QE 92k .K593 Knowlton, Frank Hall Contributions to the geology and paleontology of San Juan County, Nev Mexico !i DEPARTMENT OF THE INTERIOR Fkanklin K. Lane, Secretary UNITED STATES GEOLOGICAL SURVEY George Otis Smith, Director Professional Paper 98— S CONTRIBUTIONS TO THE GEOLOGY AND PALEONTOLOGY OF SAN JUAN COUNTY, NEW MEXICO 4. FLORA OF THE FRUITLAND AND KIRTLAND FORMATIONS BY F. H. KNOWLTON Published December 18. I9I6 Shorter contributions to general geology, 1916 (Pages 32T-353) WASHINGTON GOVERNMENT PRINTING OFFICE 1916 DEPARTMENT OF THE INTERIOR Franklin K. Lane, Secretary UNITED States Geological survey George Otis Smith, Director Professional Paper 98 — S CONTRIBUTIONS TO THE GEOLOGY AND PALEONTOLOGY OF SAN JUAN COUNTY, NEW MEXICO 4. FLORA OF THE FRUITLAND AND KIRTLAND FORMATIONS BY F. H. KNOWLTON Published December 18, 1916 Shorter contributions to general geology, 1916 ( Pages 327-353) WASHINGTON GOVERNMENT FEINTING OFFICE 1916 4. FLORA OF THE FRUITLAND AND KIRTI^VND FORMATIONS. By F. H. Knowlton. fflSTORICAL SUMMARY. San Juan County, in the extreme northwest corner of New Mexico, comprises but a small portion of the hirger more or less clearly defined structural area known as the San Juan Basm. Certain structin-al elements, particularly the upper series of coal-bearing rocks, form a more or less continiious rim aroimd this basin and dip toward the center, where, however, the}' are deeply buried. As the object of the present study is to as- certain the bearing of the fossil plants on the age of this series of coal-bearing and related rocks, it is desirable to give first a brief histor- ical setting for the geologic facts here set forth. It is not necessary in this connection, however, to go further back than the beginning of the time embraced within the history of the so- called "Laramie problem" — that is, to about 1S75. The San Juan district -^ns studied by W. H. Holmes,' of the Hayden Survey, during the field season of 1875. The results of his studies of the rocks in the vaUey of San Juan River are displayed m a generalized section in Plate XXXV of his report. The uppermost member of this section is referred to the Wasatch and is divided into two parts, the lower of which is the Puerco marls of Cope, now included in the Puerco and Torrejon formations. Immediately below the Puerco marls is the so-called " Upper coal group," made up of 800 feet of soft sand- stones and marls, which is called Laramie?. This in turn rests on massive sandstone 120 feet thick, caUed the Pictured Cliffs group, and on the evidence of invertebrates was referred to the Fox HlUs unit. Except for the preparation of a number of economic reports on the coal of this region, little systematic geologic work was done in the region until 1899, when Ci-oss ^ established the ' Holmes, W. H., Geological report on the .San Juan district: U. S. Geol, and C.eog. Survey Terr. Ninth Ann. Kept., for 1875, p. 241, 1877. .2 Cross, Whitman, U. S. Geol. Survey Geol. Atlas, La Plata folio (No 60), 1899. units of the Upper Cretaceous section, which subsequently have been so widely identified in Colorado, Wyoming, and Montana. These units, in ascending order, are the Mancos shale, Mesaverde formation, and Lewis shale. Con- cerning the Lewis shale Cross wrote as follows : Above the Mesaverde formation occurs another formation of clay shale, reaching an observed thickness of nearly 2,000 feet, •wliich is very much like the Mancos shale but contains fewer fossils. The only identifiable form thus far found in this shale occurs also in the Mancos shale, so that . this division is still apparently below true Fox Hills. This formation is called the Lewis shale. Continuing, Cross says : Still above the Lewis shale is a second series of sand- stones, shales, and clays, bearing some resemblance to the Mesaverde formation but differing in detail. The lowest member of tliis complex is the "Pictured Cliff sandstone " of Holmes's San Juan section, wliich he placed in the Fox Hills upon the e\ddence of invertebrate remains. The remainder was refeiTed to the Laramie, but without fossil evidence. The present survey has failed to bring t« light valid ground for assigning any of the beds in question to the Laramie, wliile there is some reason to believe that more than the lower sandstone belongs to the Montana group. In 1905 Schrader^ made a reconnaissance examination of the Durango-Gallup coal field in which he adopted the stratigraphic classifi- cation established by Cross and described the uppermost coal-bearing sandstone as Laramie ?, though he presented no fossil evidence. In 1906 more detailed examinations were made m the region by Taff and Shalcr. Taff * studied the Durango coal district, which lies just off the southern foothills of the San Juan and La Plata mountains, and in his report the upper coal-bearing rocks here under considera- tion were referred without question to the Lara- mie, though he gave no details as to the reason for this reference. Shaler's report •' deals with that part of the Diu-ango-GaUup field lying s Schrader, F. C, The Durango-Gallup coal field of Colorado and New Mexico: U. S. Geol. Survey Bull. 2So. pp. 241-258, 1906. < Taff, J. A. , The Durango coal district, Colo.: U. S. Geol. Suri^ey Bull 316, pp. 321-337, 1907. s Shaler, M. K., .\ reconnaissance survey of the western part of the Durango-Gallup coal field of Colorado: Idem, pp. 376-426. 327 328 SHORTER CONTRIBUTIONS TO GENERAL GEOLOGY, 1916. west of longitude 107° 30'. The beds here dis- cussed were also referred by Shaler to the Lara- mie without qualification, on the basis, it is stated, of fossils studied by T. W. vStanton and F. H. Knowlton, though the evidence is not presented in detail. On looking over the origi- nal reports on the plants collected by Shaler and others in this region, I find that they were few and fragmentary collections ordy, and the ten- dency was to regard them as mainly older than Laramie. The status of the "Laramie" in the region under consideration was so well summed up by Lee ' in 1912 that his statement is quoted entire as follows: The "Laramie" formation occurs within the area described in this paper only in the San Juan Basin. It is more than 1,000 feet thick in the southern rim of the basin but is thiimer in the eastern rim, probably due to post-Cretaceous erosion. At Dulce it is only 225 feet thick. The formation lies conformably on Lewis shale and probably for this reason more than any other has been called Laramie, although Dr. Cross several years ago called attention to the fact that investigation had ' ' failed to bring to light vaUd ground for assigning any of the beds in ques- tion to the Laramie, wliile there is some reason to believe that more than the lower sandstone belongs to the Mon- tana group." Since that time a considerable number of fossils, both of invertebrates and of plants, have been collected from these beds in the Durango region. The base of the formation^the Pictured Cliff sandstone — con- tains marine invertebrates, and the lower part of the coal-bearing rocks above this sandstone contains brackish- water invertebrates, several of which occiu- in the Mesa- verde of other fields. But liigher in the formation the rocks contain fresh-water invertebrates wluch Dr. Stanton regards as Laramie and fossil plants wliich Dr. Knowlton regards as older than Laramie. The fossil plants have been given in the table * * * and from this table as well as from the accompanying statement by Dr. Knowl- ton [see below] it will be seen that the flora differs but little from that of the Mesaverde farther to the south. The name "Laramie" is here used for this formation not because the writer wishes to argue for the Laramie age of the rocks, but because the name is in use and because in tliis paper the writer is intentionally avoiding the introduction of new names for rock formations. It must be noted, however, that while the formation is called "I^aramie " it contains a flora wliich denotes Montana age, having notliing in common with the Laramie flora of the Denver Basin. The statement by me to which Lee alludes is as follows : n Near Dulce, N. Mex., and near Durango, Colo., there have been obtained two collections of plants from above the Lewis shale in coal-bearing rocks that have been referred to the so-called "Laramie " of this region. These collections are very full and embrace a number of easily ' Lee, W. T., Stratigraphy of the coal fields of northern New Mexico: Geol. Soc. America Bull., vol. 23, pp. 607-fl08, 1912. recognized species; hence their identification is satis- factory and complete. These collections prove clearly that these beds do not belong to the Laramie, since, so far as known to the writer, not a single species there present has been found in beds of tliis age. On the other Iiand, the plants indicate beyond question that they belong to the Montana, there being, for instance, Ficus spedosissima, Ficus sp. (narrow, tliree-nerved type), Ficus sp., type of F. lanceolala, a palm, etc., wliich link them with the Mesaverde floras to tlie south and the beds aheady dis- cussed in the Raton Mesa region. Associated with these, however, and tending to give them a sUghtly higlier posi- tion, though still witliin the Montana, are such forms as Brachyophyllum, Cunninghamites, Cieinitzia, Sequoia, etc., all of wliich are beyond doubt Montana types not found in the Laramie. A number of collections were made by J. H. Gardner in the Ignacio quadrangle, east of Durango, Colo., from beds regarded as the Laramie of that area. The plants in these collections, almost species by species, are iden- tical with the forms from near Dulce and near Durango, and I have no hesitation in saying that they occupy the same stratigraphic position and are of the same age, viz, Montana. It now remains to consider certain dinosaur- bearing beds first reported near Ojo Alamo, N. Mex., which may have a bearmg on the "Laramie" of the San Juan Basin region. In 1908 James H. Gardner, then of the United States Geological Survey, found reptilian vertebrate remains near the head of Coal Creek, 1 mile southeast of Ojo Alamo, "in variegated sands, shales, and conglomerates, indisputably above the unconformity at the top of the Laramie.^ These remains were studied, by C. W. GUmore, who states that this famia "appears to represent a typical fauna of the so-called Laramie or better Ceratops beds." It appears that dinosaiu- remams had been known at the Ojo Alamo locality as early as 1902, but it was not imtil 1904 that a systematic attempt was made to collect them. In this year Barnmn Brown, of the American Museum of Natural History, made a reconnaissance trip to the locality and obtained "a small but interesting collection of fossils." These, however, were not described in print until 1910.' Concerning the stratigraphic relations of these beds Brown says: Less than a mile south of the store at Ojo Alamo the Puerco formation rests unconformably on a conglomerate - Gardner, J. H., in Knowlton, F. H., The stratigraphic relations and paleontology of the "Hell Creek beds," "Ceratops beds," and equiva- lents, and their reference to the Fort Union formation: Washington Acad. Sei. Proc, vol. 11, p. 32.3, 1909. 3 Brown, Barnum, The Cretaceous Ojo Alamo beds of New Mexico, with description of the dinosaur genus Kritosaurus: Am. Mus. Nat. Hist. Bull., vol. 28, pp. 267-274, 1910. FLORA OF THE FEUITLAND AND KIETLAND FORMATIONS. 329 that is composed of red, gray, yellow, and white pebbles. The position of these beds is below what may be called the type of the Puerco or basal Eocene. * * * Below the conglomerate there is a series of shales and sandstones, evenly stratified and usually horizontal, in wliich there is much less cross-bedding than commonly occurs in the Laramie of the northern United States. The shales below the conglomerates that contain numer- ous dinosaur and turtle remains I shall designate as the Ojo Alamo beds. They are estimated to be about 200 feet tliick, but owing to lack of time I was unable to deter- mine their relations to the underljing formations. Also in 1910 appeared a paper by Gardner ' that involved incidentally the beds vinder dis- cussion. He records the Puerco as resting luiconformably on the underlying beds, as Brown had reported, and he states that at Ojo Alamo he "obtained dinosaurs from beds unconformably above the ' Laramie ' and below the Wasatch." Tlaesebeds are, of course, the Ojo Alamo beds of Brown. From the preceding brief statements it appears that for more than 40 years the upper coal-bearLng rocks of the San Juan Basin have been regarded more or less definitely as of Laramie age, though during quite half of this tune there has been growing an undercurrent of opinion that the beds may be older than the Laramie. The object of the present study is to ascertain the bearing of the fossil plants on this question, and it may be stated here that the plants appear fully to sustain the opinion that the beds are older than the Laramie. The fossil plants upon which the present report is based were obtained by C. M. Bauer during his geologic studies of the season of 1915 in that portion of the San Juan Basin, embraced within San Juan County, N. Mex. The stratigraphic results of his investigations are set forth at length in the paper by Bauer,- and in the present connection it is necessary only to give them in brief outline. In that portion of the coal-bearing and associated rocks previously referred to as the "Laramie" of that region Bauer has recognized two htho- logic tmits. The lower of these units, called the Fruitland formation, is about 300 feet thick and includes the coal-bearing beds. Above it is a series of shales, called tlie Kirtland shale, with an intercalated sandstone, the Farmington sandstone member, and the whole formation has a maximmn thickness of nearlj^ 1,200 feet. 1 Gardner, J. H., The Puerco and Torrejou formations of the Naci- miento group: Jour. Geology, vol. 18, pp. 702-741, 1910. 2 U. S. Geol. Survey Prof. Paper 98, pp. 274-275 ( Prof . Paper 98-P). Above the Kirtland shale, without observed stratigraphic break, is the Ojo Alamo sandstone, which Bauer thinks shotild be grouped with the imderlying Kirtland and Fruitland formations. A graphic presentation of the older and newer views is given below: Older interpretat ion . Ojo Alamo. Unconformity- ■ Laramie. ' Pictured Cliffs. Newer interpretation. Ojo Alamo sandstone. Conformity Farmington sand- stone member. Shale. Fruitland formation. Pictured Cliffs sandstone. THE FLORA. The material on which the present report is based comprises 20 collections, of which 15 are from the Fruitland formation, 3 from the Kirtland shale (2 from the extreme top and 1 from the base of the formation), and only 1 from the Ojo Alamo sandstone. The bulk of the material comes from the lower or coal- bearing portion of the section, and much of this is preserved on a red baked shale, indi- cating proxunity to coal. The material in the single collection from the Ojo Alamo beds is so fragmentary that it can not be identified witn satisfaction. It includes portions of a large leaf of unknown affinity, a small willow-hke leaf, and a large leaf that appears to be an Araha of the type of Aralia nofata Lesquereux, a species very abundant and widely distributed in the Fort Union for- mation. Nothing like this has been noted in the underlymg beds, and to a certain extent it argues for the Tertiary age of the Ojo Alamo beds, though obviously the evidence is not strong. For the present, therefore, the dictum based on the evidence of the fossil vertebrates that these beds can not be separated from tlie 330 SHORTER CONTRIBUTIONS TO GENERAL GEOLOGY, 1916. underlying beds must be accepted, though the WTiter can not escape the impression that they may ultimately be shown to be of Tertiary age. Below is a complete list of the forms repre- sented in the collections from the Fruitlantl and Kirtland formations. Asplenium neomexicanum Knowlton, n. sp. Onoclea neomexicana Knowlton, n. sp. Anemia hesperia Knowlton, n. sp. Anemia sp. Sequoia reichenbachi (Geinitz i lleer. Sequoia obovata? Knowlton. Geinitzia formosa Heer. Sabal montana Knowlton. Sabal? sp. Myrica torreyi Lesquereux. Myrica? neomexicana Knowlton, n. sp. Salix baueri Knowlton, n. sp. Salix sp. a Knowlton. Quercus baueri Knowlton, n. sp. Ficus baueri Knowlton, n. sp. Ficus curta? Knowlton. Ficus prsetrinervis Knowlton. Ficus leei Knowlton. Ficus prselatifoUa Knowlton, n. sp. Ficus sp. Ficus rhamnoides Knowlton. Ficus squarrosa Knowlton. Ficus sp. Ficus eucalyptifolia? Knowlton. Laurus baueri Knowlton, n. sp. Latirus coloradensis Knowlton. Nelumbo sp. Heteranthera cretacea Knowlton, n. sp. Pistia corrugata Lesquereux. Leguminosites? neomexicana Knowlton, n. sp. Pterospermites undulatus Knowlton. Pterospermites neomexicanus Knowlton, n. sp. Pterospermites sp. Ribes neomexicana Knowlton, n. sp. Carpites baueri Knowlton, n. sp. ^ PhylUtes petiolatus Knowlton, n. sp. PhylUtes neomexicanu.s Knowlton, n. sp. Unassigned plant (a). Unassigned plant (b). The above list comprises 40 forms, of which 6 are so fragmentary that they have not been given specific names, 2 have not been assigned, even gcnerically, 16 are regarded as new to science, leaving 16 species knowTi previously in other areas. Although new species as such have little value in fixing the age of the rocks in which they occur, it not infrequently happens that impor- tant and far-reaching conclusions may be drawn from a consideration of their obviously close relation with species whose stratigraplaic relations are known. Thus the species de- scribed as Asplenium, nco-mexicanum is not to be distinguishetl from the fragment described as Asplenium sp. Knowlton, from the Mcsa- verde of Dutton Creek, Laramie Plains, Wyo. Onoclea neomexicana belongs to a very long lived type, which, it was previously supi)osed, began in the Fort Union and is still hvmg. The present form carries the tyjie much Turther back. Anemia hesperia is not closely related to any previously described fossil species from this country. Myrica ? neomexicana is so poorly reprbscnted that its generic reference has been questioned, and hence its relationship is obscured. The specimen of Salix haueri is also poorly preserved as regards nervation and is of little value in fixing its affinity. Quercus haueri is most closely related to an unpublished species front the Vermejo formation of southern Colorado, differing in its slightly smaller size and less prominent teeth. Ficus haueri is, in a way, of the tj^je of Ficus de7weriana, a large Denver species, but it differs veiy markedly in nervation and is not considered to be closely related to that species. Ficus prselatifoUa is most nearly related to Ficus planicostata lati- folia, a form first made known from Black Buttes, Wyo., but later recorded from a num- ber of horizons, including Montana and Lara- mie. Laurus haueri is of the tj-pe of Laurus socialis, a Tertiary form, but differs in nerva- tion as well as in size. Heteranthera cretacea belongs to a living tv'pe that has not before been detected in a fossil state. The single minute leaflet described under the name Legu- minosites f neomexicana is so small and obscure that its affinity can not be estabhshed. Ptero- spermites neomexicanus is probably most closely related to Pterospermites undulata of Point of Rocks, Wyo. Riljes neomexicana is without known close relations among fossil forms. The species of Carpites and Phyllitcs have been des- ignated by these nondescript names because they are without recognized affiliations. The species common to the Fruitland and Ku'tland formations and the Laramie formation of the Denver Basin are Sequoia reichenbachi, Sahal montana, Myrica torreyi, and Ficus prse- trinervis. In working up the flora of the Laramie of the Denver Basm of Colorado it has been found that this flora embraces ap- proximately 125 species, of which nine are known to be common to the Montana. As all four of the above-named species are mduded in tlie nme common to the Laramie and Mon- FLORA OF THE FRUITLAND AND KIETLAND FORMATIONS. 331 tana, it follows they can not be used as an argu- ment for the Laramie age of the Fruitland and Kirtland formations, because mthout col- lateral data it woidd be impossible to decide their age in the San Juan Basin. As a matter of fact, while these four species are known to occur in the Laramie, their principal distribu- tion is in beds older than Laramie. Thus, Sequoia reichenbachi is known to range from the upper part of the Jui-assic entirely thi-ough the Cretaceous, bemg perhaps most abundant, at least as regards mdividuals, hi the middle Mon- tana. Sabal montana is the prmcipal species of palm in the Montana. Myrica torreyi was de- scribed originally from Black Buttes, Wyo., but it has since been demonstrated to be most abimdant and widely distributed m the Mon- tana. Finis prsetrinervis was &"st fomid in the Vermejo formation of Colorado and New Mexico, where at certain locahties it is exceed- ingly abmidant. The species common to the Fruitland and -Kirtland formations and the Montana group ai'e as follows: *AspIenium neomexicanum. *tSequoia reichenbarhi. *tSeiuoia obovata? *tGeiiiitzia formosa. *tSabal montana. *tMyrica torreyi. fSalix sp. a. fFicus curta? ■ fFicus prsetrinervis. fFicus leei. *Ficiis rhamnoides. *Ficus squarrosa. fFirus eucah-ptifolia? fLaunis coloradensis. *Pistia corrugata. *tPterospermites undulatus. The species in the above list that are marked with an asterisk (*) are fomid also m the Mon- tana of Wyoming, mainly at Point of Rocks; those marked with a dagger (f) are fomid in the Vermejo of Colorado and New Mexico. Of the 40 forms making up the known flora of the Fruitland and Eartland fonnations, 16 have been found m other areas, and the above list brings out the fact that no less than 15 of these forms are known to occm* m the Montana. A further analysis of the hst shows that 12 of the 15 forms occur in the Vermejo formation of Colorado and New Mexico, 10 occur in the Mesaverde, or rooks of about this age, in Wyoming and elsewhere, and 6 species are com- mon to both these areas. On the basis of this showmg the conclusion seems justified, there- fore, that the Fruitland and Kirtland fonma- tions are of Montana age. Fanuly POLYPODIA-CEa;. Asplenium neomexicanum Knowlton, n. sp. Plate LXXXIV, figures 5-9. Asplenium sp. Knowlton, V. S. Geol. Survey Bull. 163, p. 20, pi. 3, fig. 11, 1900. Frond at least firm in textme; outhne of whole frond not kno^vn but apparently simple or once forked, base abruptly rounded and slightly cordate, apex abruptly narrowed to an actmunate point; margin finely toothed, the teeth small and apparentlj" spinose; stipe slen- der, its length not known but at least 6 centi- meters; nervation very distinct, consisting of a rather strong midvein and nimierous veins at an angle of emergence of about 45°, usually forking at or very near the base and occasion- ally once above, the veins or nearly all of them entering the teeth; sori long, narrowly linear, attached to the upper side of the veins. This very interesting species is represented by about a dozen fragments, five of which are here figured. Although none is of sufficient completeness to show the outline of the whole frond, they include parts that give what seems to be a fairly complete knowledge of its appear- ance. Two examples (figs. 8 and 9) show the configiiration of the base and a portion of the slender stipe, and two (figs. 5 and 6) show the apical portion, figure 5 being particularly com- plete to the tip. It appears that as a rule the rachis forks at an angle of approximately 45° a very short distance above the base of the frond, producing two broad, probably short, obtusely pointed lobes. In the specimen shown m figme 8 (right-hand leaf), however, there is no evidence of forking within its pre- served length -of 5 centinictei-s, and whether it was forked at a stiU higher point or was entire can not be determined. The marginal teeth may be noted at a nmnber of points in the speci- mens figured but ai-e especially distinct in figure 6. The nervation is very distinct and is well shown in all the specimens. Most of the nerves fork at the base and some of them also above the middle, but here and there one may" be noted that is simple and imforked tlu-ough- out. Mly its distinctness or identity. Occurrence: Kirtland shale, 3 inches below base of Ojo Alamo sandstone, IJ miles east- northeast of Pina Veta China, San Juan County, N. Mex. Lot 40 (6966). Family SCHIZIACE.E. Anemia hesperia Knowlton. n. sp. Plate LXXXIV, figure 3. Fronds presumably dimorphous; sterile frond roughly deltoid, 2.5 centimeters long, 3 centi- meters broad, dipinnate, the racliis slender, pinnae tliree or four pau-s, alternate, lanceolate, decreasing toward the apex; pinnules confluent, cuneate, erose-dentate at apex; nerves few, slender, at an acute angle, once or twice forked; fertile frond not known. 1 Newberry, J. S., New York Lyceum Nat. Hist. Armals, vol. 9, p. 30, 1868; U. S. Geol. Sur\-ey Mon. 35, d. 8, pl.23,fig. 3; pi. 24, flgs. 1-5, 1898. FLORA OF THE FRUITLAXD AXD KIETLAND FORMAIIO-N'S. 333 The little specimen figured is all that was found of this form. It represents the upper portion of the frond, but whether it is tlie whole frond or a mere fragment can not be determined. From its apparent affinities it seems probable that the species was dimorphous, and tliat this specimen is only the sterde portion, the fertile frond remaining unknown. Anemia hesperia finds its closest affinity with the hving species of the group including Ajic- mia wrightii Baker, Aneynia cicutaria Kunze, and Anemia cuneafa Kunze. These are all natives of Cuba and are found growing in crevices of rocks along shaded rivers. Occurrence: Fruitland formation, about 10 mdes south of Jewett and 2J mUes east of Chaco Kiver, San Juan County, N. Mex. Lot 14 (6947). Anemia sp. Plate LXXXIV, figure 4. The genus Anemia is very widely distributed both areally and vertically, but it rarely hap- pens that specimens arc well enough preserved to convey any adequate knowledge of the whole frond. The present material is so frag- mentary that it does not admit of satisfactory identification. The specmien figured is suffi- cient to show clearly that it belongs to this genus, but beyond tUat no judgment can be ventured. It was evidently a large species, with pinnaj cut deeply mto deltoid, very fuiely toothed lobes. The nervation is of the usual type. Occurrence: Fruitland fonnation, 30 miles south of Farmington and 1 mile east of reser- vation fine, San Juan Countv, N. Mex. Lot 76 (6953). Family PINACEiE. Sequoia reichenbachi (Geinitz) Heer. Araucarites reichenbaohi Geinitz, Charakteristik der Schich- ten und Petrefacten des sachsisch-bohmischeu Kreidegebirges, pt. 3, p. 98, pi. 24, fig. 4, 1842. Sequoia reichenbachi (Geinitz) Heer, Flora fossilis arctica, vol. 1, p. 83, pi. 43, figs. Id, ab, 5a, 1868. Lesquereux, U. S. Geol. Survey Terr. Rapt., vol. 6, p. 51, pi. 1, figs. 10-lOb, 1874; U. S. Geol. Sun-ev Men. 17, p. 35, pi. 2, fig. 4, 1892. The collections contain a few poorly pre- served examples that are merely of sufficient value to indicate the presence of this widely spread species. Occurrence: Fruitland formation, 30 nules south of Farmington and 1 nule east of reser- 55178°— 16 2 vation line, San Juan County, N. Mex. In cHnker above highest coal. Lot 75 (6956). Sequoia obovata? Knowlton. Sequoia obovata Knowlton, U. S. Geol. Survey Prof. Paper 101, p. — , pi. 30, fig. 7 (in press). Sequoia brevifolia Heer, Lesquereux, U. S. Geol. and Geog. Survey Terr. Bull., vol. 1, p. 365 [1876]; Ann. Kept, for 1874, p. 298 [1876]; Tenth Ann. Kept., for 1876, p. 500 [1878]; U. S. Geol. Survey Terr. Rept., vol. 7, p. 78, pi. 61, figs. 25-27, 1878. Knowlton, U. S. Geol. Survey Bull. 163, p. 27, pi. 4, figs. 1-4, 1900. In the present collections from the San Juan Basm is a single small, poorly preserved speci- men that is clearly a Sequoia and from the few leaves retained appears to belong to this species. Tliat tliis may be so is rendered likely by the fact that characteristic and well- identified specunens have previously been coUected in tliis same region. Occurrence: Fruitland formation. Hunters Wash, 30 miles south of Farmington and 1 mile east cff reservation line, San Juan County, N. Mex. Lot 78 (6952). Geinitzia formosa Heer. Plate LXXXV. figure 3. Geinitzia formosa Heer, Rreideflora von Quedlinburg: Schweiz. Gesell. Neue Denkschr., vol. 24. p. 6, pi. 1, fig. 9; pi. 2, 1871. Newberry, U. S. Geol. Survey Mon. 26. p. 51, pi. 9, fig. 9, 1876. Knowlton, U. S. Geol. Smvey Bull. 163, p. 28, pi. 5, figs. 1-2, 1900; U. S. Geol. Sun-ey Prof. Paper 101, p. — , pi. 31, figs. 1-3 (in press). Tlie figured specimen is a fragment of a branch of considerable size showing leaf bases and traces of the leaves. It is of somewhat doubtfid vafidity in tins comiection. Occm'rence: Fruitland formation, 17 miles south of San Juan River and 2 miles east of Rio Chaco, San Juan County, N. Mex. Lot 26 (6949). Family PONTEDERIACE.S;. Heteranthera cretacea Knowlton, n. sp. Plate LXXXV, figure 5. Leaf evidently tliick in texture, elliptical or very shghtly ovate-eUiptical, abruptly nar- rowed or rounded to a very narrow, short basal portion, rounded and slightly pointed at apex; three nerves arise in the basal portion of the blade or petiole, one passing up the center of the blade and the other two divichng the space 334 SHORTER CONTRIBUTIONS TO GENERAL GEOLOGY, 1916. between the first and the margin; from tlicse three arise several others of equal strength, there being altogether 10 or 12, close, parallel, and all curving around to and apparently entering the tip; no other nervation discernible. This curious little leaf is absolutely perfect except a minute portion of the tip. It is about 22 mUlimeters long includmg the narrow basal portion or petiole, which is 3 millimeters in length and 15 miUimeters wide. It is almost elliptical, though it is perhaps 1 miUimeter broader in the lower portion than in the upper. There are 11 or 12 nerves, only three of which arise in the basal portion, the others arising from them and aU running into the tip. The genus Heteranthera is a small one, comprising about nine species, two of which occur in tropical Africa and the others in America. Only three species are found ui the United States. They are herbs growing in mud or shallow water, with creepmg, ascending, or floating stems and petioled leaves which may be cordate, ovate, oval, reniform, or even grasslike. Of the tlu'ee United States species the one most smiQar to the present form is Heteranthera limosa (Swartz) WUldenow, the smaller mud plantain, which ranges from Vir- gmia to Kentucky and Missouri, south to Florida and Louisiana, and thence throughout tropical America. The living species bears numerous oval or ovate leaves 1.5 to 2.5 centi- meters long on petioles 5 to 12 centmieters long. The several nerves all arise from or near the top of the petiole and arch around to the tip. It win be noted that the fossil fonn agrees closely with this livhig species, the leaf bemg more nearly elliptical or ovate-eUiptical and more abruptly pointed at the apex. The nerves as they pass from base to apex are very similar in both forms, but m the fossil leaf they do not all arise from the top of the petiole, as apparently they do in the living species. It is believed that the generic reference can hardly be cjuestioned. This species undoubtedly resembles and indeed may be identical with a little leaf from Point of Rocks, Wyo., which was referred by Lesquereux * to Lemna scutata Dawson and which I afterward,- probably incorrectly, re- 1 Lesquereux, Leo, U. S. GeoL Sun-ey Terr. Rept., vol. 7, p. 102, pi. 6L flg. 5 [not flg. 31, 1878. 2 Knowlton, F. H., U, S. Geol. Survey Bull. Iii3, p. 31, 1900. garded as merely a small leaf of Pistia corrugata Lesquereux. The leaf from Point of Eocks is much smaller thiin the leaf here described and is more nearly circidar, but the "petiole" and the disposition of the nerves is much the same m both. It is certainly clear that the present leaf, as wcU as the one from Pomt of Rocks, is not the same as the type specimens of Dawson's Lemna scutata, and it is also reasonably certain that the present leaf can not belong to the genus Lemna. Occurrence: Fruitland formation, Coal Creek, 35 mUes south of Farmington and 1 mQe east of reservation line, San Juan County, N.Mex. Lot 8 1(6955). Family ARACE.ffi. Pistia corrugata Lesquereux. Plale LXXXV, fi,5ure 4. Pistia corrugata Lesquereux, V. S. Geol. and Geog. Survey Terr. Ann. Rept. for 1874, p. 299 [1876]; U. S. Geoi. Survey Terr. Rept., vol. 7, p. 103, pi. (51, fig?. 1. 3, 4, 6, 7, 9-11, 187S. Knowlton, U. S. Geol. Survey BulL 103, p. 31, 1900. One of the collections from the San Juan Basin contains a single example with its coun- terpart that undoubtedly belongs to Pistia cor- rugata as described by Lesciuereux from speci- mens collected in the Montana group at Pomt of Rocks, Wyo. As may be seen from the fig- ure, it is considerably broken and adds httle or nothing to our knowledge of the species. It is about the same as the specimen shown in figure 7 of Lescjuereux's plate in volume 7 of the Hayden Survey reports, in that it appears to be attached to the side of the thick stem, though it is perhaps really terminal and has been distorted in position dm-uig entombment. There appears to be a mass of rootlets by the side of the base, but these are so matted and compressed that their character can not be made out. The nervation is the same as that figured by Lesc^uereux, namely, an mdetermi- nate number of veins arising in the base of the bliide and spreading out and variously anasto- mosing above, producing very irregular polyg- onal meshes. Pistia corrugata is said to be very abundant at Point of Rocks, Wyo., Lesquereux describ- ing it as "covering by itself only large surfaces of shale." It was also found bi beds of similar age at Superior, Wyo., by Schultz in 1907, hi beds believed to be of Mesaverde age m the FLORA OF THE FRUITLAND AND KIRTLAND FORMATIONS. 335 Wind River Basin by Woodruflf in 1909, and in the Jiulitli River formation along Milk River, Mont., by Pepperberg in the same year. The questionable reference to tliis species of the specimens collected in 190S near jilesua Moun- tain, N. Mex., by Gardner is now rendered more probably correct by the evidence of the present specmien. The genus Pistia has had a very interesting geologic history. It is represented in the liv. uig flora by a smgle variable and widely dis- tributed species {Pistia stratioles Lmne). It is mainly tropical, ranging in tliis country from Florida to Texas and thence tlnough the West Indies, Mexico, and Central America to Para- guay and Argentina. In Africa it ranges from Natal to Senegambia and Nubia, and thence to Madagascar and the Mascarene Islands. In Asia it is found throughout the East Indies and thence to the Pliilippines. Of the four or five fossil species recognized, thi-ee are found m North America. The oldest of these is Pistia nordenslioldi (Heer) Berry,' which occurs in the Magothy fonnation of Maryland and very abimdantly in the Black Creek formation (Turonian) of North CaroHna. It was first described by Heer = under the name Chondrophyllum nordenskioldi, from specimens found in the Atane beds of Greenland. The next species in point of age is Pistia cor- rugata Lesquereux, wliich has already been dis- cussed. The youngest American fossil form is Pistia daibornensis, from the Claiborne group (nud- dle Eocene) of Georgia, recently described by Berry.' Tliis differs markedly from the other forms by its obovate shape and strongly rotuse apex. The only accepted European fossil species is Pistia mazeli Saporta and Marion/ which occm-s in the lignites of Fm-vean (Provence), France, and is of approximately the same age as Pistia corrugata. Pistia mazeli appears to be most closely related to the living form; in fact in the figures of the two species given side l^y side by Saporta and Marion it is ahnost uupossible to 1 Berry, E. W., Torrey Bot. Cluh Bull., vol. 37, p. 189, pi. 21, flgs. 1-1.5, 1910. 2 Heer, Oswald, Flora fossilis arctica, vol. 3, p. 114, pi. 30, fig. 4b; pi. 32, flgs. 11,12, 1S74. » Berry, E. W., U. S. Geol. Survey Prof. Paper S4, p. 137, pi. 26, flgs. 1-2, 1914. * Saporta Gaston, and Marion, A. F., L't^vohitiou du r&'HJe v<5g6tale, Phandrogames, vol. 2, p. 37, flgs. 114c, 114d, 1S.S5. note marks of distinction. In some ways it appears that the Claiborne species {Pistia daibornensis) is most closely related to Pistia mazdi, except that it is more retuse at the apex. However, its full character has not yet been made out, as it still depends on a single example in whicli little of the nervation has been retained. The other two species {Pistia corrugata and Pistia nordenshioldi) are themselves closely related and are also close to the Uving species, the main difference being their larger size and more anastomosing nerves. The significant point brought out by this cUscussion is the fact that tliis peculiar plant, now so widely distributed over both hemi- spheres, was estabhshed in essentially its present form in late Cretaceous time, and even then occun-ed in both hemispheres. It is rather remarkable that so little has been ascertained regarchng its Cenozoic history, in view of the great niunber of locahties that have been investigated, yet the evidence that it must have persisted with very little change is brought out by the above exposition. Occurrence: Kirtland shale, 1\ miles north- east of Pina Veta China, San Juan County, N. Mex. Lot 96 (6065). Family PALMACEiE. Sabal montana? Knowlton. Plate LXXXV, figure 2. Sabal montana Knowlton, IT. S. Geol. Survey Prof. Paper 101, p. — , pi. 32, fig. 3 (in jwess). Sabalit.fs yrai/anus (Lesquereux) Lesquereux, IT. S. Geol. Survey Terr. Rept., vol. 7, p. 112, pi. 12, fig. 1 [not pi. 12, fig. 2], 1878. Most fossil palms an* very difficult to identify and as a consequence are more or less unsatis- factory as stratigraphic criteria. The leaves are commonly of very large size and it is dif- ficult to prociu'c specimens that are anywhere near perfect, essential characters being in many specimens lacking or obscure. A specimen perhaps no larg(*r than one's hand from a leaf that was possibly 5 or 6 hvt in diameter can not fail to convey a very inadequate idea of its character, particularly the range in mdividual variation, and it is not at all improbable that too many species have been established. The specimen under considei-ation is a case in pomi. It isi a fragment, only about 10 centi- 336 SHORTER CONTRIBUTIONS TO GENERAL GEOLOGY, 1016. meters long, from some portion of th(! blade showing the petiole, or its prolongation, and the attachment of numerous rays. It appears to be from the under side i)f the leaf, but it is impossible to detemiine the huigth of this pro- longation of the petiole or to estimate the num- ber of rays with anj- degi-ee of accuracy. It seems to belong to what has been named Sahal montana, as based on a considerable number of examples rather widt^ly scattered through the several Montana localities, but on the other hand it is hardly to be distinguished from Geonomites ungeri Lesquereux,' a species sup- posed to bo confined to the Raton and allied Eocene formations. Geonomites ungeri was established on a mere fragment from the middle of a leaf that was obviously of considerable size, and there is no means of knowbig tlie size and configuration of the petiole, the numbiu- of rays, etc., and consequently there is no way to com- pare it with leaves referred to Sabal montana, in wliich these features are known. However, imtil more defijiite information is available the San Juan specimen may stand as Sahal montana f . Occurrence: Fruitland formation, 18 miles south of San Juan River and 4 miles east of Chaco River, San Juan County, N. Mex. Lot 28 (6961). Sabal? sp. Plate LXXXV, figure 1. The specimen here figured is the only one of its kind observed in the collections. It is a mere fragment from what apparently was a very large leaf, but it lacks so many essential features that it has seemed unwise to give it a specific name. When compared with the specimen figured as Sabal montana? it is seen to be very difl^erent, the pj-olongation of the racliis being especially strong and the rays very large at the point of their attaclmient. But the size of the leaf, the length of the rachis, and the munber and configuration of the rays are un- known, and this fragment simply serves to call attention to the presence in these beds of a large palm. OccuiTcnco: Fmitland formation, just across San Juan River from Fruitland, aljout half a mile above the bridge, San Jiiaii County, N. Mex. Lot 3 (6957). 1 Lesquerexix, Leo, U. S. GeoL Survey Terr. Rept., vol. 7, p. lis, pi. 11, fig. 2, 1878. Family MYRICACEffi. Myrica torreyi Lesquereux. Plate LXXXVI, figure 1. Myrica torreyi Lesquereux, U. S. Geol. and Geog. Survey Terr. Sixth Ann. Rept., for 1872, p. 392 [1873]; U. S. Geol. Survey Terr. Rept., vol. 7, p. 129, pi. 16, figs. 3-10, 1878. Ward, U. S. Geol. Survey Sixth Ann. Rept., p. 551, pi. 40, fig. 4, 1886; Bull. 37, p. 32, pi. 14, fig. 5, 1887. Kuowlton, U. S. Geol. Survey Bull. 163, p. ,34, pi. G, figs. 1-3, 1900. Cockerell, Colorado Univ. Studies, vol. 7, p. 150, 1910. Only a few fragments of this species were found, but it is so well marked that these are sufficient to attest its presence in these beds. Occurrence: Fruitland formation, 30 miles south of Farmington and 1 mile east of reserva- tion line, San Juan County, N. Mex. Lot 75 (6956). Myrica? neomexicana Knowlton, n. sp. Plate LXXXVI, figures 2-1. Leaves small, deltoid in general outUne, apparently truncate at the base and obtusely pointed at the apex; with at least 9 and probably as many as 11 or 13 lobes, the basal pair cut deeply or nearly to the midrib, be- conung almost separate, others only shghtly cut and in upper portion probably reduced to merely strong undulations; lobes obtuse, entire or more commonly strongly and obtusely toothed; nervation pinnate, the michib being relatively strong; secondaries as many as the lobes, at a low angle of emergence, craspedo- dronie, ending in the lobes; finer nervation abundant and very irregular. Although there are several specimens that obviously belong together they are so frag- mentary that the species is very inadequately represented. It is roughly triangular or del- toid, and was at least 4 centuncters in length and very probably was considerably longer. The width at base was certainly over 4 centi- meters, and if there was an adcUtional free or nearly free lobe, its width must have been 6 centimeters or more. Three specimens have been figured. In that shown in figure 3 the lobes are nearly or quite entire, but in the somewhat larger leaf shown in figure 2 the lobes are irregularly and obtusely toothed. The specimen shown in figure 4 is evidently the basal lobe of the leaf, but M'hcther it belonged to a larger leaf FLORA OF THE FRUITLAND AND KIRTLAND FORMATIONS. 337 or is another nearly detached lobe below the one shown in figure 2 is of course quite im- possible to determine. Except for its close resemblance to the lower lobe like that in figm-e 2, tliis specimen might be considered an individual leaf, but it is broken on the upper side near the base, where it was un- doubtedly connected to the other part of the blade. Owing to the fragmentary nature of these specmiens this generic reference is more or less uncertain. In some respects they are of the character of certain species of Myrica, such, for instance, as Mynca alkalina Lesque- reux,' from the Green River formation of Wyoming, but they differ markedly in many essential particulars. Occurrence: Kirtland shale, 3 inches below base of Ojo Alamo sandstone, 1\ miles north- east of Pina Veta Cliina, San Juan County, N. Mex. Lot 40 (6966) . Family SALICACEiE. Salix baueri Enowlton, n. sp. Plate LXXXVI, figures 7, 8. Leaves small, of firm texture, narrowly ellip- tical-lanceolate, about equally narrowed to both base and apex; margin entire; midrib rela- tively very thick; other nervation obscure or wanting. This little species is represented in the col- lections by some half dozen specimens, three of which are here figured. They are small leaves 3 or 4 centimeters in length and about 14 millimeters in greatest width. They are pre- served on a very coarse grained matrix which has obscured or obhterated nearly aU traces of nervation except the very thick midrib. An occasional secondary appears to be at an angle of about 45° and much cm^ed upward near the margin. These little leaves are so obscurely preserved that their fuU character can not be made out, and they are consequently of comparatively little stratigraphic value. They are of about the same size and shape as specimens of Salix Integra Goppert from Black Buttes, Wyo., as figured by Lesqucreux,^ but the absence of iLesquereux, Leo, U. S. Geol. Survey Terr. Rept.,vol. 8, p. 149, pi. 45 A, figs. 10-15, 1883. 2 Idem, vol. 7, p. 167, pi. 21, flgs. 1, 2, 1878. most of the nervation makes it impossible to compare them satisfactorily with this species. Occurrence: Fruitland formation, 10 mUes south of San Juan River and 4 mUcs east of Chaco River, San Juan Comity, N. Mex. Lot 16 (6958). Salix sp. a Knowlton. Plate LXXXVI, figure 9. Salix sp. a Knowlton, U. S. Geol. Survey Prof. Paper 101, p. — , pi. 38, fig. 1 (in press). The collection contams a number of willow leaves that are indistinguishable from a form described and figured from the Vermejo for- mation of the Raton Mesa region. The leaf figm'ed here, which is one of the best, is about 8 centimeters in length and 1.8 centimeters in width. It has a very thick midrib and only faint indications of secondaries, which appear to be close, parallel, and at an angle of about 45°. It is so indistmctly preserved that it has not been thought desu-able to give it a specific name. Occurrence: Fi-uitland formation, 10 miles south of San Juan River and 4 mdes east of Chaco River, San Juan County, N. Mex. Lot 16 (6958). Family FAGACE.ffi. Quercus baueri Knowlton, n. sp. Plate LXXXVI, figures 5, 6. Leaf small, of coriaceous texture, ovate- elliptical, about equally rounded to both base and ape.x; margm provided with relatively strong, rather obtuse teeth; micbib veiy strong; secondaries about four pairs, strong, alternate, craspedodrome, terminating in the teeth; finer nervation obscure. The specimen here figured is nearly perfect and is 3.5 centimeters long and 2 centimeters wide. It was evidently a rather thick and coriaceous leaf, as is attested both by its thick nervation and by its general appearance. This species has some resemblance to an mi- pubhshed species of Quercus from the Vermejo formation but differs in its shghtly smaller size and less prominent teeth. Occmrence: Fruitland formation, from chnker above highest coal bed, 30 miles south of Farmington and 1 nfile east of reservation line, San Juan Basin, N. Mex. Lot 75 (6956). 338 SHORTER CONTRIBUTIONS TO GENERAL GEOLOGY, 1916. Family MORACE^. Ficus baueri Knowlton, n. sp. PLate LXXXIX, figure 2. Leaves of medium size and firm texture, ovate, rather abruptly rounded to the appar ently tnmcate base, acuminate at apex; mid- rib strong, straight; secondaries three or four pairs, very remote, alternate, at an angle of 45°, strong below, much thimier and almost disappearing above, probably camptodrome; norvilles thin, very obscure, oblique to the midrib; fuier nervation not i-etained. The specimen figiu-ed, although it lacks a considerable portion of the leaf, is sufficient to give a good idea of this species. It is very regularly ovate, about 115 centimeters in length and 7 centimeters in greatest width, which is .above the middle of the blade. It is remarkable for its strong midrib and its few, remote, alternate secondaries. This species is in a way of the tyjie of Ficus denveriuna Cockcrell,' a weU-known Denver species. It differs markedly in nervation, however, as it has only three or four pairs of very remote secondaries, while the Denver form has not less than ten or twelve pains of relatively close parallel secondaries. The man- ner in which the secondaries reach the margin is also very different in the two fonns, and hence they can not be considered as at all closely related. Occurrence: Fruitland formation, in cUnker above highest coal bed, 30 miles south of Farmington and 1 mile east of reservation line, San Juan Basin, N. Mex. Lot 75 (6956). Ficus curta? Knowlton. Plate LXXXVIII, figure 3. Ficus curta Knowlton, U. S. Geol. Survey Prof. Paper 101, p. — , pi. 42, fig. 5 (in press). The typo of tliis species comes from the Vermejo formation of the Canon City coal field, Colorado, and is described and figm-ed in the report cited. The specimen here figured lacks most of the margin, but so far as can be made out it appears to agree with the leaf from Colorado. Occurrence: Fruitland formation, 30 miles south of Fannington and 1 mile east of reser- 1 A new name for Ficus spectabilis Lesquereux, which proved to be preoccupied. See U. S. Geol. Survey Terr. Kept., vol. 7, pi. 33, figs. 4-0, 1878. vation line, San Juan County, N. Mex. Lot 75 (6956). Ficus prjBtrinervis Knowlton. Ficus pixtrinervis Knowlton, U. S. Geol. Survey Prof. Paper 101, p. — , pi. 41, figs. 1-4; pi. 42, fig. 1 (in pres.s). The types of this species come from the Vermejo formation of the Raton Mesa region of Colorado and New Mexico and are described and figured in the report cited. It is a well- marked and exceedingly abundant form in the Vermejo formation and appears to be also abundant and well defined in the San Juan Basin. Occurrence: Fruitland formation, 30 miles south of Fannington and 1 mile east of reserva- tion fine, San Juan County, N. Mex. Lots 75 (6956) and 78 (6952). f^cus leei Knowlton. Plate XC, figure 2. Ficus leei Knowlton, U. S. Geol. Siu^'ey Prof. Paper 101, p. — , pi. 39, figs. 1-6; pi. 40, figs. 1, 2 (in press). The types of this fine species come from the Vermejo formation of the Raton Mesa region of Colorado and New Mexico and are described and figured in the report cited. It is one of the most abundant and well-marked forms of the Vermejo formation. Ficus leei is also fairly abundant in the San Juan Basin, though the leaves are somewhat smaller than the average size of those from the type area, but they do not otherwise differ. Occurrence: Fruitland fomiation, Amariilo Canyon, 10 miles south of San Juan River and 4 miles east of Chaco River, N. Mex., lot 16 (6958); 30 miles south of San Juan River and 4 2 miles east of reservation line, San Juan County, N. Mex., lot 74 (6963). Kirtland shale; 14 miles northeast of Pina Vcta China, San Juan County, N. Mex., lot 40 (6966). Ficus prslatifolia Knowlton, n. sp. Plate LXXXVII, figure 4. Leaves large, broadly ovate, truncate or slightly heart-shaped at base, probably obtuse above; nervation strongly three-ribbed from the top of the petiole, the midrib stronger, with several pairs of subopposite secondaries high up above the base; lateral ribs with six or FLORA OF THE FKUITLAND AXD KIRTLAND FORMATIONS. 339 seven secondary camptodronie branches on the outside; nervilles numerous, strong, mostly unbroken. This species is represented by a niunber of examples, one of the best of which is figurecl. It was probably not less than 13 or 14 centi- meters in length and about 9 centimeters in width, but the exact length can only be inferred. It was conspicuously longer than broad. The strong, three-ribbed nervation and other details are wcU shown in the figure. This fonn appears to be most closely related to Fims planicostata latifolia Lesquereux ' (later called Ficus latifoUa (Lesquereux) Knowl- ton), which was described originally from specimens collected at Black Buttes, Wyo., but has since been discovered at a number of other locahties. If only the basal portion was present it would be extremely difficult to dis- tinguish the San Juan leaf from the previously named species, but the shape of the whole leaf is very different. Thus, in Ficus -planicostata latifolia the blade is very much broader than long, but in the one imder consideration the reverse is true. Occurrence: Frmtland formation, 30 miles south of Farmington and 1 mile east of reserva- tion hue, San Juan County, N. Mex. Lot 75 (6956). Ficus sp. Plate LXXXIX, figure 1. The smgle much-broken specimen figured is the only one noted. It is a rather large leaf (about 12 centimeters long and 6.5 centimeters wide) eUiptical-ovate, with apparently a shghtly heart-shaped base, and entire margin. The entire upper part is missing. The nervation consists of a rather slender midrib and an un- knowni number of thin, remote, alternate secondaries, the lower of which has several tertiary branches on the lower side. None of the fuier nervation is retained, owing to the coarse-grained matrix. This form is so poorly preserved that it is hardly worth while to institute comparisons between it and various named species. Occiurence: Fruitland formation, Amarillo Canyon, 10 miles south of San Juan River and 4 rmles east of Chaco River, San Juan County, N. Mex. Lot 16 (6958). 1 Lesquereu-x, Leo, U. S. Geol. Survey Terr. Rept., voL 7, p. 202, pL 31, fig. 9, 1878. Ficus rhamnoides Knowlton. Plate LXXXVII, figure 3. Ficus rhamnoides Knowlton, U. S. Geol. Survey Bull. 163, p. 47, pi. 10, figs. 1-3; pi. 11, fig. 1, 1900. This species was described originally from specimens foimd in the Montana gi-oup at Point of Rocks, Wyo. The specimen figured here, although lacking aU the lower half of the leaf, agrees m everj' particular with the original specimens. Occurrence: Fniitland formation, 13 mUes south of San Juan River and 1 mile east of Chaco River, San Juan County, N". Mex. Lot 23 (6960). Ficus squarrosa? Knowlton. Plate LXXXVl, figure 10. Ficus squarrosa Knowlton, U. S. Geol. Survey Bull. 163, p. 45, pi. 8, fig. 2, 1900. The smgle example figured is referred with doubt to this species. It is a much smaller leaf than the type but does not appear to differ essentially otherwise. It is so poorly preserved, however, that it seems best to question the full identification. Occurrence: Fruitland fonnation, 30 rmles south of Farmington and 1 mile east of reserva- tion line, San Juan Coimty, N. Mex. Lot 76 (6963). Ficus sp. Plate LXXXVIII, figure 1. Leaf large, apparently nearly cncidar in gen- eral outUne, rather broadly heart-shaped at base, probably rounded above; nervation strongly marked, consisting of seven ribs from the top of the petiole, the central or midrib slightly the stronger, with two pairs of opposite, remote secondaries, other ribs (three on each side) about equally dividing the broad blade into four areas, the inner pair of ribs joining the lower secondaries of the midrib; the lower ribs wnth tertiary branches on the outside which join by a series of broad loops just inside the margin; nervilles numerous, very strong, mainly broken ; fuier nervation producing quad- • rangular areolae. This form is undoubtedly very weU marked, but unfortunately it lacks nearly all the margin except at the base and for a distance of some 4 centimeters above it. This leaf was probably 11 or 12 centimeters long and hardly less than 340 SHORTER CONTRIBUTIONS TO GENERAL GEOLOGY, 1<>16. 12 centimeters broad. The rather deeply heart- shaped base, seven strong ribs, strongly looping tertiaries, and strong, broken nervillcs would make it easy of recognition though it lacks so much of the blade. This leaf suggests at once Ficus wardii Knowlton,' from the Montana group at Point of Rocks, Wyo. That species, however, is smaller and has a shallower heart-shaped base and only five instead of seven ribs. The lower or outer ribs in Ficus wardii have numerous regular, parallel tertiary branches quite unlike the tertiaries in the present form. These char- acters should make it easy of recognition in the futm-e. Occurrence: Fruitland fomiation, 30 miles south of Farmington and 1 mile east of reserva- tion Ime, San Juan Comity, N. Mex. Lot 75 (6956). Ficus eucalyptifolia? Knowlton. Plate LXXXVII, figures 1,2. Picas eucalyptifolia Knowlton, U. S. Geol. Survey Prof. Paper 101, p. — , pi. 44, figs. 1, 2 (in press). The types of this species come from the Ver- mejo fonnation of Rockvale, Colo., and are described and figured in the report cited. The San Juan Basin material embraces a nmnber of leaves that appear to belong to this species, but as they are rather poorly preserved it has seemed best to question the reference. They are, so far as can be made out, of the same size and shape as the types, but the nervation, with the exception of the strong midrib, is obscure. The secondaries appear to be very thin, evenly spaced, and parallel, but their termination near the margin can not be seen. Occurrence: Fruitland fonnation, 13 miles south of San Juan River and 1 mile east of Chaco River, San Juan County, N. Mex. Lot 23 (6960). Family LAtTRACEJE. Laurus baueri Knowlton, n. sp. Plate LXXXIX, figui'e 5. Leaf evidently of thick texture and probably evergreen, lanceolate, narrowed in about equal degree to the wedge-shaped base and appar- ently acuminate apex (actual base and apex not presci'ved) ; margin cntu-e, provided with a thick "cord" wluch makes the actual margm; midrib relatively thick, straight; secondaries 1 Knowlton^ F. H., Flora of the Montana fonnation: U. S. Geol. Sur- vey Bull. 163, p. 48, pi. 9, fig. 1, 1900. very thin, alternate, at an angle of 30° or 40°, much cm-ved upward and disappearing just in- side the margm or each joining by a series of veiy thui loops to the one next above ; nervjlles all veiy much broken and irregular and form- ing different-sized areolae. The example figured is the only one observed. It is a nan-owly lanceolate, slightly unequal- sided leaf about 9 centimeters long and a little over 2.5 centimeters wide. It is remarkable in that the margin is formed by a thick fibrous "cord" nearly 1 millimeter in width, wliich otherwise resembles a secondary branch. The midrib is very thick for the size of the leaf, but the secondaries are~ thin and dehcate. The actual leaf substance is retained as a thin mem- branaceous carbonaceous fUm, which shows all the details of the nervation as completely as could be seen in a living leaf. Wlien this car- bonaceous film is removed it is fomid that the details of nervation are very faintly impressed on the matrix. It seems probable, from the thick midrib, the woody marginal "cord," and the faintly impressed secondaries, that the leaf was originally tliick and coriaceous and not unlikely was evergreen. Tliis species is of the general type of certain of the leaves referred by Lesquereux - to Laurus socialis, a well-known Tertiaiy form. It differs from that species, however, in its larger size, fewer secondaries, which curve upward for a longer distance, and above all in the presence of the marginal "cord." Occurrence: Fruitland formation, 2 miles east of Chaco River, San Juan Basin, N. Mex. Lot 14 (6948). Laurus coloradensis Knowlton. Plate LXXXVIII, figures 4, 5: Laurus coloradensis Knowlton, U. S. Geol. Sirrvey Prof. Paper 101, p. — , pi. 45, fig. 3 (in press). The type of tliis species comes from the Vennejo formation at Rockvale, Colo., and is described and figured in the report cited. The two leaves here figured are somewhat smaller than the type but do not appear to differ essentially in any other particular. Qccurrence: Fruitland formation, 13 miles south of San Juan River and 1 mile east of Chaco River, San Juan Coimty, N. Mex. Lot 23 (6960). = Lesquereux, Leo, U. S. Geol. Survey Terr. Kept., vol. 7, p. 21.3, pi. 36, flgs. 1-4, 7, 1S78. FLORA OF THE FRUITLAND AND KIRTLANE> FORMATIONS. 341 Family NYMPH^ACEa;. Nelumbo sp. Plate LXXXVI, figure 11. The San Juan Basin collections include the specimen here figured, which appears to belong to Nelumbo. It is a fragment from near the central part of what was a perfoUate leaf of considerable size, though none of the margin is now retained. The leaf was at least 12 centi- meters in diameter and veiy likely was nearly t'w'ice this size. It was evidently very thick, as is proved by the fact that the ribs seem deeply embedded in the leaf substance — so deeply, in fact, that it is impossible to deter- mine their exact number, though there were apparently as many as 18 or 20. There is also evidence that some of them were imforked. None of the other details can now be made out. It is hardly worth while to attempt com- parisons between this and described species of the genus, as so many of its characters are obscure or missing. It appears, however, to be of the type of the common livmg Nelwnho lutea (Wnidenow) Persoon, the water chinkapin of lakes and streams. Occun-ence: Fruitland formation, 30 miles south of Farmmgton and 1 mUe east of reser- vation line, San Juan Coimty, N. Mex. Lot 76 (6953). Family MIMOSACEa;. Leguminosites? neomexicana Knowlton, n. sp. Plate XC, figures 3, 4. Leaflet minute, about 11 millimeters in length and 5 miUimeters in width, ovate- lanceolate, romided and apparently truncate at base, acuminate at apex; midrib very thick for the size of the blade; secondaries ex- tremely thm, about four pairs, alternate, camptodrome, each joining the one next above and forming a bow far inside the margin ; finer nervation not retained. It must be confessed that this leaflet is of Httle stratigraphic value, its maui interest being in the evidence it affords of the presence of vegetation of this tj^jc in these beds. It is so very small and on the whole so poorly pre- served that comparisons with other forms referred to this type wiU not be attempted. Occurrence: Kirtland shale, 1^ miles north- east of Pma Veta Chma, San Juan County, N. Mex. Lot 40 (6966). FamUy STERCULEACE.ffl. Pterospermites undulatus Knowlton. Plate XO, figure 5. Pirrospennites undulatus Rnowlton, U. S. Geol. Survey Bull. 163, p. 67, pi. 16, fig. 3; pi. 17, fig. 2; pi. 18, fig. 4, 1900. The example figured, which unfortunately is much broken, appears to be referable to this species. It is of about the average size of those from Point of Rocks, Wyo., and with the exception of bemg a little more rounded below does not differ essentially. OccmTence: Fruitland formation, 30 miles south of Farmington and 1 mile east of reser- vation Une, San Juan County, N. Mex. Lot 75 (6956). Pterospermites neomexicanus Knowlton, n. sp. Plate XC, figure 6. Leaf evidently tliin in texture, ovate, abruptly truncate and slightly heart-shaped at base, acmninate at apex; margin entire below, becomuig slightly undulate in the mid- dle and few-toothed in the upper tliird, the teeth small, sharp, pomting outward, and separated by very shallow sinuses; midiib slender; secondaries about four pairs, sub- opposite, remote, the lower pair arising near the top of the petiole, with five or six branches on the lower or outer side, the lowest with several tertiary camptodi'ome branches on the lower side; upper secondaries simple or occa- sionally branched; nervilles few, irregular; finer nei-vation not retained. This leaf is regularly ovate, about 8 centi- meters long and 5.5 centimeters wide at the broadest point, wliich is just below the middle. It is otherwise distinguished by its few sec- ondaries and by the margin entire below and imdulate and finely toothed above. Occurrence: Fruitland formation, 30 miles south of Farmington and 1 mile east of reser- vation line, San Juaii County, N. Mex. Lot 75 (6956). Pterospermites sp. Plate LXXXIX, figiu-e 3; Plate XC, figure 1. One of the collections contains a fragment of a very large leaf that apparently belongs to the genus Pterospermites. It could hardly have been less than 18 centimeters in length and was at least 14 centimeters in width. It 342 SHORTER CONTRIBUTIONS TO GENERAL GEOLOGY, 1916. was apparently broadly ovate, rounded at the base, with the margin strongly undulate- toothed, both the teeth and the sinuses sepa- rating them being rouiided. The secondaries are strong, occasionally forked, approximately parallel, and craspedodrome. Nei-villes rather scattered, somewhat irregular but usually unbroken. Finer nervation producing an uTeg- ularly quadrangular mesh. Associated in the same collection is the specimen shown in figure 1, wliich represents the basal portion of a large leaf that is prob- ably the same species as that showni in Plate LXXXIX, figure 3. It is deeply heart-shaped at the base. The midrib is very strong, and the secondaries are at a very low angle. The petiole is preserved for a length of 1.5 centi- meters and is very thick and strong. This leaf, which is obviously too much broken to admit of satisfactory diagnosis, appears to be congeneric with Pterospermites undulatus Knowlton,' from Pomt of Rocks, Wyo., but it was much larger and has a more markedly undulate margin. The secondaries appear to emerge from the midrib at a lower angle than in the Point of Rocks form, but this pomt is obscm-e. The finer nervation is about the same in both forms. Occurrence: Kirtland shale; 1^ miles north- east of Pina Veta China, San Juan County, N. Mex. Lot 40 (6966). FamUy GROSStTLARIACEffi. Ribes neomexicana Knowlton, n. sp. Plate LXXXIX, figure 4. Leaf small, firm in texture, broader than long, three-lobed, the lateral lobes strongly toothed (central lobe much broken); base truncate or very sliglitly heart-shaped; nerva- tion not well retamed, consistmg at least of a faMy strong midrib and two lateral ribs that arise at or near the base and supply the lateral lobes, each apparently \\ath several branches on the outside that end in the marginal teeth. This little leaf, the only one of its kmd in the collections, is broadly ovate in general outline, about 2 centimeters long and nearly 3 centi- meters broad. It appears to be rather deeply three-lobed, and each of the lateral lobes is pro- I Knowlton, F. H., U. S. GeoL Survey Bull. l&J, p. 67, pi. 17, fig. 2, 1900. vidcd with five or si.x strong pointed teeth. The middle lobe is so much broken that its exact shape can not be made out, though it was doubtless toothed like the others. Occurrence: Fruitland formation, 10 miles south of San Juan River and 4 miles east of Chaco River, vSan Juan Coimty, N. Mex. Lot 16 (6958). INCERTiE SEDES. Carpites baueri Knowlton, n. sp. Plate LXXXVIII, figure 2. Fruit spheroidal, about 11 by 13 millimeters in short and long diameter, much compressed; surrounded by an exocarp fully 1 miUuneter thick; "stone" deeply sulcate at one end but not otherwise marked. The example figured is the only one found in the collections and consequently its exact character is difficult to determme; in fact, it is impossible to be certain of its orientatioia. The sidcation on one side is probably opposite the pomt of attachment. The nature of the outer covering is difficult to mterpret. It could hardly have been fleshy, as it is so uni- form in thickness and so distinct, and proba- bly it was an exocarp sunUar to that in certain species of Carya. The umer portion, or " shell, ' ' is without markings except the deep furrow on one side. It is perhaps needless to add that its affinity is not known. Occurrence: Fruitland formation. Coal Creek, 35 mUes south of Farmmgton and 1 mUe east of reservation line, San Juan Coimty, N. Mex. Lot 81 (0955). Phyllites petiolatus Knowlton, n. sp. Plate XCI, figure 3. Leaves small, membranaceous in texture, elliptical-lanceolate, long wedge-shaped at base, apparently narrowly acuminate at apex; margin perfectly entire; petiole very strong, more than 2.5 centimeters in length; midrib very strong below, becoming thin in the upper thu-d of the leaf; secondaries four pairs, alter- nate, at an angle of about 50°, each running up for a long distance and disappearmg in or near the margin or joining the secondary next above; nervdles numerous, very thin, mainly unbroken, at right angles to the secondaries; finer nervation obscure. FLORA OF THE FKUITLAND AND KIETLAND FORMATIONS. 343 The leaf figured is the best one observed. It is about 7 centimeters long and slightly over 2 centimeters in width, the petiole, as already stated, addmg at least 2.5 centime- ters to the total length. This species niay be known by its elliptical-lanceolate shape, long wedge-shaped base, long, thick petiole, and only foui" pau"s of alternate secondaries. Occurrence: Fruitland formation, 30 miles south of Farmmgton and 1 mile east of reser- vation Ime, San Juan County, N. Mex. Lot 75 (6956). Phyllites neomexlcanus Knowlton, n. sp. Plate XCI, figure 2. Leaf small, rather thick in texture, slightly obovate-eUiptical, rather abruptly wedge- shaped at base, apparently roimded and rather obtuse above; margin entire; midrib relatively strong; secondaries about five pau's, subopposite, remote, at an angle of about 50°, somewhat turned upward, camptodrome or just barely reaching the margin; nervilles mainlj^ imbroken, somewhat oblique to the secondaries. This little leaf is 6.5 centimeters m length and 3 centimeters in width just above the middle. Occurrence: Fruitland formation, 30 miles south of Farmington and 1 mde east of reser- vation Ime, San Juan Coimty, N. Mex. Lot 75 (6956). Unassigned plant (a). Plate XCI, figures 4-9. In one of the collections there are several specimens of a plant whoso exact affinity has not been ascertained. The most complete ex- ample, shown in figm-o 8 (enlarged in fig. 9), is pinnate, with several naiTOwIy lanceolate fern- like "pmnae," about 15 miUimetors long and 5 millimeters broad, with two close rows of small, scythe-shaped organs (pinnules or leaf- lets). Some of the little "pinnules" (see fig. 4) are short-petioled; others are nearly or quite sessile below, becommg confluent above. In the upper portion of some of the "pinnas" the "pinnules" are set so closely that the blades distinctly overlap. The nervation of the "puinules" is pecuUar. A single vein arises from the petiole at the lower margin of the blade and forks just above its point of origin; the lower branch usually traverses the length of the blade and occasion- ally branches or forks, and tho upper one dis- tinctly forks once or twice into equal branches entirely after tho manner of a fern. The size and general appearance of this little plant, as well as the shape and clo§e overlap- ping of many of tho foliar organs ("pinnules" or "leaflets"), are somewhat suggestive of Selagmella, but if it were of that genus it should belong to a typo m which there should be present two other rows of minute scalelike leaves, and nothing of the kind has been de- tected. Moreover, the nervation does not agree with the nervation of SelagineUa. On first inspection it suggested SelagineUa falcata Lesquereux,' from Point of Rocks, Wyo., but closer study shows that the "leaves" are much broader in the present material and have a quite different nervation. The pinnate arrangement of the naiTOW "pinnae," the appearance of the "pumules," and above all the forking nervation suggest a small dehcate fern of the asplenoid type, but Mr. W. R. Maxon, to whom the specimen was shown, is quite certain it is not a pteridophyte. Considering these imcer tain ties as to its affin- ities, I will not vontm-e to place it biologically until more and better material is available. Occmrence: Ku'tland shale, li miles north- east of Pina Veta Clnna, San Juan County, N. Mex. Lot 43 (6965). Unassigned plant (b). Plate XCI, figure 1. In one of the collections of red baked shale there is a specimen that is worthy of brief mention, although it is very fragmentary and hence difficult of allocation. It consists of a fragment of a stem about 4 centimeters in length and about 4 rmllimeters in diameter. It bears apparently opposite or subopposite leaves, whose sheathing bases cover the stem for a considerable distance below each node. The leaves are lanceolate and slightly con- stricted at the base and presumably acute at the apex, though no tips are preserved. The leaves are retained for a length of about L5 centimeters, but may well have been many times this length. They are unkeeled ami pro- 1 Lesquereux, Leo, U. S. Geol. Stirvey Terr. Eept., vol. 7, p. 40, pi. 61, figs. 12-15, 187S. 344 SHORTER COKTRIBUTIONS TO GENERAL GEOLOGY, 191G. vided with numerous close, fine, parallel veins ■with cross veinlets. The affinity of this plant has not been deter- mined. The leaves with theu' sheathing bases suggest a sedge, but this is hardly more than a suggestion, for they do not agree with the com- bined characters of this group. The plant is evidently a monocotyledon and in general ap- pearance som.ewhat resembles some of the leafy-stemnied forms of C^'pripedium or Habenaria, but this resemblance is perhaps hardly more than superficial, and the fact re- mains that the specimen is too fragmentary to pennit complete identification. Occurrence: Fruitland formation, 30 miles south of Farmington and 1 mile east of reser- vation line, San Juan County, N. Mex. Lot 75 [6956]. PLATES LXXXIV-XCI. 345 PLATE T.XXXIV. Page. Figures 1,2. Onoclea neomcxicana Knowlton 332 3. A yiem in h'speria Knowlton 332 4. Anemia sp 333 5-9. Asplcnium neomexicanum Knowlton 331 346 U. S. GEOLOGICAL SURVEY PROFESSIONAL PAPER 98 PLATE LXXXIV FLORA OF THE FRUITLAND AND KIRTLAND FORMATIONS. U. S. GEOLOGICAL SURVEY PROFESSIONAL PAPER 98 PLATE LXXXV FLORA OF THE FRUITLAND AND KIRTLAND FORMATIONS. PLATE LXXXY. Page. Figure 1. Sabal? sp 336 2. Sabal montana? Knowlton 335 3. Geinitzia formosa Ileer 333 4. Pistia corrugala Lesquereux •. . . 334 5. Heteranthera crctacea Knowlton 333 347 PLATE LXXXVI. Page. Figure 1. Myiica lorreyi Lesquereux 336 2-4. Mijrica? neomexicana Knowltmi 336 5, 6. Quercus haueri Knowlton 337 _ 7,8. Salix haueri Knowlton 337 9. Salix sp. a Knowlton 337 10. Fkus squarrosa? Knowlton 339 11. Nelumbo sp 341 348 U. S. GEOLOGICAL SURVEY PROFESSIONAL PAPER 98 PLATE LXXXVI ■•"!.. . >v FLORA OF THE FRUITLAND AND KIRTLAND FORMATIONS. U. S. GEOLOGICAL SURVEY PROFESSIONAL PAPER 98 PLATE LXXXVII FLORA OF THE FRUITLAND AND KIRTLAND KORMATIONS. PLATE LXXXVII. Figures 1, 2. Ficus eucalypl!folia? Knowlton 340 3. Ficus rhamnoidrs Knowlton 339 4. Ficus prxlatifolia Knowlton 338 349 PIATE I-XXXVIIl. Page. FiouHE 1. Ficus sp 339 2. Carpitcs baueri Knowlton 342 3. Ficus curtaf Knowlton 338 4. 5. LauTus coloradensis Knowlton 340 350 U. S. GEOLOGICAL SURVEY PROFESSIONAL PAPER 9a PLATE LXXXVIII FLORA OF THE FRUITLAND AND KIRTLAND FORMATIONS. U. S QEOLOQICAL SURVEY PROFESSIONAL PAPER 98 PLATE LXXXIX , ^" FLORA OF THE FRUITLAND AND KIRTLAND FORMATIONS. PLATE I,XXXTX. Page. FiGUKK 1. Ficus sp 339 2. Fieus haueri Knowlton 338 3. Pterospennites sp 341 4. Ribes neomexicana Knowlton 342 5. Lauras baueri Knowlton 340 351 PLATE XC. Page. Figure 1. Pterospermites sp 341 2. Ficus leei Knowlton 338 3. Leguminosiles' neomexicana Knowlton 341 4. Same, X 2} 341 5. Pterospermites undulatus Knowlton 341 6. Pterospermites iieomexicaniis Knowlton 341 352 U. S. GEOLOGICAL SURVEY PROFESSIONAL PAPER 98 PLATE XC FLORA OF THE FRUITLAND AND KIRTLAND FORMATIONS. PROFESSIONAL PAPER 98 PLATE XCI i^^ / i€ 2 - fe;>j-^^ 9 FLORA OF THE FRUITLAND AND KIRTLAND FORMATIONS. PIATE XCI. Page. Figure 1. Unassigned plant (b) 343 2. Phylliles neoiiuxicanus Knowlton 343 3. Fhyllites petiolatus Knowlton 342 4. Unassigned plant (a) 343 5. Same specimen, X2J 343 6. Unassigned plant (a) 343 7. Same specimen, X2i 343 8. Unassigned plant (a) 343 9. Same specimen, X2i 343 353 o /