Pl. 7, shows a phylloclad, or, more accurately speaking, a part of a phyllo- clad, similar to those described above, X 9, which bears on its sides two leafy flattened branches resembling the habit of Moriconia. These are axillary to lateral leaves of the main shoot, as may be distinguished on the uninjured left side of the phylloclad. The Moriconia-like lateral branches are covered with decussately opposite leaf pairs, which resemble those of the living Thuja and the fossil Moriconia, both in their disposition and in the strong flattening of the leafy axis. They also resemble the ultimate branches of Phyllo- cladus trichomanioides Don. In fig. 5, Pl. 7, is shown another ` phylloclad, X 8, with a leafy shoot still attached to its side, which bears in the axils of its lateral leaves what appear to be immature male cones or aments. | Figure ı, Pl. 8, shows a fragment of what was evidently a large specimen, X 7, and fig. 2, Pl. 8, a similar one, X 8. In both of these figures leaves may be seen attached to the flat surface of the phylloclads in the manner characteristic of the species. Figure 2 24 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN presents somewhat intermediate conditions between the phylloclad and the flattened leafy lateral branch, for it is at once markedly phylloclad-like and at the same time shows several leaves attached to the flat surface. Figures 6, 7, 8, Pl. 7, show the characters of the flattened lateral leafy branches, X 7, as they appear when detached from their axes. This remarkable genus is apparently not referable to any fossil plant heretofore described and, moreover, entirely differs in its internal structure, as we shall show presently, from the living genus Phyllocladus, so that it is not permissible either on the one hand to refer it to the somewhat vague Cretaceous genus Thinn- feldia, or on the other to the equally problematic Protophyllocladus. Turning our attention now to the internal structure of this remark- able fossil plant, we have in fig. 5, Pl. 28, a general view of a trans- verse section through one of the phylloclads, X то. There is evi- dence in the form of woody cylinders of the presence of three hranches in the substance of the flat axis. The most central of these is the largest, while the uppermost is the smallest. The narrow wings of tissue stretching on either side of the woody cylinders con- stitutes the lamina of the phylloclad, in which lie the traces of two or three leaves. The magnification, however, is not great enough to show these. It is not easy to recognize any assimilatory tissue in the lamina on account of the very dark conditions of the cells which compose it. Regarded superficially the lamina is covered with numerous stomata, which are surrounded, in addition to the guard cells, by four or five accessory cells. The presence of these accessory cells does not appear to be of much importance in tracing relation- ships with other conifers in which the surface characters have been described, since the presence of accessory cells is a very common feature in plants of various afhnities, which have the xerophytic habit. Figure 8, Pl. 28, shows a transverse section, X 40, through the base of one of the phylloclads. There is a single woody cylinder present in this case, which is winged above and below. We have not found any stems belonging to the species which are without these laminar wings, even in their lowest portion. In the pith may be seen sclerotic cells, similar to those occurring in Brachyphyllum and Geinitzia. We have not observed sclerotic tissues in the cortex of the species under discussion. Leaf-traces may be seen on either side of the cen- tral cylinder and in the substance of the lateral wings. Figure 7, Pl. 28, shows a transverse section, X 20, through the blade of another CRETACEOUS CONIFERALES 25 phylloclad, somewhat larger in size than that shown in fig. s. There are also threé woody cylinders present, the central one of which is again the largest. Figure 6, Pl. 28, shows, in transverse section, х 40, one of the leaves on the edge of the same phylloclad as that shown in fig. 7, in the region where it becomes free from the surface of the flattened axis. Figure т, Pl. 29, shows a transverse section, X 30, of the largest stem in our possession. Here, as in fig. 8, Pl. 28, traces of foliar strands may be seen in the lateral wings. Figure 2, Pl. 29, shows the structure, in transverse section, X 45, of the central cylinder in fig. 7, Pl. 28. The wood is very similar in general appearance in cross- section to that found in the case of Brachyphyllum, Geinitzia, etc. Figure 3, Pl. 29, shows the structure of part of the woody cylinder in fig. 1, Pl. 29, X 180. Itmay readily be seen that the resin cells, which as a rule are characteristic of the Podocarpineae, to which the living Phyllocladus belongs, are absent. Figure 4, Pl. 29, shows the appearance, in longitudinal approximately radial section, X 180, of the woody cylinder in the same specimen, which furnished fig. 8, Pl. 28. It may be noted that many of the pits are flattened by mutual contact in the manner characteristic of araucarineous woods, but that this feature is not universal, so that the wood is of the type which we propose to describe under a new generic name (see this Memoir, p. 54). Figure 5, Pl. 29, shows a longitudinal section, x 180, of the branch which furnished figs. 1 and 3, Pl. 29. The same type of wood is present. Figure 6, Pl. 29, shows a small part of the extreme right-hand portion of fig. 5, Pl. 29, X 500, to illus- trate the very typical araucarineous pitting. It may be seen from the details of external and internal structure that we have in this species a very peculiar type of conifer which presents a marked similarity in superficial habit to the podocarpi- neous genus Phyllocladus, but difters strikingly from the genus in important features of internal structure. There can be no question that in this case we have to do with one of the many peculiar arau- carineous types which we find to be so characteristic of the Kreischer- ville flora. If Professor Seward and others are right in their con- tention that the genus Thinnfeldia is of filicineous affinities, then our genus, which is beyond question coniferous, has nothing to do with Thinnfeldia, with which certain somewhat similar remains from the Cretaceous have been considered to belong. Our species, more- 26 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN ever, presents certain differences even in external appearance, notable among which is the presence of leaves on the flat surface of the lamina of the phylloclads, which make it impossible to refer it to Protophyllocladus Berry,** even if the internal structure of the fossils in question did not remove them from any close relationship with the Podocarpineae of the present day. It seems for these reasons necessary to propose a new generic name for these remains, and as we owe much to the kindness and courtesy of the owner of the Androvette pit, it seems a not unfitting recognition to name the new genus in his honor. The specific name is given from its place of discovery, Staten Island. Locality: Androvette pit. Collected by E. C. Jeffrey and Arthur Hollick. Plate 3, figs. 1-5, specimens in Mus. Staten Island Assoc. Arts and Sci. Plate 7, figs. 1-8; Pl. 8, figs. 1, 2; Pl. 28, figs. 5-8; Pl. 29, figs. 1-6, specimens in Jeffrey collection, Cambridge, Mass. Genus RARITANIA gen. nov. Raritania gracilis (Newberry) comb. nov. Plate 6, figs. 4-7; Pl. 9, figs. 1-4; Pl. то, figs. 14-17; Pl. 19, figs. 3-6; Pl. 20, fig. 1 Frenelopsis gracilis Newb. Monog. U. S. Geol. Surv. 26 (Fl. Amboy Clays): 59. pl. 12. f. 1-3a. 1895. Our specimens, depicted in figs. 4—7, Pl. 6, are undoubtedly spe- cifically identical with those described under the above name by Newberry from the Cretaceous of New Jersey; but he was clearly at fault in referring the species to the genus' Frenelopsis Schenk, which is characterized by articulated branches with opposite leaves,24 whereas in his specimens, as well as in ours, the branches are not articulated and the leaves are as Newberry describes them, “ spirally arranged." They are so minute, however, that they are not visible to the unaided eye, and their exact arrangement is difficult to deter- mine satisfactorily from impressions alone, even with the aid of a hand lens. The dichotomously forked branches, together with the minute prickle-like foliar organs, give a superficial appearance almost ? Bull. Torrey Club 30: 440. “Е. Hoheneggeri (Ettingsh.) Schenk, Paleontogr. 19': 13. pl. 4. f. 5-7; pl. 5. f. 2, 2; pl. 6. f. 1-6; pl. 7. f. 1. 1871. (Culmites priscus Ettingsh.?, Abh. K. K. Geol. Reichsanst. 1%: No. 2, p. 24. pl. 1. f. 5; pL 3. f. 4-8; Thuites Hoheneggeri Ettingsh., ibid. 26. pl. r. f. 6,7. 1851.) CRETACEOUS CONIFERALES 27 exactly like that of the living pteridophyte genus Psilotum, and it has seemed somewhat strange that Newberry did not recognize this resemblance, especially as he had previously referred another plant with similar habit to the genus.?* Figures 4-6, Pl. 6, show the general form and appearance of the species, natural size, and fig. 7, on the same plate, represents a por- tion of one of the ultimate forking branches, enlarged in order to show the presence of the foliar organs and their appearance under an ordinary hand lens. In consideration of these facts it is evident that the species cannot properly be included in the genus Frenelopsis, and, as we shall indi- cate further on, the internal structure shows that it is a conifer and not a pteridophyte, we have adopted for it a new generic name, Raritania, in order to correlate it with the geological formation of which it is characteristic. It may be pertinent to remark, however, that we expect to demonstrate, in a subsequent memoir, the proba- bility that certain other American Cretaceous plants have been prop- erly referred to the genus Frenelopsis, although these aré in general gnetold remains, resembling the living genus Ephedra, and con- sequently not related to the coniferous genus Frenela. The determination of the correct botanical affinities of Raritania, in view of the confusion of names and alliances above indicated, was therefore recognized to be a matter of considerable importance, and diligent search was made for specimens which could be subjected to critical examination. The result was the separation of a number of fragmentary leafy twigs from the lignitic debris which almost cer- tainly belong to the genus and in all probability to the particular species under discussion. Figures 1-4, Pl. 9, represent external views of several of these twigs, X 10, and figs. 14—16, Pl. ro, are views, X 5, of two of the same specimens and one of a cone, attached to a peduncle which has the same type of leaf as the twigs. Unfortunately this cone is very poorly preserved, and it has not been possible to learn anything from its internal structure, either as to its exact botanical affinities or, in the case of the peduncle, as to its rela- tion to the twigs associated with it. Incidentally it may be remarked 5 Psilotum inerme Newb. Ann. N. Y. Lyc. Nat. Hist. 9: 38. 1868. (= Cabomba inermis (Newb.) Hollick, in Newb. Monog. U. 8. Geol. Surv. 35 (Later Ext. Fl. N. Amer.) : 92. pl. 22. f. 2; pl. 23. f. 2. 1898.) 28 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN that this cone has much the appearance of those which have been figured in connection with branches of Inolepis imbricata Heer,?* from the Cretaceous of Greenland, but the leaves do not agree, while the external resemblance between the peduncle and the twigs in our specimens is so strong that there can scarcely be any doubt that these are generically if not specifically identical with each other. Figure 17, Pl. 10, shows the same cone, X 10, and there is enough of the surface sculpturing still remaining to indicate that the scales were spirally arranged and were not, externally at least, divided into two segments, as in the case of the Abietineae. From the preceding discussion it may be inferred that the results obtained from the studies of the external characters of the specimens were not entirely conclusive either as to their identity with the genus Raritania or as to their exact botanical relationship; but, as may be learned from the discussion which follows, the internal structure was found to afford a much more satisfactory field of investigation. Figure 3, Pl. 19, shows the internal structure, in transverse section, X 30, of the twig shown in fig. 3, Pl. 9. The woody cylinder is immature and is badly preserved. The cortex is confluent with the decurrent leaf bases and is characterized by the presence of a consid- erable number of stone cells, which have mostly disappeared in charred specimens like that under discussion, but are more apparent in badly preserved material. There are indications of four leaves on the periphery of the section, each of which shows a more or less apparent resin canal, which exactly subtends the foliar traces. Figure 4, Pl. 19, shows a transverse section, X 45, of the stouter twig shown in fig. 1, Pl. 9. An attached leaf may be seen on the upper side. It contains a resin canal which communicates with the surface of the stem by means of a crack, a result of the condition of fossilization. On the right is a resin canal in the cortical tissues which marks the position of the decurrent portion of another leaf. On the upper side of the section the tissues of the phloem are partially preserved and these strongly resemble those found 1 Geinitzia (see this Memoir, pp. 40-42). The woody cylinder is 15 remarkably well preserved in this specimen and presents the same general features of microscopic structure as in Brachyphyllum (see this Memoir, p. 33); but the pith is relatively smaller than in either of the other two araucarineous types mentioned. Figure “FI. Foss. Arct. 3 (Kreide-Fl.) : 72. pl. 16. f. 12-16; pl. 23. f. бе. CRETACEOUS CONIFERALES 29 5, Pl. 19, is a view of a longitudinal section, X 20, of a rather badly preserved twig, illustrating the transverse cracks, which are a fre- quent feature of fossilization in these and other similarly preserved specimens, which feature, as previously suggested by the senior author," led Newberry into the error of referring certain partly decorticated twigs of /’iddringtonites to Frenelopsis,8 on account of the resemblance of these cracks to joints or articulations. Figure 6, Pl. 19, shows a longitudinal section, X 20, passing through the broken apex of a leaf on the upper surface of another specimen. Ву the use of a lens sclerites can be made out in the cortex of the upper surface and in the pith (nearly opposite the termination of the leaf). Figure 1, Pl. 20, is a longitudinal section of the wood, X 500, showing the radial pitting. The pits are fewer in this species than in any other which we have examined, on account no doubt of its extremely xerophytic habit and it is not easy in all sections to find the araucarineous type of pitting. This can, however, generally be observed in the ends of certain of the tracheids, as 15 shown in the figure last mentioned, in which the pits may be seen to be obviously flattened by mutual pressure in the typical araucarian fashion. We therefore conclude, from the internal structure of the twigs of these remains, that there is no reasonable doubt as to their araucarineous affinities, and if we have correctly identified them as the remains of Raritania this determines the correct systematic posi- tion of the genus. Locality: Androvette pit. Collected by E. C. Jeffrey and Arthur Hollick. Plate 6, figs. 4-7, specimens in Mus. N. Y. Bot. Gard. Plate 9, figs. 1-4; Pl. 10, figs. 14-17; Pl. 19, figs. 3-6; Pl. 20, fig. 1, specimens in Jeffrey collection, Cambridge, Mass. Genus WIDDRINGTONITES Endlicher WIDDRINGTONITES Кетсни (Ettingshausen) Heer Plate 5, figs. 1-4; Pl. 8, figs. 7-11; Pl. 20, figs. 3-5 Widdringtonites Reichii (Ettingsh.) Heer, Fl. Foss. Arct. 6%: 52. pl. 28. f. 5. 1882.—Hollick, Monog. U. S. Geol. Surv. 50 (Cret. Fl. S. М. У. апа N. Eng.) : 44. pl. 4. f. 6, 7. 1906. * Bull. N. Y. Bot. Gard. 3: 411. 19 3 Monog. U. S. Geol. Surv. 26 (Fl. pu Clays) : 58. pl. z2. f. 4, 5 (Frenelopsis Hoheneggeri (Ettingsh.) Schenk?). 30 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN Frenelites Rechii Ettingsh. Kreidefl. Niederschóna 246. pl. г. f. 104- roc. 1867.—Hollick, Trans. М. Y. Acad. Sci. 12: 29. pl. I. f. 23. 1892. This is apparently the most abundantly represented of any conif- erous species thus far found in the Kreischerville deposits. The remains occur in almost every layer of the plant-bearing beds in the Androvette pit, and relatively large masses of the delicate branchlets, generally more or less matted together, can often be obtained, although well preserved specimens are difficult to isolate from the lignitic debris. The exact botanical relationship of the genus has never been satis- factorily determined by any of the authorities who have studied and described it, although affinity with Frenela and with JJ'iddringtonia is inferred by Ettingshausen and Heer respectively, in the generic names which they adopted. Specimens are represented natural size by figs. 1-3, Pl. 5. Figure 3 is that of a partly decorticated branch, with some of the smaller leafy branchlets attached, and fig. 4, on the same plate, shows a portion of a branchlet as it appears when enlarged under a hand lens, showing the phyllotaxy. Figures 7-11, Pl. 8, represent views of the external features of small twigs, X 10, which seem to be referable to this species. These figures resemble very strongly most of the published illustrations of Jl'iddringtonites, and since some of our twigs show internal structure it is possible to determine their true botanical relationship. Figure 3, Pl. 20, is a transverse section, X 40, of the specimen shown infig. 8, Pl. 8. Several of the leaves are included in the figure at dif- ferent planes of section. Оп the upper side of the figure one may be seen cut through its region of greatest expansion. On the opposite side a leaf appears as cut through its apex. On the sides of the figure are shown the bases of three other leaves. The xylem cylinder is small and contains a rather narrow pith, in which sclerotic elements are sometimes present, although none of these appears in the plane of the section. Figure 4, Pl. 20, shows a portion of the upper margin of fig. 3, Х 60. It is possible to make out here with some distinct- ness that a single resin canal is present in the leaf. Near it lies a fibrovascular bundle which is accompanied by dorsally directed borders of transfusion tissue. Figure 5, Pl. 20, shows the radial CRETACEOUS CONIFERALES 3I structure of the wood in longitudinal section, X 500. The pitting, when fully established, as it is about the middle of the figure, is obviously araucarineous. If we are right in correlating the speci- mens with structure preserved, shown in our figs. 3- 5, Pl. 20, with Widdringtonites Reichii, it follows that this species, in spite of its external resemblance to the section /Viddringtonia of the cupres- sineous genus Callitris, is in reality another of the many araucari- neous conifers which flourished during that portion of the Cretaceous period when the Kreischerville clays were deposited. If these con- clusions are justified by the facts, as they appear to be, it is a further proof that external appearances alone constitute a very unsafe guide in the matter of inferring the affinities even of late Mesozoic conifers with those now living. That this view of the matter is reinforced not only by the study of the internal structure of the twigs, but also from a similar study of certain cone scales associated with them, may be appreciated by reference to the facts in connection with remains of this kind elsewhere described in this Memoir. Locality: Androvette pit. Plate 5, figs. 1-4, collected by Arthur Hollick. Specimens in Mus. Staten Island Assoc. Arts and Sci. Plate 8, figs. 7-11; Pl. 20, figs. 3-5, collected by E. C. Jeftrey and Arthur Hollick. Specimens in Jeffrey collection, Cambridge, Mass. Genus THUITES Sternberg THUITES sp.? Plate 8, figs. 12-18; Pl. 27, figs. 4-6; Pl. 28, figs. 1-4 The new genus Androvettia is not the only araucarian with verti- cillate leaves which has come under our notice from the Androvette deposits. In figs. 12-17, Pl. 8, are shown the surfaces of three twigs, X 10, in each case represented from the two opposite sides, which have the appearance and the whorled leaves characteristic of the genus Thuites as commonly understood, and of the specimen figured by Newberry from the Cretaceous of South Amboy, N. J., under the name Thuya cretacea (Heer) .29 In fig. 2, Pl. 28, the base of the specimen shown in figs. 14, 15, Pl. 8, is shown in transverse section, X 30. Two opposite leaves may be seen, one on the right and the other on the left of the figure. The plane of section in the opposite direction is not quite accurately trans- ® Monog. U. $. Geol. Surv. 26 (Fl. Amboy Clays): 53. pl. zo. f. 1, Za. 32 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN verse, so that the base of only one of the leaves of the decussate pair can be made out. Figure 3, Pl. 28, shows a similar section of the same specimen, X 50, in which the upper and lower leaves are those best developed in the plane of section. In both of these figures there are present abnormal cavities in the substance of the leaves, the exact nature of which cannot be made out. Figure 4, Pl. 28, represents the central portion of fig. 3, X 100. It may be noticed that the elements of the fibrovascular cylinder are very immature and somewhat disturbed, as a result of fossiliza- tion. It has been found difficult to secure ап old enough twig of this species to determine the structure of the secondary wood, although the structure of the leaves makes it very probable that we have here to do with an araucarineous conifer, since the distribution of the transfusion tissue is not lateral to the foliar strands as in Thuja and the other Cupressineae, but has the same tendency to surround the phloem of the bundle which has been noted in the other araucarineous twigs previously described. However we are not left entirely to surmise in this case, for in fig. 18, Pl. 8, is shown a young twig, X Io, with a single pair of verticillate leaves, which is sufh- ciently mature to make it possible to determine the structure of its wood as araucarineous. In fig. 4, Pl. 27, is shown a transverse section of this fragment, X 35. On the two surfaces of the figure may be seen furrows, which mark the line of separation of the two leaves. In the center appears the fibrovascular cylinder, in which the wood is well developed. Subtending the ends of the woody cylinder and opposite clear interruptions in its continuity are the traces of the leaves. Figure 5, Pl. 27, represents a transverse section, X 35, of the opposite and upper end of the same fragment near the level of separation of the tips of the leaves from the surface of the stem. On the lower side of this figure one of the leaf traces may clearly be seen as a dark spot in the substance of the leaf. At this level the traces of the decussate pair of leaves are leaving the central cylinder, as may be perceived by noting the corresponding two deep opposite bays in the woody cylinder. The central cylinder is sur- rounded as a whole by a dark boundary, which appears with par- ticular clearness in this figure. ‘The nature of this boundary has not been easy to decipher. It is not possible to decide whether it is merely a zone of collapsed sieve tubes, such as often occurs in the conifers, and which we have found in the case of Brachyphyllum, or CRETACEOUS CONIFERALES 33 of a sclerenchymatous sheath. On the whole the former supposition appears more probable. Figure 6, Pl. 27, shows a portion of the section shown in fig. 4, Pl. 27, Х 45. The very small wood elements can here be seen somewhat more clearly. There are no resiniferous elements present in the wood of this species and in this respect it resembles most of the Araucarineae described elsewhere in this Memoir. A longitudinal section of the wood, X 500, such as is shown in fig. 1, Pl. 28, makes it clear that here too we have to do with an araucarineous conifer of the same general type as Brachy- phyllum, for the characteristic type of pitting in the wood is unmis- takably present. Іп all of the examples of conifers with verticillate foliage and araucarineous internal structure, which we have found in the Kreischerville deposits, cortical resin canals are universally absent and in this feature as well as in their verticillate foliage they markedly resemble each other. Since our genus .Indrovettia is cer- tainly an araucarian, and since the same statement is true of the last described fragment and is probably true of the three similar twigs described above, which we have provisionally referred to the genus Thuites, it seems likely that other Cretaceous coniferous genera with verticillate leaves, viz., Moriconia, Inolepis, etc., are probably araucarineous as well. This probability can, of course, only become a certainty when it becomes possible to investigate the internal structure of the last mentioned genera. Locality: Androvette pit. Collected by E. C. Jeffrey and Arthur Hollick. Specimens in Jeffrey collection, Cambridge, Mass. Genus BRACHYPHYLLUM Brongniart BRACHYPHYLLUM MACROCARPUM Newberry Plate 4, figs. 12-14; Pl. 9, figs. 7, 8; Pl. 11, figs. 1, 2, 4, 5; Pl. x2, figs. 1-6; Pl. 13, fig. 1 Brachyphyllum тасгосагрит Newb. Monog. U. $. Geol. Surv. 26 (Fl. Amboy Clays) : 51, footnote. 1895.—Hollick and Jeffrey, Amer. Nat. 40: 201-203. pl. 3. f. 1-5; pl. 4. f. 1-6; pl. 5. f. 1-4. 1906. Thuites crassus Lesquereux, Rept. U. S. Geol. Surv. Terr. 8 (Cret. and Tert. Fl.) :32. 1883. Brachyphyllum crassum (Lesq.) Lesq. Proc. U. S. Nat. Mus. то: 34. 1887.—Newb. Monog. U. S. Geol. Surv. 26 (Fl. Amboy 4 34 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN Clays): 51. pl. 7. f. 1-7. 1895. Not B. crassum Tennison- Woods, Proc. Linn. Soc. New South Wales 7: 660. 1883. There can be no question in regard to the identity of our specimens with the species described and figured by Newberry from the Cre- taceous of New Jersey, or in regard to their affinities with the Araucarineae, as we have elsewhere shown;*" but the determination of these facts does not mean that all of the coniferous remains which from time to time have been referred by various writers to the genus Brachyphyllum necessarily belong in it, or that all are necessarily araucarineous in their affinities. Our determinations, thus far, must be regarded as limited to the particular species under discussion. Since the publication of our preliminary paper in the American Naturalist (1. c.), however, we have had the opportunity to examine a great deal of additional material, some of it in an admirable condition of preservation, and as a consequence are able to give a much fuller account than that which first appeared. Detached leaves and fragments of twigs are abundant elements in the lignitic debris of the Androvette pit, but only one large specimen, obtained from the Drummond pit, has thus far been found. This latter is represented, natural size, in fig. 12, Pl. 4. It 15 partly decorticated, but shows the characteristic method of branching and the club-like shape of the ultimate twigs. Figures 13, 14, Pl. 4, and figs. 7, 8, Pl. 9, represent termini and lower portions of twigs, Ж то, showing the phyllotaxy and the closely imbricated, striated, scale-like leaves. It is also probable that some of the coarser lignite fragments with araucarineous structure may be referable to this species, as may be found discussed under the several types of such wood further on. Figure 1, Pl. 11, presents a general view in transverse section, Ж 11, Of the structure of a well preserved although somewhat flat- tened stem of B. macrocarpum. In the center may be seen the woody cylinder enclosing a rather large pith, which even with the low power of magnification used is seen to be occupied by masses .of sclerenchyma. Surrounding the fibrovascular cylinder is a cordon of resin canals, which appear as holes in the photomicrograph. External to these on the upper side of the figure is a dark line which marks a zone of periderm deep under the surface of the stem. Still ® Amer. Nat. 40: 189-215. pls. 1-5. 1906. CRETACEOUS CONIFERALES 35 further out are other resin canals belonging to the adnate leaves. The periphery of the section has a somewhat mottled appearance, which is due to the presence of strands of sclerenchyma just underneath the epidermis. The fact that the leaves are free from the surface of the stem merely by a narrow margin may be seen by following the outline of the section. Figure 2, Pl. 11, shows a similar section, X 28, of a much younger and smaller stem than that which appears in fig. 1. On the margin of the section may be seen the indications of three leaves. The distribution of resin canals is the same as in fig. 1, but is less easily made out. The central cylinder is so small that it can scarcely be seen without magnification by means of a lens. Figure 4, Pl. 11, is the central cylinder and immediately surround- ing tissues, X 15, of the section shown in fig. 1. Figure 5, Pl. 11, shows the central cylinder of the same specimen in longitudinal sec- tion, X 15. In this view the fact that the sclerenchyma occurs in nests may clearly be made out. Figure 1, Pl. 12, is a part of the section shown in fig. 4, Pl. II, X 40. In the upper part of the field may be seen a mass of cortical sclerenchyma. Below this the cortex is occupied by two resin canals. Inferior to the line joining these lies the phloem, which, as was pointed out in the preliminary article, contains none of the bast fibers characteristic of the Sequoiineae, with which Brachyphyllum has been placed by most authors, on account of its general external appearance. The phloem terminates above in a discontinuous dark line, which is constituted by masses of col- lapsed sieve-tubes similar to those which occur in many living con- ifers. Below the phloem is the xylem, which presents the illusory appearance of two annual rings, due to inequality in growth. The inner part of the xylem is very badly preserved and only occasionally shows the presence of the protoxylem clusters, which exhibit, how- ever, no features of special interest. Internal to the wood is the pith, consisting of thin-walled parenchyma and islands of scleren- chyma, one of which appears in the field. Figure 2, Pl. 12, is a view of another part of the section shown in fig. 4, Pl. 11, X 40. On the upper left-hand may be seen the layer of deep periderm which has been mentioned above. On the extreme left, just outside the phloem, a leaf trace may be seen on its way outward. It has not: yet undergone the divisions which characterize it in its more external' course. Figure 3, Pl. 12, is a portion of the same section, X 180,, 36 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN showing the wall of the central cylinder. Тһе xylem may be seen to be made up of tracheids, without any admixture of parenchyma- tous elements. External to it is the phloem, separated from it by a slight interval, due to decay. Figure 4, Pl. 12, is a similar section, X 40, in which may be seen a forking leaf trace. In Brachyphyllum the foliar trace divides in the outer cortex or in the mesophyll of the leaf into a large number of very fine strands, which finally become lost in a continuous band of transfusion tissue which occurs under- neath the palisade parenchyma of the leaf. Figure 6, Pl. 12, shows this transfusion tissue in transverse section, X 180. It appears as a stripe of tracheary cells running across the center of the field from side to side. Figure с, Pl. 12, illustrates the character of the tissues in the vicinity of the Е. as they appear in transverse section, x 40. The dark masses just beneath the epidermis are the strands of fibrous tissue which are partly responsible for the striated appear- ance of the leaves of Brachyphyllum in well-preserved impressions. The striations stand out most clearly in partially decayed material, which throws the sclerotic ridges into stronger relief. "There is, however, a finer striation, which appears even in well-carbonized material, due to the presence of rows of depressed stomata. Figure 1, Pl. 13, shows the structure of the central cylinder of a very young branch of Brachyphyllum macrocarpum, in transverse section, X 40. Severalfibrovascular bundles are present which have not as yet been united with each other by the formation of a con- tinuous woody cylinder such as is found in the older stem. On the right of the figure may be seen a leaf trace making its way outwards into the cortex. It subtends a clearly marked gap in the central cylinder. The anatomical fact represented by this figure makes it difficult to accept the hypothesis that the araucarineous conifers have a close relation to the Lycopodineae, an hypothesis recently revived with some force of argument by Seward,3! but it has been pointed out by the junior author?? that none of the Lycopodineae, living or extinct, have foliar gaps, while in the Araucarineae, both living and extinct, foliar gaps, so far as our knowledge goes, are always present. In our investigations we have had the opportunity to examine this feature in a number of hitherto structurally unknown Araucarians, which show every indication of representing an older type than those "Philos. Trans. Roy. Soc. London В 198: 305-411. pls. 23, 24. 1906. “Philos. Trans. Roy. Soc. London В 195: 119-146. pls. 1-6. 1902. CRETACEOUS CONIFERALES 37 now living. Foliar gaps have been observed to be invariably present in all these cases. This is one of the many features which make it very difficult to accept the hypothesis of the lycopodineous origin of the Coniferales, and other equally cogent reasons for rejecting this hypothesis will be discussed further on. Locality: Drummond pit. Plate 4, fig. 12, collected by John M. Dunnigan. Specimen in Mus. Staten Island Assoc. Arts and Sci. Androvette pit.. Plate 4, figs. 13, 14; Pl. 9, figs. 7, 8; Pl. 11, figs. 1, 2, 4, 5; Pl. 12, figs. 1-6; Pl. 13, fig. 1, collected by E. C. Jeffrey and Arthur Hollick. Specimens in Jeffrey collection, Cam- bridge, Mass. Cone of BRACHYPHYLLUM sp.? Plate 9, figs. 5, 6; Pl. 11, fig. 3; Pl. 14, fig. 3 A single fragmentary cone, X 10, viewed from opposite sides, is shown in figs. 5, 6, Pl. 9. It was apparently ellipsoidal in shape. The upper part is only represented by the denuded axis, somewhat flattened, which was doubtless originally covered with scales simi- lar to those yet attached to the lower part. These scales are strik- ingly like the leaves which cover the branches of Brachyphyllum and the relationship of the cone to that genus would naturally be inferred from these external characters alone, although it may be seen that the scales are somewhat less broad than the leaves on the shoots, represented by figs. 7, 8, on the same plate, and they may not belong to the same species. Figure 3, Pl. 11, shows a transverse section, X то, through the peduncle or lower part of this cone, and the general features may be seen to be almost identical with those of the stem of Brachy- phyllum macrocarpum shown in figs. 1 and 2, on the same plate, thus making probable at least the generic relationship. Figure 3, Pl. 14, shows a transverse section, X 35, through the axis. On the left side of this figure may be seen a break in the sur- face of the cone. The cortical tissues of the axis are occupied by very numerous resin canals. Inside the resin canals lie very small fibrovascular strands, which can scarcely be seen with the degree of magnification used in this case. The center of the axis is occupied by a large pith which is free from stone cells, although these are abundant in the pith of the peduncle, as may be made out in fig. 3, PL I1. 38 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN It is unfortunate that in no instance have we found the cones and branches attached, but other investigators have also suffered the same disappointment. Thus Newberry remarks :33 “ Unfortunately, none of the specimens establish beyond all doubt the connection between the cones and the branches, but some of the cones are borne on pedicels which are marked with scales essentially like those of the branches under consideration | В. macrocarpum]." The cones figured by Newberry (1. c., pl. 7, figs. 3, 4, 6) are too imperfect for com- parison with ours, but his fig. 3 has a striking resemblance to the cone described and figured by Velenovsky under the name Pinus pro- topicea,?* from the Cretaceous deposits of Bohemia, where Brachy- phyllum likewise occurs, and the latter author also figures a basal por- tion of one of these cones (1. c., fig. 12) showing scales somewhat suggestive of ours; but all of the above mentioned figures represent specimens which are much larger. A cone found in the Drummond pit, depicted in fig. 10, Pl. 3, and described on p. 68 of this Memoir, should be incidentally mentioned here, as may be understood by reference to the discussion in connection with it. Locality: Androvette pit. Collected by E. C. Jeffrey and Arthur Hollick. Specimens in Jeffrey collection, Cambridge, Mass. Genus GEINITZIA Endlicher Geinitzia Reichenbachi (Geinitz) comb. nov. Plate 5, figs. 7-10; Pl. 8, figs. 3, 4; Pl. 16, figs. 2-4; Pl. 17, figs. 1-4; Pl. 18, figs. 1-4 Ятаисатиез Reichenbachi Gein. Charakter. Schichten u. Petref. Sächs.-Böhm. Kreidegeb. 3: 98. pl. 24. f. 4. 1842. Geinitzia cretacea Endl. Synops. Conif. 281. 1847. Sequoia Reichenbachi (Gein.) Heer, Fl. Foss. Arct. 1: 83. pl. 43. f. Id, 2b, 5а, 5d, 544, 8, 8b. 1868.—Hollick, Monog. U. S. Geol. Surv. 50 (Cret. Fl. 5. N. Y. and N. Eng.) : 42. pl. 3. f. 4, 5 1906. “Sequoia Couttsiz, Heer.” Hollick, Trans. М. Y. Acad. Sci. 12: 30. pl. I. f. 5. 1892. The specimens which we have figured and included under the above name are unquestionably identical with most of the fossil coniferous * Monog. U. $. Geol. Surv. 26 (Fl. Amboy Clays): 52. ^ Gymnosp. Böhm. Kreidef., зт. pl. 7. f. 4. CRETACEOUS CONIFERALES 39 remains commonly referred to Sequoia Reichenbachi. In both the clay layers and the lignitic debris of the Androvette pit the leafy twigs of this species are the most abundant of the many so-called Sequoia fragments; but examination of their internal structure has demonstrated that they do not belong to that genus, as will be sub- sequently described, and hence the generally accepted generic name must be abandoned. The species is also clearly not referable to the genus 4raucarites as defined by either Sternberg? or Endlicher,3% but belongs in the genus Geinitzia of the latter author, in which he included it under the name Geinitzia cretacea,?" ignoring the earlier specific name given by Geinitz. In view of the above mentioned facts and in accordance with the accepted rules of nomenclature, we have there- fore adopted for the name of this species the new combination Geinitzia Reichenbachi (Gein.). The general appearance of the leafy twigs, natural size, as they occur in Ше clay layers, is well indicated in figs. 7-10, Pl. 5, and the details of the external characters in the enlargements shown in figs. 3, 4, Pl. 8. These latter two figures represent a single speci- men, X IO, viewed from opposite sides, and they show both the falcate form of the leaves and their tetragonal cross-section. It will probably be conceded by paleobotanists that our figures serve to identify the specimens with the species beyond any reason- able doubt; but the fact should be borne in mind that different authorities have included a number of more or less different forms under it, some of which may have been referred to the original species incorrectly, and this has perhaps resulted in crediting it with an erroneous and altogether too extended geographic and strati- graphic range. As commonly recognized its geographical distribu- tion covered the United States, Canada, Greenland and Europe, while stratigraphically it apparently extended from the upper Juras- sic to the end of the Cretaceous period. We, therefore, desire to have it understood that the facts of internal structure and the conclu- sions in regard to botanical relationship which are next described and discussed are referable only to the species as it occurs in the Creta- ceous deposits at Kreischerville or in their equivalents elsewhere. 35 Presl, in Sternb., Verst. 2: 203. 1838. # Gen. Plant. 263. 1836-40. °" Synops. Conif. 281. 1847. 40 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN Figure 2, Pl. 16, shows a general transverse section, X 30, of the leafy twig just referred to. The bases of two leaves may be distinctly made out, one on the left and the other on the right of the figure. There are less clear indications of the bases of three other leaves on the remaining portion of the periphery. In the center of the stem lies the woody cylinder, which encloses a larger pith than is found in the case of the living Sequoia gigantea Torr. The pith which occupies the center of the figure is largely taken up by stone cells, which appear dark on account of the thickness of their walls. Figure 3, Pl. 16, shows a longitudinal section of the same twig, x зо. The pith is well preserved but the tissues of the central cylinder are less perfectly represented. On the right and left certain irregularities of contour mark the bases of leaves. Figure 4, Pl. 16, is a transverse section of a leaf, X 40. Three resin canals may be seen, a median large one and two smaller lateral ones. Тһе mag- nification is not sufficient to show that the transfusion tissue, which is abundantly present, bends outward from the flanks of the bundle and tends to envelop the resin canals, thus presenting a marked contrast to the transfusional borders of the foliar strands in Sequoia, which are directed towards the sides and not towards the dorsal surface of the leaf. Figures 1, 2, Pl. 17, show other transverse sections of the same twig, X 28. In fig. 2 may be seen on the upper side a foliar gap in the woody cylinder, which subtends an outgoing leaf trace. Figures 3; 4, Pl. 17, show other similar sections of the same twig, X 40. These somewhat numerous figures are introduced in order to give as complete an idea as possible of the structure of this important Creta- ceous fossil, which is now for the first time studied from the stand- point of internal structure. Figure 1, Pl. 18, shows, in longitudinal section, X 40, the com- position of the pith in the same twig. It is largely occupied by stone cells such as do not occur in either of the living species of Sequoia, but are common in the pith of Brachyphyllum. The wall of the stone cells is strikingly laminated from the action of the charring process, to which is due the preservation of most of the Cretaceous material from Kreischerville. This lamination is accompanied by a consid- erable swelling of the cell. It is only sclerenchymatous elements which are affected in this way, tracheids, for example, not being noticeably influenced by heat. In branches which are less charred CRETACEOUS CONIFERALES 4! and less well preserved, Ше tracheary tissues have suffered greatly from decay, while the stone cells are much better preserved than they are in the material illustrated in our figures. This swelling up and lamination of sclerenchyma makes it necessary, in many cases, to study the structure of both badly and well preserved branches which present the anatomical peculiarities of Brachyphyllum macro- carpum and Geinitzia Reichenbachi. Figure 2, Pl. 18, shows a transverse section of the wood, X 180. It may be noticed that the resin cells, which are so characteristic of the wood of the living species of Sequoia, are quite absent in this case. Figure 3, Pl. 18, shows a longitudinal section of the wood, X 500, in which the pitting it obviously araucarineous. Figure 4, Pl. 18, shows another similar section, under the same degree of magnification, with alternating pits. In the tracheid to the right of that containing these pits may be seen a single row of pits, some of which are in contact and others show a round contour. We have in fact in this wood the same type of structure that is found in Brachyphyllum, as described on p. 35 of this Memoir. In the case of the remains commonly called Sequoia Reichenbachi we have accordingly to do, not with a species belong- ing to the modern genus Sequoia, but with one of araucarineous affinities, and this instance is another of the many in which we have learned the futility of attempting to definitely refer Mesozoic leafy twigs to genera still living on the sole basis of external habit. In this connection it is interesting to recall that Schimper?? expressed his ideas in regard to the probable relationship of the Cretaceous so-called Sequoias in the following words: “Les especes de l'epoque crétacée reproduisent par leurs feuilles tétragones courbees en faux l'aspect des Araucaria, avec lesquels elles ont été quelquefois con- fondues. Elles pourraient bien former un genre particulier." Locality: Androvette pit. Plate 5, figs. 7, 8, 10, collected by Arthur Hollick. Plate 5, fig. 9, collected by William Т. Davis. Figures 7, 9, specimens in Mus. Staten Island Assoc. Arts and Sci.; figs. 8, 10, specimens in Mus. N. Y. Bot. Gard. Plate 8, figs. 3, 4; Pl. 16, figs. 2-4; Pl. 17, figs. 1-4; Pl. 18, figs. 1-4, collected by E. C. Jeffrey and Arthur Hollick. Specimens in Jeffrey collection, Cambridge, Mass. % Palaeont. Veg. 2: 314. 4,2 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN GEINITZIA Sp. Plate 8, figs. 5, 6; Pl. 18, figs. 5, 6; Pl. 19, figs. 1, 2 In figs. 5, 6, Pl. 8, are shown opposite sides of a leafy twig, X 10, which has a close resemblance to the species last described, although the leaves may be seen to be more slender and less closely arranged on the twig. We have been unable to definitely refer this specimen to any one of the so-called Sequoia species, although it is suggestive of certain forms of 8. subulata Heer, from the Cretaceous of Green- land,*? and some others of the closely allied Cretaceous species, with one or another of which it may ultimately be proved to be identical. The internal structure, however, shows it to be so closely allied to G. Reichenbachi that no matter how it may be referred it cannot possibly be regarded as a Sequoia, but should be included in the genus Geinitzia. Figure 5, Pl. 18, shows a transverse section of this specimen, X 40. "The stem is much slenderer than is the case in the average branches of G. Reichenbachi, but it presents the same general fea- tures of structure. The leaves in the specimen under discussion, however, contain but a single resin canal. Figure 6, Pl. 18, shows a somewhat larger and better preserved specimen of apparently the same species, in transverse section, X 25. e leaf has the same single resin duct. In the pith may be made out stone cells similar to those found in С. Reichenbachi. Figure 1, Pl. r9, shows the structure of the wood and phloem, as seen in transverse section, X 150. The wood is without resin cells, as in С. Reichenbachi. ‘The phloem is well preserved and shows general features such as are found in all those araucarians of which Brachyphyllum and Geinitzia may stand as types. Figure 2, Pl. 19, shows a longitudinal approximately radial section of the wood of this species, X 180. In a single tracheid the pits are flattened by mutual contact, while in the others they are generally round. There can be no doubt that in this species also we have to do with an arau- carineous and not with a sequolineous conifer, and have further evidence which confirms the suspicion that other, if not all of the so-called Sequoias from the Cretaceous do not belong to that genus but to some genus or genera which represent a peculiar and probably extinct type of vegetation. 9 Fl. Foss. Arct. 3 (Kreide- P 102. pl. 27. f. 3b, 7a, 76, 8b, 8c, 15a; pl. 28. f. 3-6b; pl. 29. f. 2c, 7b; pl. 34. f. Та, CRETACEOUS CONIFERALES 43 Locality: Androvette pit. Collected by E. C. Jeffrey and Arthur Hollick. Specimens in Jeffrey collection, Cambridge, Mass. Genus EUGEINITZIA gen. nov. Eugeinitzia proxima sp. nov. Plate то, fig. то; Pl. 25, figs. 1-3 Remains consisting of peltate, usually six-sided cone scales, with a median depression occupied by a process similar to that found in the peltate end of cone scales of Sequoia; pelt attached to the axis of the cone by a relatively slender stalk; center of the stalk occupied by a fibrovascular cylinder surrounding a medullary core; resin pas- sages abundant in the cortical tissues, but wanting in the pith and in the wood of the central cylinder; a layer of periderm occupying, apparently, the lower side of the stalk. "These cone scales, of which we have found a considerable number of specimens, most of them in a fair condition of preservation, bear a superficial resemblance to those of Sequoia and are hardly distin- guishable from many similar scales referred to that genus and to the fossil genus Geinitzia by various authors in their descriptions of Cre- taceous plants, although we have not been able to definitely correlate our specimens with any particular species. Figure ro, Pl. 10, shows the external appearance and characters of one of these scales, X 5. The resemblance to Sequoia may be seen in the well-defined process which occupies the median depres- sion, while the six-sided outline of its peltate end is a feature strongly suggestive of Geinitzia. The stalk, by which the pelt is attached to the axis of the cone is, however, much slenderer than in Sequoia and more nearly like that of Geinitzia. There were no indications of the attachment of seeds, either to the stalk or to the peltate portion of the scale. Figure r, Pl. 25, is a transverse section, X 40, through the stalk of the cone scale shown in fig. 10, Pl. 10. The center of the figure is occupied by a fibrovascular cylinder surrounding a medullary core, just as is found to be the case in the cone scales of Sequoia. There are no resin canals in the pith or in the wood of the central cylinder. Resin passages are, however, abundant in the cortical tissues. On what appears to be the lower side of the stalk of the cone scale is a layer of periderm. Figure 2, Pl. 25, shows a transverse section, X 40, of the same cone scale at a slightly higher level. The woody 44 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN cylinder has grown larger and shows a concave depression on its upper surface. The bundles at this stage begin to separate from each other and, contrary to the conditions found in Sequoia, diverge widely into the very much flattened peltate portion of the cone scale. So quickly does the expansion of the narrow stalk into the flattened terminal shield-portion take place, that it 15 almost impossible to secure a section showing the transition from the one to the other. The small bundles arrange themselves around the margins of the disk and become completely surrounded with transfusion tissue, а feature of marked contrast to the scale bundles in Sequoia and at the same time one which indicates a strong affinity with the Araucarineae. In Sequoia, as has been pointed out by the junior author,*” there are in the upper portion of the bundles of the cone scales lateral wings of transfusion tissue connecting the bundles with one another. In fact in the cone scales of the Sequoiineae and Cupressineae there is present the same type of transfusion tissue which is shown in the leaf-bundle of Cunninghamia sinensis, as figured by De Bary.*! In this type of transfusion tissue the tracheidal cells originate from the flanks of the bundles and do not encircle the whole strand as in the vegetative leaves of the Abietineae or envelop the phloem part of the strand as is often the case in the bundles of the cone scales of living repre- sentatives of the Araucarineae. It has not been possible to trace any of the strands of the peltate portion of the cone scales under discussion into cicatrices of ovules, so that the attachment of the ovules in this case was probably on the stalk of the peltate scale near the axis of the cone. This view of the matter is strengthened by the discovery of one immature and carbonized cone composed of peltate scales, in which there are the remains of four ovules on the peduncular portion of each scale. Unfortunately the specimen is completely carbonized and cannot be sectioned to discover its pos- sible affinities with the isolated mature scales described above. Figure 3, Pl. 25, shows a transverse section, X 100, of one of the numerous small bundles found about the margin of the disk in the mature scales under discussion. It may be clearly made out that there is a radial disposition of the transfusion tissue around the bundle, which appears as a cluster of smaller cells in the center of the larger trans- fusion elements. “Mem. Boston Soc. Nat. Hist. 5: 449. pl. 70. f. 2 = a Anatomy of the Vegetative Organs of d Phanerogams and Ferns, fg. 183, p CRETACEOUS CONIFERALES 45 It seems highly probable that the cone scales of this type are of araucarineous affinities, especially as they are associated with numer- ous leafy branches of so-called Sequoias, which we have shown to be in reality twigs of araucarineous conifers, and this view of the matter is rendered the more probable by Endlicher's reference of Sequoia Reichenbachi, undoubtedly an araucarineous conifer, to the genus Geinitzia (l. c.), to which the isolated cone scales described above may well belong, since they show all the external characters of the scales found attached to complete cones in this genus. It appears then not unlikely that the cone scales just described may belong to the above or to an allied species, and if such should be determined to be the.case, the original specific name would have to be substituted for the one here proposed. Locality: Androvette pit. Collected by E. C. Jeffrey and Arthur Hollick. Specimens in Jeffrey collection, Cambridge, Mass. Genus PSEUDOGEINITZIA gen. nov. Pseudogeinitzia sequoiiformis sp. nov. Plate 10, fig. 11; Pl. 25, fig. 4 Remains consisting of peltate, four-sided scales; outer end of the fibrovascular cylinder of the stalk somewhat more gradually expanded than in Eugeinitzia, thus more nearly resembling Sequoia. Only a single specimen of this type was found. It is shown, x 5, in fig. тт, Pl. то. The condition of preservation was such that it was only possible to secure fairly good sections through the stalk. Part of one of these, in transverse section, X 100, is shown in fig. 4, Pl. 25. There is the same tubular fibrovascular cylinder as in the scales previously described, but in this type the expansion of the outer end is somewhat more gradual and resembles more or less the conditions found in Sequoia itself. It has not been possible to secure satisfactory sections of the peltate portion, but enough was made out to make it certain that the bundles surrounding the margin of the peltate disc in this case have the same encircling zone of transfusion tissue as in the scales described above. There seems in fact good reason to consider this scale, as well as the other, as araucarineous and as possibly belonging to the one genus, since the outline of the cone scale may not, perhaps, be regarded as a sufficiently important feature to bring about a generic separation, where the internal struc- 46 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN ture so closely agrees. It seems probable that where the cone axis of Geinitzia was much elongated the peltate expansions of the cone scales are hexagonal by contact with numerous other scales and that on the other hand, where the cone axis was short, in analogy with the living Sequoia, that the scales may have been generally four-sided. Among American figures which indicate elongated cones with six- sided scales may be cited the following,*? referred to Sequoia gracil- lima (Lesq.) Newb., while the shorter type, with four-sided scales, may be clearly seen in the cone attached to a leafy branch, figured by Fontaine* and referred to Sequoia ambigua Heer. We have thought it advisable to distinguish between those Geinit- 2445 which have the long cone and hexagonal scales and those which have the short cone with tetragonal scales, by making the two new genera, Eugeinitzia and Pseudogeinitzia. In that case our first described scale would come under Eugeinitzia and the second under Pseudogeinitzia. It.may, of course, turn out that the length of the cone is not a determining factor in the number of angles of the cone scale, in which case the shape of the scales may alone be taken into consideration in bestowing the names suggested. Locality: Androvette pit. Collected by E. C. Jeffrey and Arthur Hollick. Specimen in Jeffrey collection, Cambridge, Mass. Genus PROTODAMMARA Hollick and Jeffrey PROTODAMMARA SPECIOSA Hollick and Jeffrey Plate 4, figs. 1-11; Pl. то, figs. 1-3; Pl. 14, figs. т, 4, 5; Pl. 15, figs. 1—6; Pl. 16, fig. 1 Protodammara speciosa НоШсК and Jeffrey, Amer. Nat. до: 199. pl. г. f. 5-13; pl. 2, f. 1-5. Mch., 1906. Dammara minor Hollick, Monog. U. S. Geol. Surv. 5o (Cret. Fl. S. N. Y. and N. Eng.) : 40. pl. 2. f. 35-37. 1906. “ Dammara microlepis Heer (?)." Hollick, Ann. М. Y. Acad. Sci. II: 57. pl. 3. f. ga, 9b. 1898. These little cone scales, which we have thus far only found detached, are very abundant in the lignitic debris of the Androvette pit. Their araucarineous affinities have been determined beyond “Berry. Bull. N. Y. Bot. Gard. 3: pl. 48. f. ar, 22. 1903. v *! Monog. U. S. Geol. Surv. 15 (Potomac or Younger Mesoz. Fl.) : pl, 120. f. 6, 6a. CRETACEOUS CONIFERALES 47 question (1. c., Amer. Nat., pp. 199, 200, pl. 2, f. 1-5) and their relationship to Brachyphyllum is not improbable (J. c., p. 204). In figs. 1-9, Pl. 4, are shown a number of the scales, natural size, which indicate the variation in size and shape that may be found among them. Figures 10, 11, Pl. 4, represent respectively a narrow and a broad form, X 10, in which the acuminate apical process may be seen, as well as the three seed scars and the numerous longitudinal resin canals on the lower surface. Figures 1-3, Pl. 10, show other typical specimens, similarly enlarged. Figure 5, Pl. 14, shows a transverse section, X 40, through the base of one of the cone scales. In the lower part of the figure may be seen indications of the seven resin canals, which are ordinarily found on the inferior surface. The upper part of the scale is largely sclerified and the superficial tissues are covered by a layer of peri- derm. The magnification is not sufficient to show the very small fibrovascular bundles, which are present in the base of the scale. Figure 3, Pl. 15, shows the center of the upper portion of the last mentioned figure, X 100. Four minute fibrovascular bundles may be made out, three of which have their wood orientated upwards and a single one above has its xylem directed downwards. At a lower level than is represented in this figure there is a single bundle present in the scale, in which the wood is uppermost. Аса slightly higher level a single bundle of opposite orientation is derived from the upper surface of this bundle, after which the original bundle divides into three, as may be seen in the figure. After the separation of the upper bundle the scale broadens considerably, as is shown in trans- verse section, X 30, in fig. I, Pl. 15, being characterized at this stage by two thin lateral wings. Figure 2, Pl. 15, shows another trans- verse section of the cone scale, X 30, at about its thickest and broadest part, where it gives attachment to the three ovules. The bundles of the upper and lower series lie nearly in a line about the transverse middle of the scale. It is not easy to distinguish the upper and the lower series from one another, nor has it been possible to make out just when the upper, originally single, bundle divides into three for the supply of the three ovules. Figure 5, Pl. 15, shows a transverse section, X 40, through the median part of the scale at about the same level as that shown in fig. 2. Below may be seen the central of the seven resin canals, which characteristic- ally occupy the lower portion of the scale. Above the resin canals 48 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN may be seen a lighter stripe, mainly composed of the abundant transfusion tissue, which surrounds the bundles on all sides and at the same time connects them with one another. Figure 4, Pl. 15, shows the upper part of a transverse section, X 80, through nearly the same region as fig. 5. Above may be seen the central ovular bundle which passes into the median ovule. This and its two lateral companions appear to bend very sharply upwards from the complex of bundles occupying the horizontal middle of the ovuliferous scale below the attachment of the ovules. Below may be made out one of the lower series of bundles which lies immediately above a resin canal. Figure 6, Pl. 15, is a transverse section, X 180, through one of the inferior bundles of the cone scale, showing it surrounded by a thick cordon of transfusion tissue on all sides. The tracheids of the bundle may be distinguished by their small size, their central position and the dark hue of their walls. The lighter colored cells which surround the bundle are transfusion elements with small bor- dered pits, very similar to those which surround the fibrovascular bundles of the needles in living pines. Similar cordons of transfu- sion tissue have been described by Seward and Ford in the case of the bundles of the cone scales of species of draucaria,** and have been likewise observed by one of us in species of 4gathis. It is of interest to note in this connection that this probably ancestral type of relation of the transfusion tissues to the foliar trace is best marked in the vegetative leaves of the Abietineae and in the cone scales of the Araucarineae. It may be stated in general that the cone scales of the Araucarineae, living and extinct, can often be distinguished in an Isolated condition from those of the Cupressineae and Sequoi- ineae by the cordon of transfusion tissue which surrounds the fibro- vascular bundles on all sides, since in the last two mentioned tribes the transfusion tissue had characteristically the same disposition as is found in their foliage leaves, viz., on the flanks of the bundles only. Figure 1, Pl. 14, represents a longitudinal section, X 30, of a cone scale of Protodammara, a little to one side of the median longi- tudinal line. To the left is the spinous process, which is a usual feature of the cone scales of this genus, and which represents the end of the sterile bract, on the hypothesis that the araucarian ovuliferous scale is made up of the fusion of a pair of superposed scales equiva- lent to those found in the female cone of the Abietineae. The bulg- “ Philos. Trans. Roy. Soc. London B 198: 363. f. 27. 1906. CRETACEOUS CONIFERALES 49 ing upper part of the cone scale ends to the right in a minute pit, which marks the point of attachment of the median ovule. Passing downwardly to the right of this pit may be seen a dark stripe, the fibrovascular supply of the ovule. Below the projecting convexity of the upper surface of the scales are to be seen stone cells in the substance of the scale. Lower down still is a dark somewhat upwardly concave stripe, which is one of the fibrovascular bundles of the lower series. It is jacketed above and below with transfusion tissue. Below the transfusion tissue is the large central resin canal of the cone scale. The lower surface of the scale ends in a free surface or apophysis, the inferior limit of which is indicated by a marked depression. Figure 4, Pl. 14, shows another longitudinal section of the same scale, X 40, in a somewhat different plane. In this case the ovular depression can still be seen on the upper surface. The central resin canal of the scale is to be seen passing out into the base of the terminal spine. Figure 1, Pl. 16, shows another approximately median plane of longitudinal section, X 30, in which the median resin canal is seen to have passed out into the terminal spine. The figure in this case Is inverted to save space. Locality: Androvette pit. Collected by E. C. Jeffrey and Arthur Hollick. Specimens in Jeffrey collection, Cambridge, Mass. Genus ANOMASPIS gen. nov. Anomaspis tuberculata sp. nov. Plate xo, figs. 5, 6; Pl. 25, fig. 5; Pl. 26, fig. 1 Remains consisting of peltate cone scales, irregularly pentagonal (?) in outline; pelt without any median depression, surface flat or somewhat rounded, covered with polygonal tubercles; stalk slender below, rapidly expanding into the pelt above; substance of the scale occupied by numerous irregularly disposed strands of sclerenchyma, in the meshwork of which are the fibrovascular bundles; bundles irregularly disposed, ending in a cordon of transfusion tissue in the tubercles. In addition to the cone scales previously described, several other types were found in the material from the Androvette pit. Super- .ficial views, X 10, of the one now under discussion, are shown in figs. 5, 6, Pl. 10, together with a similarly magnified figure of a small 5 50 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN cone scale of Sequoia gigantea Torr., in fig. 7, Pl. то, for compari- son, in order to illustrate their striking external contrast. Figure 5 shows a somewhat lateral view so as to bring out the character of the slender stalk or peduncle rapidly expanding into the peltate disc. Figure 6 shows the characteristic tuberculate, rounded surface, with- out any median depression. Figure 5, Pl. 25, shows the internal structure of the species as seen in transverse section, X 12. Figure 1, Pl. 26, shows a view of a portion of the same section, X 40. The substance of the scale may be seen to be occupied by numerous irregularly disposed strands of sclerenchyma, interspersed in the meshwork of which are the fibro- vascular bundles. The intensely black cells which make up a large portion of the rest of the scale are for the most part cells surrounding resin canals. In the living Sequoia, a transverse section of the cone scale of which, X 8, is represented in fig. 6, Pl. 25, the bundles are arranged in a lower and an upper series and have not the scattered disposition shown in our fossil. Тһе numerous irregularly disposed bundles in the cone scales under discussion end in a cordon of trans- fusion tissue in the tubercles described above as ornamenting the flat- tened surface of the peltate expansions. The structure of these terminations of the fibrovascular strands of the cone scales suggests araucarineous affinities. Locality: Androvette pit. Collected by E. C. Jeffrey and Arthur Hollick. Specimens in ‘Jeffrey collection, Cambridge, Mass. Anomaspis hispida sp. nov. Plate то, figs. 4, 8, 9 Remains consisting of peltate cone scales, similar in external appearance to those of 7. tuberculata, except that the surface of the pelt is not tuberculate but is covered with rather minute hispid projections. The exterior of one of these scales, X ro, is shown in fig. 4, Pl. 10. The internal structure does not differ materially from the spe- cies last described, but unfortunately in neither species are the tissues well preserved, so that it has not been possible to diagnose their structure very satisfactorily; but, so far as can be judged, their relationship with the Araucarineae may be assumed, at least pro- visionally. Cone axes of this species, X 10, are shown in figs. 8, 9, „то. i CRETACEOUS CONIFERALES 51 Locality: Androvette pit. Collected by E. C. Jeffrey and Arthur Hollick. Specimens in Jeffrey collection, Cambridge, Mass. Genus SPHENASPIS gen. nov. Sphenaspis statenensis sp. nov. Plate ro, figs. 22, 23; Pl. 26, figs. 2-4 Remains consisting of cone scales, wedge-shaped below, somewhat peltate above; surface of the pelt hollowed in its upper part, the lower lip of the excavation conspicuously projecting as an 1. cuspidate process; internal structure showing an upper and a low series of 1. bundles, the latter accompanied by resin nl the phloem sides of both series surrounded by a thick zone of trans- fusion tissue. This type of cone scale was found somewhat sparingly in the lignitic debris of the Androvette pit. “The characteristic wedge- shaped base and projecting cuspidate process on the lower surface are well shown, X 10, in the side view represented by fig. 23, Pl. то, which resembles the dental alveolus of a sea urchin. The hollowed outer surface is somewhat comparable to that of Sequoia and Geinitzia, but it lacks the process at the bottom of the depression and is clearly differentiated by the projecting cuspidate rim of the lower lip of the depression. Figure 22, Pl. го, shows the appearance, X то, of the inner basal portion and part of the upper surface of the scales, viewed from in front and slightly from above. A transverse section of the scale, X 15, 15 shown in fig. 2, Pl. 26. A series of numerous bundles stretches across what appears to be the upper side. On the lower there are fewer bundles and a number of large cavities which represent more or less disorganized resin canals. Figure 3, Pl. 26, represents a transverse section, X 15, at a higher level, in an inverted position to that shown in the previous figure. There is now a space in the center which represents a trans- verse section of the terminal depression. Figure 4, Pl. 26, is a por- tion of the latter section, X 100, showing one of the upper series of bundles, in which the phloem side is surrounded by a thick zone of transfusion tissue. This condition, which is found in both series of bundles, is a clear indication of araucarineous affinities. On account of their characteristic wedge-like shape and somewhat peltate ends we have adopted for these scales the generic name Sphenaspis, and a specific name which indicates the place where they were discovered. 52 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN Locality: Androvette pit. Collected by E. C. Jeffrey and Arthur Hollick. Specimens in Jeffrey collection, Cambridge, Mass. Genus DACTYOLEPIS gen. nov. Dactyolepis cryptomerioides sp. nov. Plate то, figs. 12, 13 Remains consisting of cone scales composed of an upper and a lower segment, the upper one divided into as many as seven sub- divisions or finger-like processes, the lower one undivided; each of the upper subdivisions containing a fibrovascular bundle which is completely surrounded by a cordon of transfusion tissue; lower seg- ment also containing fibrovascular strands. This type of cone scale is of somewhat rare occurrence in the macerated debris from the Androvette pit. Specimens, X го, are shown in figs. 12, 13, Pl. 10. It is of particular interest because its general appearance suggests at once the living genus Cryptomeria and also certain species of the extinct genus Voltzia. We have not been able to make out any indication of the number or mode of attachment of the seeds. Most of the specimens are poorly pre- served; but in one instance we were able to obtain a section which afforded sufficient data to clearly indicate its affinities. Each of the finger-like processes of the upper segment in this specimen was found to contain a fibrovascular bundle completely surrounded by a cordon of transfusion tissue, thus betraying its araucarineous relationship. In the lower segment the fibrovascular supply was too much car- bonized to decipher, although its presence was quite apparent. It is evident, in spite of the external appearance, that these scales cannot be related to Cryptomeria; but their resemblance to certain species of J’oltzia may have considerable significance, and if there should be any close affinity between our specimens and the latter genus, this will have to be removed from the Sequoiineae, where it is generally supposed to belong, and placed in the Araucarineae, and such an eventuality would only be in accord with the general results which we have obtained in our study of the coniferous flora of the Kreischerville deposits. The conspicuous finger-like processes naturally suggested for these scales the generic name Dactvolepis, and the resemblance to Cryptomeria suggested an appropriate spe- cific name. CRETACEOUS CONIFERALES 53 Locality: Androvette pit. Collected by E. C. Jeffrey and Arthur Hollick. Specimens in Jeffrey collection, Cambridge, Mass. Genus PITYOIDOLEPIS gen. nov. Pityoidolepis statenensis sp. nov. Plate 9, figs. 13, 14; Pl. 27, figs. 1-3 Remains consisting of cone scales having a superficial resemblance to those of the female cones of the American and Mexican nut pines, but lacking the depressions on the inner surface corresponding to the large seeds of these species. On the opposite surface is a well- marked apophysis, distinguished by a distinct transverse ridge, which appears to resemble the umbo of a nut pine, but without any evidence of annual growth as in the umbo of Pinus. A somewhat commoner type of cone scale from the Androvette deposit is one which had the general appearance of the scales of the female cone of one of the American or Mexican nut pines. The scales in question have the slender base which is found in certain nut pines, but are without the depressions corresponding to the large seeds of this section of Pinus. They have on the opposite surface a well-marked apophysis, distinguished by a distinct transverse ridge, which on superficial examination in badly preserved material appears to resemble the umbo of a nut pine. It is really not comparable to an umbo, however, as it does not exhibit in the mature scale the evi- dence of annual growth, which is inseparable from the idea of an umbo as it occurs in Pinus. Figures 13, 14, Pl. 9, representing two of these scales, X 10, sufficiently establish the truth of the above statements. Figure r, Pl. 27, shows the structure of the lower part of such a scale, in transverse section, X 25. The lower side of the scale is occupied by a series of resin canals, above which lie a series of small fibrovascular bundles. Above these again is a second series of bun- dles of inverted orientation. The lower bundles are distinguished particularly by the cordon of transfusion tissue, which surrounds them. In the upper series there is a median bundle much stronger than the others, which probably represents the vascular supply of a single median seed. Figure 2, Pl. 27, represents a similar section, X 100, showing on the lower side one of the inferior bundles with its Jacket of transfusion tissue, and above the large median bundle. 54 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN of the superior series, which probably represents the ovular supply. Figure 3, Pl. 27, illustrates, in similar section and magnification, the structure of one of the bundles of the lower series which is almost surrounded by a transfusion tissue, the tracheary cells appearing as small elements in the center of the larger transfusion cells. In the case of this species of scale again we have in all probability to do with part of an araucarıneous female cone. Оп account of its superficial resemblance to the cone scales of Pinus we have given it the generic name Pityoidolepis and have indicated its place of dis- covery in the specific name. We have found not a few fragments of flattened cones made up of scales of this species, which would in all probability, in impressions, be taken for cones of Pinus. Locality: Androvette pit. Collected by E. C. Jeffrey and Arthur Hollick. Specimens in Jeffrey collection, Cambridge, Mass. Genus ARAUCARIOPITYS Jeffrey ARAUCARIOPITYS AMERICANA Jeffrey Araucariopitys americana Jeffrey, Bot. Gaz. 44:435. pls. 28-30. 1907. This type of wood, found in the Androvette pit, we have not yet been able to definitely connect with any of the associated leafy twigs; but we have reasons for suspecting that it may represent the wood of Czekanowskia. It is clearly, however, an araucarineous conifer, as demonstrated by its internal structure and hence if the reference to Czekanowskia should turn out to be correct, this latter genus would have to be placed in the Araucarineae instead of in the Gingkoales, where it is generally considered to belong. The species is fully described and illustrated in the paper above cited, and any further discussion of its characters does not seem necessary. Genus BRACHYOXYLON gen. nov. Brachyoxylon notabile sp. nov. Plate 13, figs. 2-6; Pl. 14, fig. 2 “Brachyphyllum macrocarpum Newb.” Hollick and Jeffrey, Amer. Nat. 40: 203 [desc.], 214 ud pl. 5. f. 1-4. Mch., 1906. “© draucarioxylon sp." Ibid. f. 5, CRETACEOUS CONIFERALES 55 “Wood of Brachyphyllum." Jeffrey, Annals Bot. 20: 384-386. pl. 27. f. 1-12. Oct., 1906. Radial pits not all alternating or mutually flattened as in Jraucari- oxylon Kraus, but as often or more frequently of rounded configura- tion and not contiguous. The alternating or flattened condition of the radial pits generally confined to the terminal walls of the tra- cheids. Resiniferous elements characteristically absent. Wounds leading to the formation of traumatic resin canals, which are not present under any conditions in Araucarioxylon Kraus (Emend Hollick and Jeffrey). In our preliminary paper in the American Naturalist (/. c.) we identified certain araucarineous lignites present in the Androvette pit as probably the wood of Brachyphyllum. As a result of our fuller knowledge, derived from a continued study of the Staten Island lignites, we are no longer in the position to affirm that these frag- ments of fossil wood actually represent the wood of the genus Brachyphyllum, for, as we have determined, a number of other araucarineous conifers, which have in general been erroneously referred to other tribes of the Coniferales, possess the same general type of wood structure. The correct statement of facts appears to be that during the Cretaceous and earlier Mesozoic times there were present numerous representatives of a peculiar araucarineous tribe or sub-tribe, essentially different in anatomical structure from the genera of the Araucarineae still living. Brachyphyllum was only one of the numerous genera of this important group, which in all probability ceased to exist in the changes which ushered in the Tertiary period. The fact that on account of the richness of the display of this type of araucarian and the uniformity of ligneous structure presented by it, it is no longer possible to relate the lignitic fragments of wood to any one genus, makes the fuller knowledge of this type of wood not less but rather more important than at first appeared to be the case. igure 2, Pl. 13, shows a transverse section of this wood, X 40. Three annual rings are included in the field of view. The wood is con- spicuously without resiniferous elements, so far as can be judged from this section. Figure 4, Pl. 13, shows a longitudinal section of the same specimen, X 180. ‘The pits occur in a single row as in the case in the small-leaved species of the living Araucaria. They have the arau- carian peculiarity of being flattened by mutual contact. In fig. 5, Pl. 56 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN 13, is shown another similar section of the same specimen under the same magnification. Here occur in a single tracheid the alternating pits, which are generally present in the living araucarian genus .1gathis and in the section Colymbea of Araucaria. This о of affairs is comparatively rare in our fossils. In fig. 3, Pl. 13, 1 another longitudinal section of the same specimen similarly magni- fied. This section represents, on the whole, the commonest condition of the pitting. It may be observed in this case that the pits, which are present in a single row are not flattened by mutual contact, and thus present an agreement with the pitting found in the remaining tribes of the Conferales. This feature appears to be of great impor- tance, since the strictly flattened or alternate pitting of the living Дева 5 and Araucaria has suggested a relationship between them and the Cordaitales, in which the pits entirely cover the walls of the tracheids and are of the alternating type. The general trend of opinion of those who have in recent years investigated the Conif- erales is that the araucarian group is the oldest and that it bears a more or less close degree of relationship to the Cordaitales. Brachyphyllum and allied genera aftord reasons for strongly doubt- ing the correctness of this view, for representing as they do an older and now quite extinct branch of the Araucarineae, they nevertheless show in their pitting, and in other features, characters which are less araucarian than those found in the living genera .1gathis and Arau- caria— features, moreover, which approximate the araucarian stock to the Abietineae rather than to the Cordaitales. Figure 6, Pl. 13, shows a tangential view of the summer wood of the lignite under discussion, X 180. The tangential pitting of the tracheids and the generally shallow character of the medullary rays can be made out. Figure 2, Pl. 14, representing a transverse section of the wood, X 40, shows the nature of the wound reaction in this lignite. This phenomenon has been described by the junior author for this and sim- ilar lignites in an earlier article;* but it ıs advisable for completeness to refer to the matter again here. Contrary to the condition of affairs found in the living genera of the Araucarineae, a wound in the case of the clearly marked group of conifers for which Brachyphyllum in the present instance may stand as an example, results in the for- mation of traumatic resin canals such as occur under similar condi- tions in the Abietineae. In the figure last cited a row of somewhat ^ Annals Bot. 20: 383-394. pls. 27, 28. 1906. CRETACEOUS CONIFERALES 57 compressed vertical resin canals may be seen to stretch away from the right-hand side of the mass of wound callus, which marks the site of a healed wound. It may be seen from the above data that the type of wood found in the well-marked extinct group of araucarineous conifers represented by Brachyphyllum and by the lignitic fragments just described, differs from that present in the living Agathis and Araucaria in several important particulars, which serve to relate the group in question to the Abietineae. Since this type of wood is quite distinct and easily recognizable, it seems very desirable that it should bear a special and distinctive name, which will clearly separate it from Araucarioxylon Kraus. In the latter genus the pits are always alternating or at least flattened by mutual contact, while in the type under discussion they are quite as often remote and round in configuration as they are flattened by mutual propinquity, and seldom or never alternate. The resiniferous ele- ments which are commonly, although not universally, present in the wood of Araucaria and its immediate allies are absent, so far as we have been able to distinguish, in rather abundant material of Brachy- phyllum and related genera. Further, wound reactions, which, con- trary to what might be supposed, are frequently represented in fossil woods, supply a clear distinction between ‚/raucaryoxylon and the type under discussion; for in the latter wounds lead to the formation of traumatic resin canals, such as have not been found to occur in Araucarioxylon or in the wood of Agathis and Araucaria. We have considered it advisable to discuss at length this new type of araucarineous wood and to designate it by the generic name Brachyoxylon in order to define its affinities, since it is very abundant in the Kreischerville deposits and is found to characterize leafy twigs which on the basis of the vegetative habit alone, have been referred to almost all tribes of the Coniferales, other than the Abietineae and Araucarineae. Locality: Androvette pit. Collected by E. C. Jeffrey and Arthur Hollick. Specimens in Jeffrey collection, Cambridge, Mass. 58 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN Genus ARAUCARIOXYLON Kraus Araucarioxylon noveboracense sp. nov. Plate 21, figs. 1-3, 5, 6 “ Wood of Araucaria or an allied genus.” Jeffrey, Annals Bot. 20: 388. pl. 28. f. 14, 15. 1906. In fig. 1, PL 21, is shown a transverse section, X 40, of an draucarioxylon which is very common in the lignitic remains found in the Drummond pit. We have not found a single fragment of any similar remains in the Androvette pit, which supplied practically all the leafy twigs with structure sufficiently well preserved for more or less accurate diagnosis. This Araucarioxylon was previously figured and described as the wood of Araucaria or an allied genus, in a paper on “ The Wound Reactions of the Genus Brachyphyllum " in the Annals of Botany, above cited, but a fuller description is here supplied. Figure 1 may be compared with the similar section of Brachyoxylon shown in fig. 2, Pl. 13, having the same degree of magnification. It may be seen that in our specimen of Ягаисапоху- lon the tracheids are considerably larger than they are in the lignites presenting the structural peculiarities of the genus Brachyphyllum, and that, further, there are somewhat numerous cells present with dark contents, comparable to similar cells occurring in the wood of certain species of the living „ра из and Araucaria. Such cells appear to be absent in the lignites of the Brachyphyllum type. Fig- ure 2, Pl. 21, is part of the section shown in fig. 1, X 180. The res- iniferous a interspersed among the tracheids may be still more clearly made out in this figure and the additional fact may be dis- cerned that the tracheids are filled with a matrix similar to the muci- laginous contents found in the tracheids in wounded roots of living species of 4gathis and Araucaria, as one of us has observed. This matrix is in all probability of the nature of a fossilized gummy secre- tion, formed as a result of wounding. This conclusion gains strong probability from the fact that in the actual trunk from which this sec- tion was made, which still retained its bark, numerous citatrices were present on the outside of the stem as well as evidences of healed wounds in the deeper layers. In the article on the wound-reactions of Brachyphyllum, above cited, mention has been made of the nature of the wound-reactions in living araucarians. Figure 4, Pl. 21, CRETACEOUS CONIFERALES 59 shows the margin of a healed wound in 4gathis alba (Rumph.) Salisb., from material derived from the Botanical Garden at Buiten- zorg, Java. There is a considerable amount of callus parenchyma on the right of the figure and some of the tracheids adjacent to this are blocked with mucilage. To the left of the wound-callus the tracheary tissue presents a normal appearance. This condition is to be contrasted with the state of affairs found in the Brachyoxylon type of wood, as represented in our fig. 2, Pl. 14, which shows a row of traumatic resin canals running from the callused margin of a healing wound. We have found that Brachyoxylon presents this marked contrast to Araucarioxylon, viz., that whereas traumatic resin canals are characteristically formed as the result of wounding in the former, they are invariably absent in the latter. That a similar contrast occurs in our Araucarioxylon is clearly shown in our fig. 3, Pl. 21, which represents a transverse section of the wood, Xx 40, and shows the margin of a healed wound in this species, for there is the same absence of traumatic resin canals which is found to be characteristic of the living genera: Agathis and Araucaria. We have made a large number of observations on this point and feel that the above general statement as to the absence of traumatic resin canals holds as strictly in the case of our fossil as it does in the case of the genera last mentioned. Figure 5, Pl. 21, shows a longitudinal radial section of the wood of our Araucarioxylon, X 180. The preservation is not good and this, together with the presence of mucilage on the tracheids, somewhat obscures the structural fea- tures. It may be noticed that the pitting is strictly of the type found in Agathis and the large-leaved Araucarias of the section Colymbea, for the bordered areas are for the most part in several rows per tracheid and are uniformly alternate. We have found the condition of alternation or mutual flattening to be universal in this wood, just as it is in Araucaria and Agathis. Figure 6, Pl. 21, shows another radial section under the same magnification, which elucidates the structure of the resiniferous elements. One of these is shown in the middle of the field, which is crossed by a transverse septum. The resin cells in our species are of the nature of resiniferous parenchyma similar to that found in Cupressinoxylon Goepp. and present a contrast to the resiniferous tracheids which are found in 4gathis australis (Lamb) Salisb. 'They resemble, however, similar parenchymatous resinif- 60 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN erous elements which occur in the wood of 4. alba and other species of the living Araucarineae. It may be seen from the above description that our ./raucarioxylon presents a close resemblance to the structure of wood found in the modern genera of the Araucarineae, but differs from the extinct type of araucarineous wood which we have described under the name Brachy- oxylon, in the invariable flattening or alternation of the radial pits; in the presence of resiniferous xylem parenchyma, and in the absence of traumatic resin canals. In Brachyoxylon the pits are frequently not marked by flattening due to mutual contact but are often round as in the Abietineae. Alternation of the pits is generally absent on account of the small size of the tracheids. Resiniferous parenchyma does not occur. Traumatic resin canals in the latter genus, in con- trast to .1raucarioxylon, are present. It is obvious that we have in Brachyoxylon a clearly defined type of araucarineous wood, which is strikingly different from the ordinary Jraucarioxylon type. Wood of this type have apparently not escaped the attention of other ob- servers, although their points of difference from the ordinary Arau- carioxylon type have not been recognized. Seward, for example, has figured an araucarineous wood, which he calls Araucarioxylon Lindleii,*® which is characterized by rows of intercellular spaces, similar to those seen in our fig. 2, Pl. 14, which he compares with the resin canals of the Abietineae, but considers them not to be regular enough to merit the name of resin canals. In his fig. 3, pl. 7 (1. c.), their appearance is quite as regular as the traumatic resin canals found in the abietineous genus T'suga; while Goeppert has figured the pits of the wood as seen in actual leafy branches of the ancient arau- carineous genus Ullmannia as occuring in a single row апа not flat- tened by mutual contact. This is а much more likely description of the wood of this ancient Mesozoic conifer than that figured by Zittel*® from associated fossil woods, which were not certainly the remains of Ullmannia, and may well from the nature of their pitting have belonged to Cordaites. The question as to the relative anti- quity of the 4raucarioxylon and Brachyoxylon types of araucarineous woods has an important bearing on the subject of the phylogeny of the Coniferales, as already discussed in connection with certain other ^ Cat. Mesoz. Plants, Dept. Geol, Brit. Mus. (Jurassic Flora, Pt. II, Liassic and о 2. 56. a 6. f. 1-4; pl. 7. 2 2-5. “ Monog. Foss. Conif. pl. 20. f. 3. ^ Handb. Palaeont., Abth. | 275. f. 100. CRETACEOUS CONIFERALES 61 facts derived from the study of the internal structure of Cretaceous araucarineous cone scales. Locality: Drummond ри. Collected by E. С. Jeffrey and Arthur Hollick. Specimens in Jeffrey collection, Cambridge, Mass. CONIFEROUS REMAINS OF UNDETERMINED RELATIONSHIP SEQUOIA HETEROPHYLLA Velenovsky Plate 3, figs. 11-13 Sequoia heterophylla Vel. Gymnosp. Böhm. Kreideform. 22. pl. 12. f. 12; pl. 13. f. 2-4, 6-9. 1885.—Hollick, Trans. N. Y. Acad. Sci. 12: 30. pl. r. f. 21. 1892; Monog. U. S. Geol. Surv. 50 (Сте. Fl. S. М. У. and N. Eng.) : 41. pl. 3. f. 2, 3. 1906. Only a few fragmentary specimens of this species were found, con- sisting of impressions of leafy twigs in the clay layers. Three of the best of these are represented, natural size, in figs. 11-13, Pl. 3. No remains which could be definitely identified as belonging to the species were found in the lignitic debris, hence we have been unable to determine what its true botanical affinities may be, as in no instance was any specimen sufficiently well preserved for. sectioning and microscopical examination. The identity of our specimens with those which have been recog- nized under the above name cannot, however, be questioned; although, from what we have learned in regard to other fossil species com- monly referred to the genus, such reference must be regarded as wholly tentative. Locality: Androvette pit, Pl. 3, figs. 11, 13. Collected by E. C. Jeffrey and Arthur Hollick. Specimens in Mus. Staten Island Assoc. Arts and Sci. Old excavation, southwest of Killmeyer's hotel, Pl. 3, fig. 12. Collected by Arthur Hollick. Specimen in Mus. Staten Island Assoc. Arts and Sci. JUNIPERUS HYPNOIDES Heer Plate 5, figs. 5, 6 Juniperus hypnoides Heer, Fl. Foss. Arct. 6°: 47. pl. 44. f. 3, 4; pl. 46.f. 18. 1882.—Hollick, Trans. N. Y. Acad. Sci. 12: 29. pl. 1. 62 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN f. 1. 1892; Monog. U. S. Geol. Surv. 50 (Cret. Fl. S. N. Y. and N. Eng.) : 46. pl. 3. f. 12. 1906. Juniperus macilenta Heer [ ?], Fl. Foss. Arct. 6°: 47. pl. 35. f. 10, Iob, 11. The single specimen, represented natural size in fig. 5, Pl. 5, is the only one thus far brought to light in any of the Kreischerville material which could be definitely referred to this species; although isolated fragments, apparently belonging to it, are more or less abundantly represented in the lignitic debris. These latter, however, are difficult to distinguish superficially from other delicate coniferous twigs, unless the leaves are all preserved so that the phyllotaxy may be made out. Figure 6, Pl. 5, shows a terminal twig, with leaves attached, as it appears under an ordinary hand lens. If the two species described by Heer are to be regarded as dis- tinct, our specimen undoubtedly resembles hypnoides more nearly than it does macilenta, although specimens apparently identical with ours were described and figured by Newberry from the Cretaceous of New Jersey under the latter species.*? None of our specimens was found sufficiently well preserved for sectioning, so that we are unable to say whether the commonly accepted reference of this species to the genus Juniperus may or may not be justified. In this connection it is interesting to note that in the New Jersey clays the twigs of this tree are closely associated with cone scales of Dammara borealis Heer, which are undoubtedly closely related to our Protodammara speciosa; in fact, Newberry states that they are “sometimes apparently attached to the branch- lets,” and that “almost no other plant except this conifer is found with the cone scales, and it is difficult to avoid the conclusion that they belong together" (1. c., p. 47). This association may possess some significance, inasmuch as we have found in similar association, in the Androvette pit, the twigs of Juniperus and the cone scales of Protodammara, which latter has been determined to be araucarineous in its affinities. Locality: Androvette pit. Collected by William T. Davis. Speci- men in Mus. Staten Island Assoc. Arts and Sci. ? Monog. U. S. Geol. Surv. 26 (Fl. Amboy Clays): 54. pl. 10. f. 7. CRETACEOUS CONIFERALES 63 CZEKANOWSKIA CAPILLARIS Newberry Plate 6, figs. 1-3 zekanowskia capillaris Newb. Monog. U. 5. Geol. Surv. 26 (Fl. Amboy Clays) : 61. pl. 9. f. 14-16. 1895. Closely packed masses, irregular shaped fascicles, and isolated fragments of linear, striated leaves, occasionally forked, occur abun- dantly in the clay layers of the Androvette pit. “Two of these speci- mens, natural size, are shown in figs. І, 2, Pl. 6. Many of the frag- ments in which the forking is not apparent have much the appearance of pine needles and may readily be mistaken for them; in fact, so far as the gross appearance of some of the masses of leaves is con- cerned, these could hardly be distinguished from similar remains of Pinus Quenstedti Heer.° ‘They are evidently identical, however, with the remains described by Newberry under Czekanowskia capil- laris, but referred by him with hesitation to this genus. Some of our more robust specimens, however, such as the one represented by fig. 2, Pl. 6, may be satisfactorily compared with C. (Sclerophyllina) dichotoma Heer, from the Cretaceous of Green- land, and with C. nervosa Heer, from the Cretaceous of Portu- gal,5? which latter species has also been identified by Ward from the Cretaceous (Hay Creek series) of the Black Hills, while the more delicate ones, such as are represented by fig. 1, Pl. 6, are more nearly like certain of the forms of C. rigida Heer, from the Jurassic of Siberia,“ and as figured by Nathorst from the Triassic of Sweden.” Certain of these figures are in fact practically identical in general appearance with ours, and if the two series of figures are compared it is difficult to resist the impression of specific identity between them in several instances. It is also of interest to note, incidentally, that we find the same close association of these and certain other conif- erous remains in the Androvette pit as is mentioned by Heer in his ““Kreide-Flora " (1. c., p. бо), where he says that the remains of С. dichotoma occur “neben den Zweigen der Sequoia Reichenbachi.” ? Kreidefl. Moletein, pl. 3 * Fl. Foss. Arct. 3 oe FL): 59. pl. 17. f. 10, 11, 11b; pl. 20. f. ба. 9 Cont. Fl. Foss. аа. 17. pl. 16. f. 5-7a, 8-1 5! Nineteenth Ann. Rept. U. S. Geol. Surv. 1897-98, en II: 685. pl. 160. f. 1, 2. 5 Fl. Foss. Arct. 4 m: E зо. pl. 5. f. 8-11; pl. 6. f. 7; pl. ro. f. 2 ibid. 6 (Nachträge Jura-Fl.): 19. pl. 6. f. 7-12. 5 Kgl. Fysiogr. Sálsk. Handl. 17° (Om Nagra Ginkgováxter Нап Kolgrufvorna vid Stabbarp i Skáne): 11. pl. r. f. 9; pl. 2. f. 2-15. 1906. 64 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN The probable botanical relationship of the genus has never been satisfactorily determined, although the question has been discussed by nearly every author who has had occasion to study it. Heer says in his “ Jura-Flora ” (1. c., p. 65), “ Die auf Taf. V. und Taf. ГІ. с Pflanzen ШЕ einen so eigenthümlichen Pflanzentypus dar, das es schwer hält, für denselben die richtige systematische Stel- Dues auszumitteln” ; ànd he further remarks that the first impression is that the remains are bundles of pine needles, but that the forking of some of the leaves show that they cannot belong to this genus. Their resemblance to certain pteridophyta, especially to the genus Isoëtes, is then discussed, but is regarded as untenable, and the con- clusion is finally reached that they probably belong in the Gingkoales. This conclusion seems to have been quite generally accepted by paleo- botanists and the genus is now commonly regarded as closely related to Trichopitys, Jeanpaulia (Baiera), Gingkophyllum, etc. It is unfortunate that in none of our specimens is the basal part preserved. It is probable, however, that fig. 1 represents an upper and fig. 2 a lower portion. Figure 3 represents an enlarged frag- ment of one of the forking branches as it appears under a hand lens. In no instance have we been able to definitely identify any of the lignitic remains with the species under discussion, but it is possible that certain leafless twigs, showing well preserved internal structure, representing the genus .1raucariopitys (see р. 54 of this Memoir), may belong to it, and if such should be the case, it would be another example of a conifer of araucarineous affinities from the Kreischer- ville deposits, in this instance simulating the Gingkoales in external appearance. Locality: Androvette pit. Collected by E. C. Jeffrey and Arthur Hollick. Specimens in Mus. Staten Island Assoc. Arts and Sci. UNIDENTIFIED [TWIG Plate 9, fig. 15; Pl. 20, fig. 6 Figure 15, Pl. 9, depicts a small coniferous twig, X 10, rep- resented by a single specimen, which, however, sufficiently illustrates the external appearance of the species. It is strongly suggestive of certain twigs which have been described by Heer under the genus Brachyphyllum 5° from the Jurassic and Cretaceous of Portugal % В. о. Heer, Cont. Fl. Foss. Portugal: то. $1. ro. f. 4; B. corallinum Heer, ibid.: 21. pl. 12. f. 7-3. CRETACEOUS CONIFERALES 65 respectively, but our material is too scanty for any adequate com- parison. Figure 6, Pl. 20, shows the twig in transverse section, x 60. The leaves are somewhat rhombic in transverse section and contain a single resin canal, as appears in the upper part of our figure. The specimen is too immature for identification, since the woody tissues are poorly developed and do not warrant the expecta- tion that the pitting of the tracheids of the secondary wood will be made out in longitudinal sections, which have not been cut on account of the small amount of material. Locality: Androvette ри. Collected by Е. С. Jeffrey and Arthur Hollick. Specimens in Jeffrey collection, Cambridge, Mass. Genus CUPRESSINOXYLON Goeppert CUPRESSINOXYLON sp. Plate 20, fig. 2 Although we have had occasion to frequently refer to the monoto- nously araucarineous structure of the leafy twigs and wood from the Kreischerville deposits, most of which present marked external resemblances to certain genera of the living Sequoiineae, Cupres- sineae and Podocarpineae, it should not be inferred that indications of the presence of plants with cupressinoid structure were entirely absent. We have found scanty remains of lignite referable to the genus Cupressinoxylon, but specific designation is withheld in the hope of being able, later on, to definitely identify them with certain of the leafy branches. Figure 2, Pl. 20, shows the structure of this material in trans- verse section, X 40. Resin cells are very numerous, as in the Cupressinoxylon type of wood. Longitudinal sections of the wood serve to confirm the diagnosis, for there are numerous rows of resi- nous parenchyma present and the pitting of the tracheids presents no indications of araucarineous affinities. The lateral pits of the ray cells are few in number and large in size, which leads.to the suspicion that we may have to do with a representative of the Podo- carpineae. This conclusion is, however, to be accepted with some reserve, in spite of the attempt recently made by Gothan in his studies on the anatomy of living and fossil gymnosperm woods? to separate the wood of certain of the Podocarpineae from 5 Abh. K. Preuss. Geol. Landesanst. II. 44: 47. 1905. 6 66 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN the general Cupressinoxylon type. This group of conifers needs a much more comprehensive study than it has received before it will be possible to distinguish clearly its characters, either ligneous or reproductive. It is apparent from the data in regard to the Cupressinoxylon from Staten Island, above recorded, that the possibility of the exist- ence of cupressineous conifers at so early a period is not excluded, although it is obvious that they could not have been very numerous. We hope to return to this phase of the subject on another occasion, when more abundant material may be at our disposal. Locality: Drummond ри. Collected by E. C. Jeffrey and Arthur Hollick. Specimen in Jeffrey collection, Cambridge, Mass. Genus STROBILITES Lindley and Hutton Strobilites microsporophorus sp. nov. Plate 1o, figs. 18-21; Pl. 24, figs. 2-6 Remains consisting of small, slender cones, much longer than broad; scales lanceolate-ovate, acute, closely imbricated, bearing on their edges numerous pollen grains or microspores, each of which is provided with two lateral wings or air sacs. In the lignitic debris of the Androvette pit occur numerous minute, slender cone axes, almost never complete, covered with very small scales. The scales have often about their edges a yellowish brown substance, which on superficial examination appears to be of the nature of a resinous exudation. This substance, when viewed with the microscope in favorable cases, is seen to be made up of pollen grains or microspores, bearing two lateral wings, such as occur among living coniferous pollens in the Abietineae and Podo- carpineae. The small cones which produce these microspores are obviously male aments of a conifer. It would be inferred from their pollen grains alone that they are either abietineous or podocarpineous in their nature. Figures 18-21, Pl. 10, show the superficial characters of these male aments, X 10. Their length and slender proportions, when complete, make it difficult to regard them as belonging to the Abie- tineae, ıf such a reference were not negatived by their internal struc- ture, as will be indicated at a later stage. The outward configuration of the scales is more in favor of their podocarpineous affinities, but CRETACEOUS CONIFERALES 67 the anther sacs do not open by a stómium аз is the case in Podocarpus and Dacrydium and, moreover, their microspores are not quite of the type found in the genera just mentioned. Figure 3, Pl. 24, shows a view of a transverse section of one of these aments, X 30. Inthe center is the axis containing a number of small fibrovascular bundles, arranged in a circle. These bundles are of no diagnostic value, although well preserved, for they are entirely com- posed of spiral ringed and reticulate elements, as regards the xylem, and consequently do not show any pitted tracheids, which could be used to establish the systematic position of the remains. From the surface of the axis are, derived a number of microsporophylls, each of which bears a pair of anther sacs. The pollen in the anther sacs, as has been pointed out, is winged, with two air chambers. Figure 2, Pl. 24, shows a longitudinal view of one of these small cones, X 30, as seen in tangential section. The fact that the anther sacs are in pairs may more readily be made out in this plane of section. In the right of the section may be distinguished the end of one of the microsporophylls. It is obviously tipped with a distinct process. Figure 4, Pl. 24, shows the process, X 100. It is largely occupied by a mass of transfusion tissue which forms a cordon about the bundle entering from the sporangiferous part of the lamina of the sporophyll. Figure 5, Pl. 24, shows a transverse view of the upward process of another microsporophyll X тоо. The center of the process is occupied by the small cells which constitute the fibro- vascular bundle proper. The bundle is in turn surrounded by a cordon of larger cells with bordered pits—the transfusion elements. Figure 6, Pl. 24, shows an image of some of the microspores, X 500. One of these below the center of the photograph shows a wing with particular clearness on the left, the corresponding right wing is only partially represented. In one of the microspores above and on the right is seen a transverse section through the two wings. In living Coniferales air chambers occur in only two tribes, the Abietineae and Podocarpineae. In the Abietineae they are confined to the genera, Pinus, Picea, Pseudotsuga, Cedrus, Abies and Pseu- dolarix and are absent in Tsuga and Larix. In the case of one species of Tsuga, however, according to Lemon they аге sometimes: present. In the Podocarpineae all the genera with the possible exception of Saxegothaea have winged microspores. There are no. other examples among the living Coniferales of winged microspores.. 65 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN The fact that air chambers occur in the pollen of two groups so widely separated geographically as the Abietineae and Podocarpineae makes it possible that the presence of air cells in the pollen grains of the Coniferales is an ancient character and that in the older repre- sentatives of various tribes they may have had a much wider distri- bution than at present. The structure of the sporophylls in our fossil points strongly to its araucarineous affinities. In neither the Abietineae nor the Podocarpineae, as far as we have observed, do the leaf traces of the sporophylls become surrounded in their upper region by a cordon of transfusion tissue. This feature is strikingly araucarineous. It seems accordingly not unlikely that our fossil belongs to this tribe. This probability is enhanced by the abietineous features of wood structure found in a large number of the araucari- neous species described from the Androvette pit. Fliche, in his “ Études sur la Flore Fossile de Argonne,” has described and figured three species of Cretaceous cones under the new genus Pseudo- draucaria. These are female cones, characterized by the presence of two seeds оп the cone scale as in the Abietineae and a less pro- found fusion between the ovuliferous and sterile scales than is found in the genus draucaria. Оп the whole it appears not improbable that the slender male cones which we have found in the Cretaceous deposits at Kreischerville belonged to a generalized araucarineous type, nearer in the structure of its male sporophylls to the Abietineae than are any of the existing Araucarineae. It may well have been one of the numerous genera with the Brachyoxylon type of wood structure, which have been described in the foregoing pages. Locality: Androvette pit. Collected by E. C. Jeffrey and Arthur Hollick. Specimens in Jeffrey collection, Cambridge, Mass. Strobilites Davisii sp. nov. Plate 3, fig. 10 Cone linear-elliptical in outline, about 5 cm. in length by 2.5 cm. in width at the middle; scales numerous, relatively thin, closely imbricated. | This 18 the most perfect cone thus far obtained from the Kreischer- ville deposits, and it more or less resembles certain Cretaceous cones * Bull. Soc. Sci. Nancy II. 14: 70-84. pl. 6. f. 3, 3, 4; pl. 7. f. 1,2. 1896. CRETACEOUS CONIFERALES 69 elsewhere described and figured.*? Unfortunately the condition of the specimen was such that it could not be sectioned for microscopic examination, so that its true generic or tribal relationship could not be determined. For this reason it is placed in the comprehensive genus Strobilites and a specific name given to it in honor of the dis- coverer, Mr. William T. Davis. It is in all probability, generically at least, identical with Fontaine's 4bietites angusticarpus (1. c.), but inasmuch as this name more or less definitely implies relationship with the Abietineae, it was not thought advisable to so refer it as long as any doubt remains as to its identity. Locality: Drummond pit. Collected by William T. Davis. Speci- men in Mus. Staten Island Assoc. Arts and Sci. STROBILITES sp. Plate 3, fig. 9 This specimen is little more than an impression in the clay and the determination of its generic or tribal relationship is not feasible. It is possible, however, that it may belong with some of the so-called Sequoia cones which are found more or, less abundantly in the Creta- ceous deposits of New Jersey, or, more probably, it may represent one of the species of cones from which some one or another of the scales described in this Memoir were derived and hence no distinctive specific designation is here attempted. Locality: Androvette pit. Collected by Arthur Hollick. Speci- men in Mus. Staten Island Assoc. Arts and Sci. ® Araucarites Goepperti Presl, in Sternb. Verst. 2: 204. pl. 39. f. 4. Pinus protopicea Vel. Gymnosp. Böhm. Kreideform. 31. pl. 7. f. 4 Abietites angusticarpus Font. Monog. U. S. Geol. Surv. 15 (Potomac or Younger Mesoz. Fl.) : 263. pl. 13 CONCLUSIONS The many new facts in regard to the internal structure of Creta- ceous conifers, described in this Memoir as the result of our studies of the plant remains from the deposits at Kreischerville, must inevi- tably have a more or less important bearing on the question of the evolution of the conifers in general and particularly the relationship of the Araucarineae to the other tribes of the Coniferales. This must follow whether our reference of certain branches and cone scales to recognized Cretaceous genera, known previously from impressions alone, is accepted or not. Thus, even if it be denied that the branches and cone scales we have described as belonging to or closely related to the genus Geinitzia are recognized as such or not by other students of Cretaceous plants, it nevertheless results that a very marked superficial resemblance to a well-defined and characteristic sequoiineous type, both in the habit of the leafy twigs and in the superficial appearance of the cone scales is entirely illusory, since the structure of these remains shows them to be unquestionably araucarineous. The conclusion must further be inevitably drawn from our observed facts in regard to this extensive Cretaceous conif- erous flora, that the superficial appearance of conifers of this period is in general of very slight value in determining their actual botanical affinities, and that even the reproductive organs, i. e., cones and cone scales, of Cretaceous conifers, cannot always be regarded as conclu- sive evidence of relationship, since these are externally sometimes as deceptive as to their true affinities as are the leafy twigs. Unless it be assumed that the Kreischerville flora is entirely unique and exceptional, a conclusion quite unwarranted by the nature of the impressions associated with the remains with structure preserved, its composition and the affinities of the forms represented must be of considerable importance from the evolutionary standpoint. It may be stated in general that there are present in these beds conifers resembling, either in their general habits or in their reproductive organs (external features) or in both, the Sequoiineae (Brachyphyl- lum, Geinitzia, Eugeinitzia, Pseudogeinitzia, Anomaspis, Sphenas- 70 CRETACEOUS CONIFERALES 71 pis), the Cupressineae (J/'iddringtonites, Thuites, Raritania, Dac- tylolepis), and the Podocarpineae (Androvettia), all of which are in reality not related to the tribes which they simulate but to the Araucarineae. While it might be possible to doubt the evidence for araucarineous affinities on the basis of the structure of either the leafy twigs or the cone scales alone, it appears quite impossible to question the independent and consonant evidence afforded by a study of the anatomy of both. If 1t be admitted that these remarkable genera represent an arau- carineous flora the question arises as to their affinities. Attention has been repeatedly called in the earlier part of this Memoir to the common Brachyoxylon type of wood possessed by this group of conifers, which differs, as we have defined it, from Araucarioxylon Kraus by the fact that the pitting is not strictly and uniformly araucarineous, but presents a transition towards the condition found in the Abietineae and other conifers. It also presents the further peculiar feature of having a characteristic type of wound reaction found in the Abietineae but not in the existing genera of the Araucarineae, which feature apparently indicates a transitional type of wood. Another type of araucarineous wood is also found in the Kreischerville deposits, which possesses all of the arau- carineous features found in Brachyoxylom, but shows an even closer affinity than the latter to the Abietineae, by the possession of the abietineous type of the medullary ray. То this type Araucariopitys belongs. Here not only the lateral walls of the ray cells are pitted but also the horizontal and terminal walls, as is the characteristic condition of the rays іп the Abietineae. Since іп Brachyoxylon and Araucari- opitys we have Mesozoic types of araucarineous wood, which both by their structure and wound reactions connect the existing type of araucarian wood (as well as 4raucarioxylon Kraus) with that which is characteristic of the Abietineae, the question naturally arises, whether the Araucarioxylon or Pityoxylon (abietineous) type is the more primitive. It is practically universally assumed at present, from the resemblance of the pitting found in araucarian woods to that found in the Cordaitales and other ancient types of gymno- sperms, that the Araucarineae represent the oldest conifers. With- out considering the argument for the primitive character of the Araucarineae, based on the supposed presence, in the case of the female cone of Araucaria of a ligular structure, comparable to that 72 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN found in the heterosporous Lycopodiales, an argument which may now be considered as having very slight support since it has been abandoned even by Professor Seward, the latest and most vigorous defender of the lycopodineous origin of the Araucarineae, we may proceed to examine the grounds for deriving the Araucarineae directly from the older gymnosperms with crowded radial pitting of the tracheids. In the first place it should be pointed out that the nature of the ligneous pitting alone cannot serve satisfactorily to establish relation- ships, since in that case the angiosperms and Gnetales must both be connected directly with the older gymnosperms, because they usually possess the same alternate type of pitting and Gingko, which shows so many primitive characters, must be allied with the Abieti- neae, since it shares with these the feature of opposite non-alternat- ing pits and the even more striking character of the bars of Sanio. As has been pointed out by Kraus, the kind of pitting found in the living Araucarineae and in Araucarioxylon Kraus, presents this general feature of difference from that found in the Cordaitales, etc. —that the bordered pits are for the most part crowded in the ends of the tracheids and do not cover the entire radial walls of the water- conducting elements. The wound reactions of Brachyoxylon and .draucariopitys con- stitute a strong objection to the derivation of the Coniferales through the Araucarineae from the older gymnosperms. It has been pointed out by the junior author® that the type of abietineous wood, characterized by the presence of resin canals in the mature secondary wood as the result of injury only, is less ancient than that in which the resin canals are a normal feature of the secondary wood, as is shown by the greater geological age of the abietineous type with normal resin canals in the wood, and by the fact that in accord- ance with the theory of recapitulation those Abietineae which do not normally possess resin canals in their secondary wood often show them in the first year's growth. The greater antiquity of those Abietineae, which, like Pinus, have resin canals normally present in the secondary wood, further seems to be placed quite beyond ques- tion by the discovery of structural remains of the ancient abieti- neous type, Prepinus, in which the leaves are characterized by a struc- ture of the fibrovascular bundles identical in general organization ? Mem. Boston Soc. Nat. Hist. 6:5. 1905. CRETACEOUS CONIFERALES 73 with that found in certain Cordaitales. If the occurrence of resin canals as a result of wounding in the wood of those more modern Abietineae (Cedrus, Abies, Tsuga, Pseudolarix), which are normally without resin canals in the secondary wood, is to be regarded as a reversionary phenomenon, as it apparently must be, in view of all the evidence now at our disposal, it involves no distortion of logical principles to regard the traumatic resin canals of Brachyoxylon and Araucariopitys as likewise reversionary features. Since the posses- sion of ligneous resin canals is a very ancient characteristic of the Abietineae, recorded by Goeppert as far back as the Carboniferous of Waldenburg, it is highly probable, on the basis of the evidence of traumatic resin canals, that Ше Brachyoxylon and Jraucariopitys types of wood are as truly derived from that of the older Abietineae as is the Cedroxylon type of Kraus found in the more modern Abie- tineae (Cedrus, Abies, Tsuga, etc.). It may accordingly be con- sidered as highly probable that the Araucarineae possessed of the Brachyoxylon and Araucariopitys type of wood are more or less closely related in their origin to the ancestral stock of the Abietineae. Further, the free ovuliferous scales found in the female cone of the Abietineae corresponds according to the weight of evidence in a general way to a reduced branch, axillary to the subtending so-called sterile bracts. In the Araucarineae, as is specially well shown in some of the fossil species described structurally in this Memoir, we have instead of the two separate scales a complex organ made up, if we may judge from the double nature of the persisting fibrovascular structures, of two scales, ovuliferous and sterile, such as are found in the Abietineae, congenitally fused together. If the fusion of ovulif- erous and sterile scales be admitted for the Araucarineae, and it can scarcely be denied in view of the strong degree of approximation of the wood structure of the older Araucarineae to that of the Abie- tineae, as described in this Memoir and exemplified by the Brachy- oxylon and Araucariopitys types of wood, it follows that the tribe of conifers which presents the ovuliferous scale in the free condition, viz., the Abietineae, is the older. But the strongest evidence in favor of the correctness of our view is furnished by the leaf structure of Cretaceous pines, which present a more archaic type of foliar bundle than is found in those species of Pinus now living. In the genus Prepinus the archaic characters of the leaf-bundle are most strongly marked of all, for 74 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN here is found true centripetal wood and a general organization of the transfusion tissues, similar to that present in certain Cordaitales. On the basis of the lines of argument derived from a consideration of the structure of the wood and of its wound reactions in Brachy- oxylon and .1raucariopitys; of the structure of the ovuliferous cone scales of recent and Cretaceous Araucarineae and of the archigymno- spermic type of leaf-bundle found in Cretaceous pines, and pre- eminently in Prepinus, we conclude that the type of araucarian charac- teristic of the Kreischerville flora is more ancient in its affinities than the araucarineous stock represented by the living 4 gathis and Ягаи- caria, and was clearly derived from abietineous ancestry and particu- larly from forms closely allied to the living genus Pinus. It will probably be urged by advocates of the araucarineous origin of the Coniferales that the comparatively slight representation of the abietineous stock in the Mesozoic is against their having been the progenitors of the modern groups of Coniferales. That the Abieti- neae were really scantily present in the Mesozoic flora is, however, open to question. In the Kreischerville deposits we have found remains of at least seven distinct species of Pinus and one species of Prepinus. It would be difficult at the present epoch to gather the water-borne relics of seven species of Pinus in the same resting place, except, perhaps, in a stream issuing from the pinetum of a botanical garden, and this in spite of the fact that there are some ninety species of the genus now in existence. There is accordingly no good reason to suppose that Pinus was not quite as well represented by species in the middle Cretaceous as it is at present. Even if we admit that the Abietineae were less numerous in the Mesozoic than at present, it appears to be a fallacious conclusion that on that account they were necessarily less primitive than the apparently more prevalent Arau- carineae. It would not be less reasonable to suppose that the richly branched and leafy crown of a tree is its oldest portion. It is highly probable in view of the various facts described in this Memoir that the Araucarineae actually did represent the abundant crown of the coniferous genealogical tree in Mesozoic time, but that by their almost complete extinction in the important changes which ushered in the Tertiary period, a lower, more ancient and overshadowed branch, the Abietineae, became a new “ leader ” and in its subsequent development very much obscured the original plan of phylogenetic development. The validity of this conclusion will be strengthened when certain experimental results, obtained by one of us in the case CRETACEOUS CONIFERALES 75 of the more modern tribes of conifers have been published. It may be stated in general that the abundance of araucarineous or any other plant remains at earlier periods of the earth’s history does not neces- sarily supply evidence of superior antiquity. Whether the arguments advanced above for the derivation of the Araucarineae from a stock closely allied to the still flourishing but very ancient genus Pinus, be accepted or not, it is quite clear that a large part of the coniferous flora of the Cretaceous and probably earlier periods, was composed of types of araucarians, differing markedly in their wood structure, their wound reactions and general organization from the existing genera Agathis and Araucaria and their near relatives of the Mesozoic. In accordance with the facts thus far determined in regard to the structure of recent and fossil araucarineous woods, the Araucarineae may be separated in three sub-tribes as follows : ARAUCARIOPITYOIDEAE Wood tracheids with both araucarineous and abietineous pitting, the araucarineous pitting ordinarily confined to the ends of the tra- cheids. Traumatic resin canals resulting from injury often present. Medullary ray cells with abietineous pitting, :. e., with pits on the horizontal and terminal walls of, as well as on those in contact with, the tracheids. Leaves probably occurring on short shoots as in Pinus. Reproductive organs not accurately known (draucariopitys, Czekanowskia (?)). BRACHYPHYLLOIDEAE Wood tracheids with both araucarineous and abietineous pitting, the araucarineous pitting confined to the ends of the tracheids. Traumatic resin canals may be formed as the result of injury. Medullary ray cells with araucarineous pitting, i. e., the walls smooth except those in contact with the tracheids. Leaves not occurring on short shoots, various. Scales of the female cones various, peltate to squamiform, persistent or deciduous (Brachyphyllum, Geinitzia, Raritania, Widdringtonites, Thuites, Androvettia, etc.). ARAUCARIOIDEAE Wood tracheids with araucarineous pitting (radial bordered pits alternating or flattened by mutual contact). “Traumatic resin canals 76 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN not formed. Rays strictly araucarineous, with pits on the walls in contact with the tracheids only. Leaves not on short shoots, various. Female cone scales squamiform, deciduous (./gathis, Araucaria, Albertia (?), Pseudo- Araucaria (?)) It appears probable that the above order of enumeration of the sub-tribes corresponds to their phylogenetic sequence, the са pityoideae being the oldest. SUMMARY 1. By special means of isolation we have been able to secure from the Androvette and Drummond clay pits at Kreischerville, Staten Island, N. Y., structural material of leafy twigs, cone scales and fragments of lignite, representing a considerable number of impor- tant Cretaceous coniferous genera. 2. Some of these remains are genera already known from. impres- sions, others represent genera which are new to science. 3. The coniferous flora of the Androvette pit consists of abun- dant remains of species of Pinus and of other conifers, which have been referred to genera supposed from the habit of their leafy branches, or of their cones, or of both, to belong to the Sequoiineae, the Cupressineae and the Podocarpineae. 4. The conifers referred on their superficial characters to the Sequoiineae, Cupressineae and Podocarpineae, turn out from micro- scopic examination of their leafy twigs and cone scales to belong to the Araucarineae. 5. They represent a special sub-tribe of the Araucarineae, for which the name Brachyphylloideae is proposed. 6. These conifers are characterized by a special type of wood, which we designate Brachyoxylon. 7. The wood structure of the Brachyphylloideae clearly allies them with both the Araucarineae (in the narrower sense) and the Abietineae. 8. Thereis good evidence that the Araucarineae have come through the Brachyphylloideae from ancestors allied to the Abietineae. 9. The abietineous remains from the Androvette pit belong to the genera Prepinus and Pinus. The species of Pinus are in general more archaic than any living species. Prepinus shows features of structure more primitive than those known to exist in any other conifer living or extinct. PLATES PLATE 1 PLATE 1 PAGE Androvette pit, Kreischerville, N. Y. Мем. N. Y. Вот. GARD. Vor. HI. Pr. т. ANDROVETTE PIT, KREISCHERVILLE, N. Y. PLATE 2 PLATE 2 PAGE Drummond pit, Kreischerville, N. Y. . : 5 : А А . 6 Mem. N. Y. Вот. Garp. Vor HI. Pr. 2. DRUMMOND PIT, KREISCIIERVILLE, N. Y. PLATE 3 PLATE 3 Figs. 1-5. Androvettia statenensis sp. nov., nat. size . 6, (7?) Pinus triphylla sp. nov., nat. size 8. Bark of Pinus sp.?, nat. size 0. Strobilites sp., nat. size : IO. Strobilites Davisii sp. nov., nat. size . 11-13. Sequoia heterophylla Vel., nat. size MEM. N. Y. Bor. GARD. Vor. III. Pr. 3. CRETACEOUS CONIFERALES. PLATE 4 PLATE 4 12-14. Brachyphyllum macrocarpum Newb. 1g. I2, nat. size. Figs. 13, 14, X 10. PAGE Vor. Ш. Pr. 4. Мем. N. Y. Bor. Garp. CRETACEOUS CONIFERALES. PLATE 5 Figs. 1-4. 5, 6. 7-10. PLATE 5 Widdringtonites Reichii (Ettingsh.) Heer igs. 1-3, nat. size. Fig. 4, enlarged. Juniperus hypnoides Heer Geinitzia Reichenbachi (Gein.) comb. nov., nat. size ZEN Ха NOOO > SQA Wg, | АА ЕЕЕ М 2 E | EN 222 © SS SSS I N. Y. Bor. Garp. Мем. МІҒЕКАТ,Е5. CRETACEOUS СО PLATE 6 PLATE 6 PAGE Figs. 1-3. Czekanowskia capillaris Newb. . Figs. 1, 2, nat. size. Fig. 3, enlarged. 4-7. Raritania gracilis (Newb.) comb. nov. . я : . 26 Figs. 4—6, nat. size. Fig. 7, enlarged. PLATE 7 Figs. 1-8. PLATE 7 PAGE Androvettia statenensis sp. : я v 22 Fig. 1. Surface of a о ж 7. Fig. 2. Surface of the base of another specimen, X 10. Fig. 3. Surface of another, larger specimen, X Fig. 4. Specimen showing lateral leafy branches, X 9. Fig. 5. Specimen showing lateral branch, bearing imma- ture male cones or aments?, Figs. 6-8. Lateral leafy branches, X 7. Mem. N. Y. Вот. Garp. NOIA CRETACEOUS CONIFERALES. PLATE 8 Figs. 1, 2. PLATE 8 A ndrovettia statenensis sp. nov. Fig. 1. Surface of a fragment of a jue о. x ч Fig. 2. Surface of another large specimen, X 8 Geinitzia Reichenbachi (Gein.) comb. nov. Opposite sides of one specimen, X то Geinitzia sp. Opposite sides ut one specimen, x IO. Widdringtonites Reichii EE Heer. dno i X IO Thuites sp.? E BS x IO р Figs. 12, 13. Opposite sides of one specimen. Figs. 14, 15. Opposite sides of another specimen. Figs. 16, 17. Opposite sides of another specimen. Fig. 18. Fragment with single pair of verticillate leaves. PAGE Vor. III. Pr. 5, Мем. N. Y. Вот. GARD. CRETACEOUS CONIFERALES. PLATE 9 PLATE 9 Raritania gracilis (Newb.) comb. nov. Fragments of leafy twigs, X то. Е Я А А : : Е . Cone of Brachyphyllum sp.? Opposite sides of one speci- men, X IO А А . . Brachyphyllum € N Fragments of leafy twigs X 10. р . Prepinus statenensis bs ма то Cone scale of Pinus sp. шы sides Е: one specimen, X IO : à Pityoidolepis statenensis sp. nov., ve то Unidentified twig, X 10 . | АСЕ Мем. N. Y. Вот. Garp. Vor. ПТ. Pr. 9. CRETACEOUS CONIFERALES. PLATE 10 Figs. 1-3. 4, 8, 9. 5, 6. 7. 10. 11 12, 13 14-17 18-21. 22, 23. PLATE 10 Protodammara speciosa Hollick and те of scales, X 10 Anomaspis hispida sp. no ig. 4. Outer surface ae ee К 10. Figs. 8, 9. Cone axes, X 10. Anomaspis tuberculata sp. nov. . Fig. 5. Stalk and pelt, X 10. Fig. 6. Outer surface of pelt, X 10. Cone scale of Sequoia іы Torr., X 10. for comparison Eugeinitzia proxima sp. nov., x 5. Pseudogeinitzia sequoiiformis sp. nov., X 5 Dactyolepis cryptomerioides sp. nov., X 10 . Raritania gracilis (Newb.) comb. nov. Figs. 14, 15. Fragments of twigs, X 5. Fig. 16. Cone?, X 5. Fig. 17. Same cone, X 10. Strobilites microsporophorus sp. nov., X 10 Sphenaspis statenensis sp. nov., X 10 Upper surface Introduced PAGE 49 Mem. N. Y. Вот. GARD. Vor. III. Pr. 10. 1 ! ls х AV A TES Пе. Pa le Ў VENIT Uu k ХА TO p CRETACEOUS CONIFERALES. PLATE 11 PLATE 11 PAGE Figs. 1, 2, 4, 5. Brachyphyllum macrocarpum Newb. : . 33 Fig. г. Transverse section of a branch, X 11. Fig. 2. Similar section of a smaller branch, X 28. Fig. 4. Central part of the section shown in fig. 1, Х 15. | Fig. 5. Longitudinal section through the center of the branch from which fig. 1 was taken, X 15. 3. Cone of Brachyphyllum sp.? Transverse section of the peduncle of the specimen shown in figs. 5, 6, Pl. 9, X 10. , . , А қ я > 237 Mem. N У. Вот. GARD. CRETACEQUSTCONIEERALES PLATE 12 Figs. 1-6. Brachyphyllum macrocarpum Newb. . -PLATE 12 igs. 1, 2. Parts of the transverse section ова in fe 4, Fig. 3. Part of the same section, showing the wall of the central cylinder, X 180. Fig. 4. Transverse section through a portion of a branch, showing forking leaf-trace, X 40. Fig. 5. Similar section through the outer portion of a leaf, 40. Fig. 6. Similar section, showing transfusion tissue, X 180. PAGE “ 33 Bot. GARD. N Y. MEM. PY T- a » Fs T yy! «е e 1 99 № RACE 4 » TOUS CONIFI CRETACI PLATE 13 PLATE 13 PAGE Fig. т. Brachyphyllum macrocarpum Newb. ‘Transverse section, show- | ing the departure of the leaf-trace from the central cylinder, X 40. : : . қ : . 33 2-6. Brachyoxylon monde sp. nov р : - 54 Fig. 2. Transverse section di the — X 4 40. Fig. 3. Longitudinal section of the same specimen, X 180. Fig. 4. Similar section of another part of the same specimen, showing flattened pits, X 180. Fig. 5. Similar section of the same specimen, showing alterna- tion of the pits, X 180. ig. 6. 'langential section of the summer wood of the same specimen, showing the tangential pits, X 180. Mem. N Y. Bor. GARD. ar ETAT и HHP Mn Д ЖА ТО 2% 4 Mee LY TET Ti AR STEER TA ОНА RL ИНИ н 2 Миа 7 иа ГА И Ы LIP ^ ІНІ ПАЯ ТЕНИ?) Ж ДІН Ж lag ОД 7, / d АУ 4 A 221) RN ДИ И РАР ИИ ЙЕЛ и HH: DIET Te за «а: амла ча LI TM MELIA Аз D AS Arig Au, чат NS CRETACEOUS CONIFERALES PLATE 14 Figs. 1, 4, 5. D PLATE 14 Protodammara speciosa Hollick and Jeffrey . Fig. 1. Longitudinal median section of a cone "v X 30. Fig. 4. Similar section of the same specimen, in a some- what different plane, X 40. Fig. 5. Transverse section through the base of a scale, O. Brachyoxylon notabile sp. nov. ‘Transverse section of the specimen shown in figs. 2-6, Pl. 13, showing traumatic resin canals in injured wood, X 40 . Cone of Brachyphyllum sp.? Transverse section dud the axis of the specimen shown in i 5, 6, Pl. 9, X 35 PAGE 46 Mem. N Y. Вот. GARD. Os Ne OMAR CRETACEOUS CONIFERAL ES. PLATE 15 PLATE 15 Figs. 1-6. Protodammara speciosa Hollick and Jeffrey Fig. 1. Transverse section through the narrow ще ats a cone scale, X 30. Fig. 2 Similar section through the u part of a scale, Fig. 3. Central part of the transverse section shown in fig. 5, Pl. 14, X 100. Fig. 4. Similar section through the upper part of a scale, showing the double system of bundles, X 80. Fig. 5. Similar section through the median part of a scale, X 40. Fig. 6. Similar section through one of the bundles of the lower series, showing the surrounding transfusion tissue, PAGE Mem. N Y Bot. Garp. Vor. Jl. P TOL. EU CRETACEOUS CONIFERALES. PLATE 16 PLATE 16 PAG Fig. 1. Protodammara speciosa Hollick and кш Longitudinal sec- tion through a scale, X 30. р : ; . 46 2-4. Geinitzia Reichenbachi (Gein.) enm nov. . я . 38 Fig. 2. Transverse section of the twig shown in = 3, 4 Pl. 8, X 30. Fig. 3. Longitudinal section of the same twig, X 30. Fig. 4. Transverse section of a leaf, X 40. Mor НЕЕ TO. GARD. АЗА Вот Мем. dad D y Vi n Це, Ни M |. 3 CRETACROUS CONIFERALES. PLATE 17 PLATE 17 PAGE Figs. 1-4. Geinitzia Reichenbachi (Gein.) comb. nov. Figs. 1, 2. Transverse sections of the twig shown in Тм 3, 4, Pl. 8, at different levels, X 28. Figs. 3, 4. Similar sections of the same specimen, X 40. Mem. N Y. Вот. Garp. Мог. III. Pr 17. CRETACEOUS CONIFERALES. PLATE 18 PLATE 18 PAGE Figs. 1-4. Geinitzia Reichenbachi (Gein.) comb. nov Fig. 1. Longitudinal section through the ai of ae speci- men shown in figs. 3, 4, Pl. 8, showing the stone cells, X 40. Fig. 2. Transverse section of the wood, X 180. Figs. 3, 4. u. sections of the wood, X 500. 5, 6. Geinitzia sp. Fig. 5. аса section à of the branch сан in fee " 6, Pl. 8, X 40. Fig. 6. Similar section of another specimen, X 25. N Y. Гот. GARD. MEM. 4 &,. ҮТҮ: ¿AAA ------ әз чт 00190039588. - 2-е... cc RETACEOUS CONIFERALES. 3 C PLATE 19 Figs. 1, 2. 3-6. PLATE 19 Geinitzia sp. . : . 42 Fig. 1. ers: section of she mee: x I © Fig. 2. Longitudinal section of the wood, X 180. Raritania gracilis (Newb.) comb. nov. 26 Fig. 3. Transverse section of Ше twig clown in па 3; Fig. 4. Similar section of an older branch, shown in fig. 1, Pl. 9, X 45 Fig. 5. Longitudinal section through part of a branch, showing cracks due to decay, X 20 Fig. 6. Similar section through another specimen, show- ing the broken apex of a leaf, X 20. MEM. N у lor. GARD. SEN Ж 8 3: ® CRETACEOUS Uf )NIFER ALES. PLATE 20 Fig. 1. о 118 РГАТЕ 20 Raritania gracilis (Newb.) comb. nov. Longitudinal section of the wood, X 500 Cupressinoxylon sp. ere section of die бо, x 40 W'iddringtonites Reichii (Ettingsh.) Heer Pl. 8, Fig. 4. Fig. 5. Transverse section of the specimen own in e 8, X 40. Part of the section shown in fig. 3, X 60. Longitudinal section of the wood, X 500. Transverse section of the unidentified twig shown in = 15, Pl. o, Х 60 PAGE 64 Мкм. N Y. Вот. GARD. МО ОРТ P А ЗО. CRETACEOUS CONIFERALES. PLATE 21 PLATE 21 PAGE Figs. 1-3, 5, 6. Araucarioxylon noveboracense sp. nov. 58 ig. I. Transverse section of the wood, X 40. Fig. 2. Part of the same section, X 180. Fig. 3. Similar section, showing injured wood, X 40. Fig. s. Longitudinal radial section, X 180. Fig. 6. Similar section, showing resin cells, X 180, 4. Wood of Agathis alba (Rumph.) Salisb., in transverse section, X 40. Introduced for comparison 59 Mem. N Y. Bot. GARD. > АУ 4 y. aig sin. LT " vage y (ТЫ ГГ , tarto HH UR сэв at, ғ 4, 202252 НЕДА + ..Х БЕМ EI SE e, ез .Х p “2-4, cas i ue a une но ъв: е Ye un. q» Dr e? 7 E ose o. a y DE a RS 3 е SEO Er a о. а > Pa ей: 4 I uos ке 34 E А A ae es ‘an i DE а 4 һе. А СЕБЕ T uu ry А ~ ay 1 26. 5 5%» = 1: а а B SO кұ * N e v eM cM - LE Y ae CRETACEOUS CONI FERALES. PLATE 22 Fig. 1. a PLATE 22 Pinus triphylla sp. nov. Transverse section of a leaf fascicle, Pinus quinquefolia sp. nov. Transverse section of a leaf fascicle, Prepinus statenensis Jeffrey. ‘Transverse section through part of a leaf fascicle, X 10 . : Я A А я : Pinus tetraphylla Jeffrey. ‘Transverse section of a leaf fascicle, X40. : 5 ; я ч ; я Bark of Pinus sp.? ‘Transverse section, X 15 PAGE єч са MESE Vol. 5. CRETACEOUS CONIFERALE PLATE 23 Figs. 1-4. PLATE 23 PAGE Pityoxylon statenense Jeffrey and Chrysler р г . 20 Figs. 1, 2. Transverse section of the wood, X 10. ig. 3. Part of the section 2 in fig. 2, X 20. Fig. 4. Similar section, X 4 Des statenensis Jeffrey. Part of the transverse section shown in fig. І, Pl. 24, X 180 . . г . I9 Cone scale of Pinus sp. Transverse section, x 40 . . 16 5 ‚RALE + OUS CONIFI SPACE URI PLATE 24 PLATE 24 AGE Fig. 1. Prepinus statenensis Jeffrey. ‘Transverse section through the axis of a short shoot, X 18. (Shown in part in и 5, РІ. 23.) 19 2-6. Strobilites microsporophorus sp. nov. Р . . 66 Fig. 2. Longitudinal tangential section, X 30. Fig. 3. Transverse section, X 30. Fig. 4. Longitudinal section through the apex of one of the sporophylls, showing transfusion tissue surrounding the bundles, X 100. Fig. 5. Transverse section through another sporophyll, show- ing the relation of the transfusion tissue to the bundle, X 100. Fig. 6. Microspores, showing the wings, X 500. Mem. N Y. Вот. GARD. PLATE 25 Figs. 1-3. + a > PLATE 25 Eugeinitzia proxima sp. nov. Fig. 1. Transverse section Фф re zb - & сопе scale shown in fig. 10, Pl. 10, X 40. Fig. 2. Similar section through the same specimen, at a higher plane, X 40. Fig. 3. Similar section through the upper part of the same specimen, showing bundle surrounded by transfusion tis- sue, X 100. Pseudogeinitzia sequotiformis sp. nov. ‘Transverse section through part of the woody cylinder in lower part of a cone scale, X 100 . я ы А : е Е A nomaspis tuberculata sp. nov. “Transverse section through upper part of a scale, X 12 Transverse section through upper Bat of a cone sele T Sequoia gigantea Torr., X 8. Introduced for comparison PAGE Mem. N Y. BoT. GARD. Vor. HI. Pr. 25 = В 2252 CRETACEOUS CONIFERA PES PLATE 26 Fig. 1. 2-4. 130 PLATE 26 PAGE Anomaspis tuberculata sp. nov. Part of the transverse section shown in fig. 5, Pl. 25, X 40 . А : Е | . 49 Sphenaspis statenensis sp. nov. n . SI Fig. 2. Transverse section through the median TT ЕР а сопе scale, X 15. Fig. 3. Similar section through the same specimen at a higher plane, X 15. Fig. 4. Part of the section shown in fig. 3, showing a bundle surrounded by transfusion tissue, X 100. Mem. N Y. Bor. GARD. V ИЛЕР / OL. А ТОЛО " de DM i 4 CRETACEOUS С: )NIFERALE5. PLATE 27 PLATE 27 PAGE Figs. 1-3. Pityoidolepis statenensis sp. nov. 53 ig. 1. Transverse section (Шош: the lower pott of a cone scale, X 25. Fig. 2. Similar section of part of the same specimen, show- ing the double system of bundles, X 100. Fig. 3. Similar section of one of the bundles of the same specimen, surrounded by transfusion tissue, X 100. 4-6. Thuites sp.? 31 Fig. 4. ans section -— ТТ BS uomen: йы in fig. 18, Pl. 8, X 35. Fig. 5. Similar section of the same specimen taken at a higher plane of the internode, X 35. Fig. 6. Part of the section shown in fig. 4, X 45. Мем. N Y. OT. GARD. CRETACEOUS CONIFERALES. ” \ суы won Ж. cor >, 4, AU. da 7 PLATE 28 Figs. 1-4. 134 PLATE 28 PAGE Thuites sp.? 31 Fig. 1. Longitudinal section T" the — X 500. Fig. 2. Transverse section through the base of the speci- men shown in figs. I4, 15, Pl. 8, X 3o. Fig. 3. Similar section of the same specimen, X 50. Fig. 4. Central part of the section shown in = 3, X 100. 22 Androvettia statenensis sp. nov Fig. 5. Transverse section па de upper Best of a phylloclad, X 10. Fig. 6. Similar section through the edge of the phylloclad írom which fig. 7 was taken, showing an attached leaf, X 40. Fig. 7. Similar section through the upper part of the phyl- loclad, shown in part in fig. 6, X 20. Fig. 8. Similar section through the basal part of a phyllo- clad, X 40. NENNE: Мем. N - ie = O Дол‏ ر MEE еі aen,‏ ی Ara ¿ai Графа? ve RALES. 1 DU GRE C) D S ОО BESE xD í PLATE 29 135 PLATE 29 Figs. 1-6. Androvettia statenensis sp. no ig. 1. Transverse section PUES the База] pan of an unusually large phylloclad, X 30. Fig. 2. edian part of the transverse section shown in Fig. 3. Part of the woody cylinder of the transverse sec- tion shown in fig. 1, Pl. 29, X 180 Fig. 4. Longitudinal section of the wood of the specimen shown in fig. 8, Pl. 28, X 180 Fig. 5. Similar section of the wood of the specimen shown in figs. 1 and 3, Pl. 29, X 180. Fig. 6. Part of the section shown in fig. 5, Pl. 29, X 500. PAGE Mem. N Y. Бот. GARD. Vole LI Pie oO: AAT Big; to ae Sas De ca жән an TE vei [i ye tro TX (SAA AINE am Ml кафе TO, CRETACEOUS CONIFERALES. INDEX mes in black-face type represent new species or new combinations. Names in ce represent synonyms or species a) mentioned. . mbers in black-face type refer to pages on which. descriptions begin. Numbers in ale refer to pages of plate 4... Р де А ا‎ insignis Font., les, 67, 73 Culmites priscus Ettin sh., E Abietites angusticarpus Font, 69 i mia права, ROE , 44 Agathis, 48, 56-60, 74-7 Дои 59, 65, T alba (Rumph.) Salisb., ie nud 120 65, 778 australis (Lamb) Salisb., бугай туе1оп Albertia Czekanowskia P 5 Androvetia 22, 31, 33, 71, capillaris Newb., 63, 90 nsis sp. nov., 725, DE 92, 04, dichotoma Heer, 6 134, ps nervosa Heer, 63 fom pe e rigida Heer, 63 pida 50, 98 tuberculat i nov., 49, so, 98, 128, D Ааа 41, 48, 55- 59, . m 74-76 Dactolepi Zn Araucariopitys, Ра Tx r 75 НО зр. поу., 52, 08 байла, 4, 46, "indie Sesa 55, sn 58-60, 71, 72 borealis He veboracense sp. nov., 58, 120 O ы. Araucarites, 35,39 2 Davis, "WiLL м T. Mentioned E col- 5 DER lector of specimens, 3, 4, 41, 62, 69 Reichenbachi Gein., 38 Dunnican, Joun M. Ме Шол. as col- B lector of specimen, 37 ud 2. ы. 2 6 Ephedra И Brachyoxy on, 57-60, 68, 71-74, 7 j ile sp. nov. 54, 104, 106 | Eum 43, um pus pul tolus 4, 5 24, 25, m 2: 33-38, | proxima- SP: "nov, ое 40—42, 47, 2 64, 70, 75 F corallinum Heer, 64 crassum Lesq., и Feistmantelia, 17, 18 crassum Tennison- 34 oblonga Ward, 18 Delgadonum Heer, vir, 2. Font, 18 a Newb. 33, 34, 36-38, | Fren 41, 06, 100, 102, 104 P cen Reich Ettingsh., 30 sp- % Си of, 37, 0б, 100 22 Frenelop рзіз, 1. 27, 2 NATHANIEL L. Quot атасыз BRITTON, Hoheneggeri (Esingsh) Schenk, 26, ; 29 Cabomba inermis Е Hollick, 37 ae Ы; Callitris, 31 Geinitzia, 24, 25, 28, 38, 39, 41-43, 45, о E 46, 51, 70, D Cedrus, 1. 15, = Celastro vi d 2 marylandicum Font., 2 eichenbacni t com CHR trophy . Bibliography cited, 38, 39, 41, 42, 88, 94, 110, 112, Clathraria, 18 114 Colymbea, 56 , 42, 04, 114, 116 Cordaites, 15, 60 Gus, LE Jj. Report on amber, 11, Cryptomeria, 52 137 138 Gingko, 72 Gingkophyllum, 64 H Номлск, A. Bibliography cited, 5 Quoted, 2, 3, 14 Inolepis, 28, A Нет 28 Isoétes, 64 та 64 JEFFREY, Е. С. а 4, hypnoides Heer, 2h 62, 88 macilenta Hee Bibliography cited, 5 L Lar Lepiserbus folios rn. 19 talas Font., 19 M MATHER, WILLIAM W. Quo MEYER, . Mentioned i rendered, 12 Moriconia, 22, 23, 33 ted, 2 "assistance Omphalomela, 18 P bic 22 ıcho о. Don., 23 7 Pinites Solmsi Seward, 1 Pinus, 12-19, 38, P > 63, 67, 69, 72-76 edi ulis „Engelm аа % of, 17, 84, Cone p of, 16, 96, 124 Pityoidolepis, 3, 54 nsis nov., 53, 96, 132 a 5, Jeffrey er Chrysler, 20, Bi moh marylandica Font., 22 Podocarpus, 6 Prepinus, 19, ‚ 76 tenio. Jefes 19, 96, 122, 124, 12 Protodammara, 46, speciosa Hollick and Hen 46, 62, , 108, А 24, 26 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN Pseudo-Araucaria, 68, 4 Bu 45, 46, 70 rmis sp. nov., 45, 98, 728 Pseudolari "m p 73 suga, 67 Psilotum inerme Newb., 27 Raritania, 26-29, gracilis (Ne) ТЕЛЕН поу., 26, 00, 96, 98, 11 RIEs, HEINRICH. Mentioned as discoverer of specimens, Saxegoth a И Heer, Seque 2; 3846 Ei ё" 61, a 25 mbigua Coutisiae ed h : 4 Reichenbach (Ge ща Неег, 38, 39, шш Heen 42 Sphenaspis, 51, statenensis 2 nov., 51, 98, 130 Strobilites, “66, 68, 69 visii sp. nov., 68, 84 microsporopho orus sp. nov., 66, 98, 126 sp., 69, 84 Strobus, 13 Succinite, 11, 12 Thinnfeldia, 22-2 Thuja, 23, 32 Thuya таса (Heer) Newb., 31 Trichopitys, 6 Tsuga, 60, pe 73 U Ullmannia, 60 Unidentified Twig, 64, 96 У Voltzia, 52 W Widdringtonia, Widdringtonites, 29-5 2. ines) Bo 29, 31, 88, Z Zamiopsis insignis Font., 22 New York Botanical Garden 1 DN 5185 00292 371 3