r.; k ^ '4rl^ ^'^' W ^^^^ WSjfliSi *£.f U^r -ft- t .*i' i/jT/i^' DEPAETMENT OF THE INTERIOR-U. S. GEOLOGICAL SURVEY CHARLES U. WALCOTT, DIUECTOK y. A$i THE STRATIGIUPHK) SUCCESSION >^ SOUTHERN ANTHRACITE COAL FIELD, PENNSYLVANIA DA.VID TTHITE EXTRA(;T from the twentieth annual report of the survey, 189K-99 PART 11— GENERAL (iEOLOGV AND PALEONTOLOGY WASHINGTON GOVERNMENT PRINTING OFFICE 1900 THE STRATIGRAPHIC SUCCESSION FOSSIL FLORAS OF THE POTTSVILLE FOR- MATION IN THE SOUTHERN AN- THRACITE COAL FIELD P K N ^ S V I^ V A N I A LIBRARY r^GW YORK -N/CAU DAVITJ WHITE 74i) CONTEXTS. Page. Introduction 755 Application of tiie term " Potts^dlle formation " 755 Agency of fossil plants in the correlation of the terranes 756 Purpose and scope of this paper 757 General description of the Southern field 759 Form and extent of the field 759 General geologic structure 759 Description of the Pottsx^ille formati(.)u in the type region 762 Composition of the formation 763 Conglomeratic nature of the Coal Measures 766 The Lykens or Pottsville coals 766 Nomenclature of the coals 768 The type paleobotanic section of the Pottsville formation 769 Fossil-plant collections . 769 Species and their observed distribution within the formation and field 771 Paleontologic divisions of the formation 773 Floras of the Lower Lykens division 790 Flora of the Zero and No. 6 coals 790 Flora of Lykens coal No. 5 791 Paleontologic features of the roof shales of coal Xo. 4 792 Comparison of the lower jiortion of the type section with reference to the Lykens coal horizons i 793 Flora of the Lower Intermediate division 794 Brief existence of a transition flora. 794 Flora of the Upj^er Lykens division _ 795 Flora of Lykens coals Nos.-3 and 2 795 Flora of Lykens coal No. 1 798 Upper Lykens zones in the type section 798 Flora of the Uj^per Intermediate division 800 Plants of beds ]M and N in the type section 800 Correlations 803 Paleontologic relations of coals developed at isolated mines in the Southern Anthracite field 803 Horizon of the lower Lykens Valley coal in the Western Middle Anthra- cite field, between Frackville and Shamokin 810 Zones of the Pottsville floras in other regions of the Ai)palacliian province 812 Clark formation 814 Quinnimont formation 815 Sewell formation 816 Lookout formation 817 Fayette samlstone 818 Campbell Ledge, Northern Anthracite field 819 Relative horizons of the l)asal Iteds of the thin sections as comi)ared to the thick eastern sections 820 751 752 CONTENTS. Page. Upper limit of the Pottsville foniiatioii 823 Flora of the base of the Lower Coal Measures in the Sduthern Anthracite field 823 Flora in the roof shales of the Twin (Buck Mountain) coal 824 The paleontologic uj)per limit of the formation 829 Lower limit of the Pottsville formation 831 The Pottsville formation in the Dauphin liasin 832 Age of the coals in Lorberry Gap 833 Section at Lorberry ( raji 833 The fault in Sharp ^lonntain 835 Section at Fishing Creek Gaj) 838 Western limit of the fault 839 Position of the formation al()ng Sharp Mountair. 841 Section at Black Spring < 'ap 842 Section at Gold 3Iine ( Jaj. 842 Section at Ransch ( rd\\ Lebanon County . S44 Yellow Springs Gap 845 Rattling Run Gap 849 Big Flats 850 Short ^lountain shaftings west of Big Flats 851 The Lykens coals in Stony Mountain 854 Geiieral conditions relating to the occurrence of the Lykens coals in the Dauphin Basin 855 Thickness of the formation in tlie Southern Anthracite field 857 Variation in the constituent terranes in the formation 862 Notes on or descriptions of some of tne more characteristic species of fossil plants of the Pottsville formation in the Southern Anthracite field 865 List of fossil plants from the Pottsville formation in the Southern Antlira- cite field - 866 Notes and descriptions 868 Age of the Pottsville formation 911 Summarv of conclusions . . 913 ILLUSTRATIONS. Page. Plate CLXXX. Outline map of the Southern Anthracite field in Pennsylvania 918 CLXXXI, CLXXXII. Section in the Sharp Mountain gap at Pottsville. . . CLXXXIII, CLXXXIV. Section of the Pottsville formation in the Lincoln mining district 918 CLXXXV. Sections of Locust ^Mountain at Tamaqua and of Sharp Mountain at Rausch Gap, Schuylkill County, and at Lorberry Gap 918 CLXXXVI. Sections at Fishing Creek Gap, Black Spring Gap, and Gold Mine Gap 918 CLXXXVII. Sections at Gold Mine Gap and Rausch Gajj, Lebanon County 918 CLXXXVIII-CXCIII. Figures of fossil plants characteristic of the several paleobotani(! zones in the Southern Anthracite field 920-930 20 GEOL. PT 2 48 ^'53 THE STHATIGRAPHIC SUCCESSION OF THE FOSSIL FLORAS OF THE POTTSVILLE FORMATION IN THE SOUTHERN ANTHJSA- CITE COAL FIELD, PExNxNSYLVANIA. By David White. IXTRODUCTIOX. APPLICATION OF THE TERM "POTTSVILLE FORMATION:"' The Pottsville formation, or "Pottsville series" or ''Pottsville eon- glomerate," as it has more often been known, is a group of largel}" arenaceous beds of highl}^ variable thickness which, in eastern Penn- sylvania, lies between the Mauch Chunk red shale, or distinctly Lower Carboniferous, and the Lower Productive Coal Measures, or distinctly Upper Carboniferous. Besides the term employed for these terranes in this report, this formation has been otherwise designated the ""Serai conglomerate" or '' Great conglomerate," b}' the early Penn- sylvania geological surve}';^ the "Pottsville conglomerate," and locally the "Lykens series," by the second geological survey;^ the "Con- glomerate series" by many other geologists;^ and more recently it has been known in the northern portion of the Appalachian trough as the "Pottsville series," the modified name published by Dr. L C. White.* The early name "Serai" is hardly adaptable to present use, CD since it was applied by Rogers to the entire Carboniferous series above ^ the red shale in the anthracite region, the lower portion being distin- I guished from the remainder only by the addition of the word "con- ^ glomerate." The claims to consideration of "Great conglomerate" '21. and "Conglomerate series" have been rejected by most geologists, — ^ since throughout the greater part of its extent the formation is found to be productively coal-bearing, while in certain districts it contains little or no conglomerate. Nevertheless, the magnificent development of the terranes displayed at Pottsville, which is cited by all authors as the type locality under the various names, and from which the later names "Pottsville conglomerate" and "Pottsville series" were derived, is overwhelmingly conglomeratic, as well as deficient in profitably workable coal. In the various districts of the ' Rogers, Geol. Pennsylvania, Vol. 1, 185cS, pp. vii, 109, 146, 148; Vol. II, Pt. I, pp. 16, 17. -.\iinual Kept., 1886, Pt. Ill; Summary Final Report, 1896, Vol. Ill, Pt. I; Atlas Southern Anthracite Field, Pts. I-VI. 3 Fontaine, Am. Jour. Sci., 3d series, Vol. VII, ISTti, pp. 4.59, 573. r of terms are used for individual terranes or local groups, which have often erroneously and without harmony been correlated with portions of, or with the whole of, the Pottsville formation. The extra-anthracitic nomenclature is not, however, involved in the immediate consideration of the formation in the type region. '"Formation" is here used in a broad sense in preference to "series," in view of the subordinate rank of the collective terranes in the geologic column, their biologic unity, and their lack of individual persistence or continuity. In general, the Pottsville formation has been understood as wholly or in part representing the Millstone grit of Nova Scotia, New Bruns- wick, and the Old World. This correlation is founded chiefly on the lithologic similarity and the coincidental occurrence of the two forma- tions at the l)ase of the Productive Coal Measures. In the case of the Pottsville the correlation has rested entirely on the order of strati- graphic occurrence and the lithology, a method of coordination that, as will be shown later, has resulted in the reference of a portion of the formation in the Southern States, where it contains the most valual)le coals of the Southern Appalachian districts, to the Productive Coal Measures. It has, moreover, been the custom to consider the lithologic representative in each State or region as contemporaneous with and equivalent in toto to the lithologic member or group in ever}" other region, including the type section. The studies, now in progress, of the plant fossils of the terranes in different regions, correlated by lithology with the Pottsville formation, clearly show the fallacy of reuartling the lithologic section in each region as covering the same time interval as that covered by every other section. They also show that in certain regions considerable thicknesses of beds which, on account of the lithology, are referred to the post-Pottsville Coal Meas- ures contain the well-marked and distinctly characteristic floras of various horizons in other sections which have been determined, on the lithologic basis, as Pottsville. It is clear that under such circumstances the correlation, especially between separate basins, must be bv means of comparative paleontology. AGENCY OF FOSSIL PLANTS IN THE CORRELATION OF THE TERRANES. From tlie foregoing it will be seen that the existing condition is one in wliicli we have, under numerous names, a large number of terranes of supi)osed Pottsville age in both the interior or Mississippi basin and the (>astern l»asins, the final correlation of which is largely or often wholly dependent on the results of paleontologic studv. Since the organic remains in these beds are comi)osed prinlominantly, if not exclusively, of plant fragments, the foremost (piestions involved in any correlation with the Peinisylvaniansectionare: Has the typical Pottsville WHITE.] PURPOSE AND SCOPE OF PAPER. 757 formation any reliable and distinct paleontologic characters or aspect? is it satisfactorily distinguishable by means of fossils from the subja- cent formations or the overlying Coal INIeasures :' and does the plant life reveal such moditications in time, or such vertical distribution, as to constitute paleontologic subdivisions, zones, or horizons ? In answer to these inquiries, it must be confessed that up to the present time nothing has been known of the plants or their associations in the type section. No one appears to have studied the fossils from the Potts- ville formation in the vicinity of the type section in the Pottsville Basin, or even in the entire Southern Anthracite coal held. In tak- ing up the task of the stratigraphic elaboration of the Pottsville flora,^ it was further discovered that neither the upper nor the lower limits of the '"Serai" conglomerate, or "Pottsville conglomerate," are closely defined by the earlier geologists, while the somewhat conven- tional lithologic boundaries proposed in the latest publications on this subject appear to lack general acceptance. PURPOSE AND SCOPE OF THIS PAPER. As will have been inferred from the above statements of existing conditions and needs, the purposes of the studies, the immediate results of which are preliminarily reported in this j)aper, are: (1) The exploitation and elaboration, from a stratigraphic stand- point, of the plant fossils of the Pottsville formation in the tvpe region in the Southern Anthracite coal field. This involves the volumi- nous collection of fossils from as many horizons as possible, ranging throughout the entire thickness of the formation, as somewhat uncer- tainly defined on the lithologic basis. (2) The critical analysis and comparative study of the plant material collected, with a view to the discover}^ of the existence of anv natural paleontologic subdivisions, zones, or horizons, and their paleontologic characters, or the species of stratigraphic value, if any such are present. (3) The discover}^ of the paleontologic limits as differing or as agreeing with the lithologic limits of the type section, and the conse- quent paleontologic definition of the formation. This entails the examination of the fossils in the terranes below the Pottsville, as well as in the lower portions of the Coal Measures above the lithologic Pottsville, and the determination of («) their relations to the floras of the latter, and (b) the significance of those relations in both the geo- logic and the paleontologic grouping of the formation. Since, as has 1 Lest the use oi the word " paleontology" in this report be considered an unwarranted assumption by those who are accustomed to understanding the term as applying exclusively to animal remains, it should be explained that fossil plants in general are not only most widely distributed and frequently the only fossils in the terranes on the east side of the Appalachian trough, but also that in many of the sections, including the type section at Pottsville, no animal fossils, with the exception of Spiror- bis, small, rare, crustacean fragments, and a few cockroach wings, have yet been discovered. 758 FLORAL ZONES OF THE POTTSVILLE FORMATION. been remarked above, marine animal fossils are ver}^ rare in the ter- ranes of the Pottsville along the eastern margin of the trough, the usually abundant plant fossils constitute the chief evidence on which correlations, in this region at least, must be based. The primary result of this work should be the paleontologic defini- tion, if such is practicable, of the Pottsville formation, and the estab- lishment of a paleontologic section which shall constitute the type section of the formation, for comparison and reference in the study and correlation of other middle Carboniferous phytiferous terranes in the Appalachian province. Two other, largelv concomitant, results that are either economic or scientific in their nature have also been reached in the process of the elaboration of the fossil plants of the formation in th(^ typical region. The first, of some economic interest, is the correlation of groups of beds, or of individual coals wrought in disconnected or somewhat isolated portions of the Southern Anthra- cite field. The other, which concerns the question of general geologic correlation, is the acquisition of data for the determination of the age of the Pottsville formation — i. e., (a) the time interval represented by the tvpe section, and (/>) the e(|uivalents, in a broad sense, of the formation in other basins of this province and in other parts of the world. Incidentally, also, through the discovery in the Pottsville of floras already more or less completely known from isolated and uncor- related terranes in other regions of the United States, the way is opened to the proper reference and correlation of those terranes with the Pottsville, or with portions thereof. However, in this report no special eflort will be made to correlate the formations of this age in the bituminous regions, except in certain special or important cases. Such a work of general correlation will be more naturally and eft'ectively done in connection with a general study of the floras of the supposedly contemporaneous formations in the Appalachian trough and of their relations to the tvpical Pottsville, a work that of necessitv is dependent on and consequent to that now in hand. In this report the details of the geolog}' of the Southern Anthracite field are coiisidei'ed only to the exttMit to which they are concerned in the ascertained occurrence, distribution, or relation of the fossils exam- ined. Beyond a general description of the field, the stratigraphic data are largely confined to the ori(nitation of coals or plant beds, or to the dcHnition of the formations in certain sections. The details relating to the areal geologv are limited, first, to questions of the area of cer- tain coals as identified and correlated at different ])()ints liv means of the fossils, and, second, to the correction of certain areal and strati- graphic errors in the existing maps, especially in those relating to the western part of the field. These errors were discovered in the course of paleontologic investigation and were worked out b\' the combined methods of stratigraphy^ and paleontology. WHITE] FLORAL ZONES OF THE POTTSVILLE FORMATION. 759 GEXERAIL, DESCRIPTIOX OF THE SOUTHERX FIELD. FORM AND EXTENT OF THE FIELD. The Southern Anthracite field, known also as the Schuylkill or Pottsviile, and as the First Anthracite field, is, as its name implies, the most southern of the four fields or regions into which the anthracite basins of Pennsylvania group themselves. It embraces an area of about 181 square miles, lying in Carbon, Schuylkill, Lebanon, and Dauphin counties. Its territory is mapped on the Hazleton, Mahanoy, Pottsviile, Catawissa, Shamokin, Lykens, Hummelstown, and Harrisburg sheets of the Topographic Atlas of the United States. Its greatest longi- tudinal extent is a little over 70 miles, from the Lehigh River at Mauch Chunk, in a direction averaging nearly S. 60° W., to within li miles of the Susquehanna Kiver at Dauphin, 8 miles north of Harris- burg. Its maximum breadth is nearly 8 miles, from the crest of Sharp Mountain across Broad ^Mountain, in the region west of Pottsviile. Eastward the field narrows to a width of about 2 miles at Tuscarora, whence it extends, between Sharp Mountain on the south and Locust Mountain on the north, in a linear-lanceolate prolongation, hardly exceeding 2i miles in width, to the Lehigh River. Owing to the structure, the margin on the northwest, in the central portion of the field, is cut in rounded westward-projecting lobes of Broad Mountain, so that at a point a short distance west of Tremont, or about 12 miles west of Pottsviile, the field is reduced to a width of 4 miles. From this point the north and south borders diverge at an angle of about 20°. At the same time a very extensive arch, the Perry Count}" anticline, penetrates the field from the west, causing the parting of the latter, as far as a point about 4 miles west of Tremont, into two narrow divergent arms or prongs, forming what is known as the ''fish tail" of the Southern Anthracite field. The northern of these prongs, the Wiconisco Basin, l3ang between Bear and Big Lick mountains, is about 16 miles long, 2 miles in greatest breadth, and ends in a rather blunt point about 3 miles west of Lykens. The other prong, which also is about 2 miles wide at the base, and which tapers gradually from the latter for 30 miles to near the Susquehanna River, is known as the Dauphin Basin. It is bounded by Sharp. Mountain on the south and by Stony Mountain on the north. GENERAL GEOLOGIC STRUCTURE. Structurally, the Southern field is a synclinorium — a complicated group of synclines producing a great and, at points, irregular basin. Besides the numerous principal axes of folding, which are conforma- ble with the usual Appalachian trend, there are other oblique, more nearly due east-west undulations, which have had much to do with 760 FLORAL ZONES OF THE POTTSVILLE FORMATION. the d«^liinitation of the fields in their entirety, as well as in complicat- ing the minor structure of the basins. The general geolog'ic features of the region are shown in the large geologic map of the State, pub- lished in 18l>3, and in the county maps accompanying the descriptive reports of the second geological survey of the State. The relation of this region to the other anthracite regions is illustrated in the new Gen- eral Map of the Anthracite Region, revised to 189(>, published sepa- rately by the survey in reduced form, as one of the miscellaneous maps in Pt. I of the Atlas of the Southern Anthracite Field. The geology of the Southern field in particular has been worked out at great pains and expense by the late geological survev of Pennsjdvania. The six parts or volumes, with two supplements, comprising the Atlas of this field, in which are presented in great detail the mine workings and the areal geology, on a scale of 800 feet to the inch; numerous cross sections, mostly on a scale of 400 feet to the inch, and a great numl)er of columnar sections, chiefly at a unit of 40 feet to the inch, represent the latest and most experienced work of the survey. Not- withstanding the fact that the field methods were of necessit}' devel- oped in the course of the work in order to meet existing conditions, and the presence of many errors in correlation, the Atlas of the South- ern Anthracite Field in Pennsylvania represents the most minuteh' detailed and most excellent economic work on sedimentary depos- its that has j'et been accomplished over an extensive area in this country. To the abundant mine maps, profiles, and sections therein contained reference will frequently be made. Wherever the work here reported results in additions or corrections to the State mine maps, they will be so described that the changes or additional matter can in most cases l)e readily applied. The flexures of the held, which ofl'er a most interesting study, were described at considerable length by H. D. Rogers in the Geologj^ of Pennsylvania. The geographic features of the ))asin are incompletely represented in Pi. CLXXX of this report, which is here presented as an index to the State maps, as well as for the purpose of indicating the localities at which fossils have })een collected. On examining the maps referred to above, it will be noted that in general the folding of the strata is closer toward the southern border of the fleld and more opcMi to tln^ north. Thus the southern limit of the l)asin, along Sharp Mountain, is somewhat overturned throughout the greater part of its length, while in the widest part of the V)asin, near the northern ])order, the undulation of the beds is com- paratively gentle. To this is possibly due the variable and interesting- topography of the district; for where the strata are more steeply inclined the ridges caused by the (n-osion of the soft shales on either side of a hard formation are naj-i-ow and sharp, while the hard ter- ranes, when but slightly flexed and nearly horizontal, form the resist- WHITE.] GENERAL GEOLOGIC STRUCTURE. 761 ant tloor of a high plateau. Along the southern border of the coal field we accordingly find that the upturned and nearly vertical hard beds of the Pottsville formation compose the crest of a narrow mountain, Sharp Mountain, between which and the upturned wall of the Pocono (Vespertine, X), the basal member of the Eocarboniferous in the Second IMountain, extends a narrow parallel valley cut in the likewise upturned soft red shales of the Mauch Chunk formation (Umbral shale, XI), which, in the Schuylkill region, reaches its maximum thickness of over 3.000 feet. The steep inclination of the Pottsville floor is similarly accountable for the rigid and cristate c-haracter of the rim of the coal field in Sharp Mountain, in Stony or Fourth Mountain on the north side of the Dauphin Basin, in Bear and Wiconisco moun- tains in the northern prong of the "fish-tail,'' and in Locust Mountain. Its more enduring composition, as compared with the terranes of the Coal iSleasures, has resulted in the erosion of irregular valleys, gener- ally corresponding to the axes of the basins. Throughout the South- ern Anthracite field, wherever the elevation of the Pottsville has been sufficient to cause its complete erosion, the consequent erosion of the underlying Mauch Chunk shales has been so rapid as to form sharply defined valleys, varying in width according to the area uncovered or the inclination of the beds. It thus happens that Broad Mountain, on which is spread the northern dilation of the coal field, comprises essen- tially an elevated undulating plateau sustained by the rigid, flexuous Pottsville floor of the Coal Measures, which here, as in the Coal Measures of the other anthracite fields, prevailingly forms anti- clinal ridges and synclinal valleys. The breaking through of the Pottsville on the anticlinal axes along the western portion of Broad Mountain is responsilile for the deep t!Ove-like valleys between the lobes of the field, as noted above, the borders being often formed by massive cliffs of the gently inclined Pottsville conglomerates. In the region north of Pottsville the elevated Pottsville formation is not entirely cut across at any point, the result being that there is a conti- nuity of the conglomerates, which bridge the axis from the New Bos- ton Basin into the ^Nlahanoy Basin of the Western Middle Anthracite field. The line of division ])etween these fields, which, as may be inferred, is somewhat conventional, is drawn along the axis south of the northward plunge of the conglomerates into the Mahanoy Basin. The irregularity of the margin of the Southern coal field along Broad Mountain is quite in contrast with the relatively straight bord- ers of the prongs of the '*fish tail," or of the southern margin of the field along Sharp Mountain from the Susquehanna River to a point east of Middleport. This condition is largely due to the closer fold- ing and increased depth of the Coal Measures toward the south, so that the soft shales beneath the Pottsville are not brought to light by the minor flexures. The effect of the latter is, however, evident 762 FLORAL ZONES OF THE POTTSVILLE FORMATION. alono- the border of the field, in the scallops at the west extremitT of the Wiconisfo Basin and in the angle of the "fish tail"' west of Tre- mont. while it also appears in Sharp Mountain itself east of Middle- port, and in the Sunmiit Hill district. The occurrence of another lobe of the field along the apparently rigid Sharp Mountain, in the region of Ijorberry Gap, seems only to have been escaped by an overthrust faidt of the basal portion of the Coal Measures, as will be shown in a later portion of this report. DESCRIPTIOX OF THE POTTSVILI.E FORMATIOX IX THE TYPE REGIOX. As has already been remarked, the Pottsville formation is, in the type region, composed chieflj'^ of massive siliceous conglomerates. It will be seen later, in the course of a comparison of various sections, that this topographicalh'^ conspicuous formation, which constitutes the floor of the Coal Measures, comprises a series of ponderous conglom- erates, which are more variable in color, composition, and assortment in the lower part, and more quartzose, dense, and light colored near the top. These conglomerates alternate near the base with washes of purple and olive mud or soft, greenish sandstone, and in the higher portion with thin beds of arenaceous shale, and arc interspersed with a number of carbonaceous beds, some of which, in portions of the field, are workable over considerable areas, their product being the most valuable of the anthracite coals. The formation, as a whole, varies greatly in thickness, the maximum of a little more than 1,200 feet being reached in the vicinity and to the west of the type section, east of which it thins remarkably. That it thins toward the west in the Southern field itself has more recently been doubted. It is clear, however, that the relative thickness of its divisions is quite different in some of the sections, if the total depth remains the same. The sandstones, like the coals, are extremely vari- al)le even within short distances.^ Northwestward the formation thins rapidly in the anthracite regions, its development being about 850 feet in the Shamokin district of the Western Middle field, or an average of about 350 feet in the Eastern Middle field, while it is recorded as averaging 2-25 feet in the Northern Anthracite field. In the Southern Anthracite field, the formation is apparently con- formable with the Mauch Chunk shales, while the line of separation bet\veen it and the superimposed Coal Measures, which are also highly arenaceous, abounding in conglomerates, has for convenience been drawn at the lowest workable coal in the type region. 1 Compare columnar sections on columnar-section sheet xi, Pt. IV B of the Atlas of the Southern Anthracite Field. WHITE.] FOTTSVILLE FORMATIOlSr IN THE TYPE REGION. 763 COMPOSITION OF THE FORMATION. The character and composition of the Pottsville formation and its rela- tion to the Umbral shale (Mauch Chunk formation) below and the Productive Coal Measures above are well shown in the magniiicent exposure in the' cutting along the Pennsvlvania Railroad through Sharp Mountain on the east side of the gap below Pottsville. A some- what detailed section of this exposure, extending from Tumbling Run Valley to the Pottsville Valley is given on Pis. CLXXXl, CLXXXIL This section includes the upper portion of the Mauch Chunk red shales and extends to the Dirt bed, the third workable coal of the Coal Measures at this point. The Pottsville formation itself may. for the present, be considered as comprising that portion, nearly 1,200 feet in thickness, of the section between the topmost bed of red shale and the "Twin" coal, which in both the first and the second geological surveys of this State has been agreed upon as the boundary between the Pottsville formation and the succeeding Lower Coal Measures. On referring to the section it will be observed that the lower por- tion represents a transition from the typical red, purplish-red, and olive-green shales of the Mauch Chunk to the almost exclusively are- naceous, ponderous quartz-conglomeratic terranes of the Pottsville formation. The conglomerates intercalated in increasing proportions in the upper beds of the Mauch Chunk consist of irregularly bedded, poorly assorted, or sometimes apparentl}' unassorted pebble or bowl- der accumulations in a matrix of coarse arkose sands colored by reddish or greenish shale washes. The pebbles are mostly of quartz, though sandstone, syenite, chloritic schist, limestone, and even red and green shales and conglomerate fragments are also present. Occasion- ally the pebbles, which are sometimes subangular, attain a diameter of 3 or 4 inches or more; but in most of the beds the coarsest materials do not exceed a g'oose egg in size. For a long distance from the base of the formation the conglomerate matrix consists of a micaceous, chiefly arenaceous medium, poorly cemented and often colored with a red or green argillaceous material. In passing upward the beds of red shale are less conspicuous, and at about l,:iOO feet below the Twin bed the last distinct stratum of t3q3ical Mauch Chunk red shale is seen. Above this the conglomeratic matter prevails almost exclusively through a long interval. Nevertheless, the olive-green shales occur here and there throughout an interval extend- ing 200 or 300 feet higher, while most of the conglomerates in the lower portion of the section derive their color from the greenish or reddish mingled sediments. The irregular l^edding and the variety of the rock materials in the pebbles, which are often imperfectly rounded, are interesting features of the lower portion oi the Pottsville forma- tion itself. This portion of the section is notably characterized by 7H4 FLORAL ZONES OF THE POTTSVILLE FORMATION. the occurrence of olive-green or slightl}' reddish mud beds, apparently red('p()sit«>d from the older formation. These muds often conclude rapid transitions from greenish conglomeratic sandstones into fine argillaceous silts of no great thickness. On the conspicuousl}^ uneven surfaces of the latter, coarse conglomeratic strata or typical bowlder beds are directly imposed, in knife-edge contacts, at a number of horizons in the lower half of the section. These irregular, interca- lated nuids. which are similar to others in the upper Y)art of the ]Mauch Chunk formation, sometimes appear as thin lenses interspersed among the irregular layers of the conglomerates. Without further detailed description of the type section, for which the reader is referred to Pis. CLXXXI. CLXXXII, it will appear that we have a series of beds of passage — i. e.. a transition series — consisting of coarse, heteroge- neous, semiassorted. conglomeratic materials, intercalated in the upper- most l>eds of red shale, above which, for a distance of several hundred feet, many of the conglomerates preserve essentialh^ the same charac- ters, although typical deposits of the red and green shales are want- ing. Subangular pebbles in imperfectly l)edded arkose conglomerates are not rare throughout the lowest third of the formation in this vicinity. Although the quartz material preponderates, pebbles of sandstone and shale are not infrequent. Occasionally some of the pebbles attain the proportions of goose eggs, and farther east, in a section near the Hacklebarney tunnel, soiue of them measure 5 to 6 inches in diameter. As already indicated, the conglomerates in the lower portion of the PottsviHe formation are prevailingly greenish, arkose, and poorly cemented, l^sually. in the more freshly cut sections, they otier little resistance, and frequenth-^ they are but slightl}'^ displayed. When, however, the erosion has been very slow, as along the summit of Sharp Mountain, the ferruginous material so cements the pebbles that the lower ledges of the formation often predominate and form, for consid- erable distances, the crest of the mountain. This featui'e is more noticeable to the eastward of Swatara (iap. In the upper half of the formation the conglomerates become more rigid, more distinctly are- naceous, and more persistent, the pe))bles l)eing better rounded, more compactly disposed, and regularly assorted. Sandstone without peb- bles is rare and is alwa3's thin in the section. Cross bedding, indica- tive of current movem«Mit from the northeast, is conspicuous. In the more shaly conglomeratic sandstones in the middle of the section con- cretionar}' weathering is especially noticeable. Generally speaking, the relatively small amount of shales and of coaly matter in the type section is, for the most part, contained in the mid- dle third. Toward the top the conglomeratic material becomes lighter colored, as well as more exclusive, and at a distance of 200 or 800 feet below the Twin coal, in that portion of the exposure opposite the WHITE] COMPOSITION OF THE POTTSVILLE FORMATION. 765 Pennsylvania Railroad bridge, occur the most massive, rigid, densely quartzitic, regularlv bedded, and persistent conglomerates of the entire section. These conglomerates constitute a close group or plexus of ponderous ledges in which the formation culminates. The}" usuall}^ form the conspicuous beds in ever}^ exposure of the formation, and in every break in Sharp Mountain through which the waters of the basin find escape these steeply inclined ledges appear as jagged, irregular teeth, picturesquely defining the jaws of the gap. Exceptions to this, however, are Lorberry and Fishing Creek gaps, at which the entire Pottsville formation appears to be absent. Usually they also form the crest of the mountain, although, as was remarked above, the lower conglomerates predominate in the older exposures. It may be noted in this place that these uppermost white or light-gray conglomerate plates, which in both their lithologic and their paleontologic characters are distinctly comparable to the Homewood sandstone in the bituminous basins, appear to have the greatest geographic extent and regularity of all the strata in the formation. Thev are amonsthe few individual beds which, although varying in thickness and in relative intervals, may be traced to sections in distant portions of the same field. As shown in Pis. CLXXXI, CLXXXII, the type exposure at Potts- ville exhibits a number of thin coals, none of which are profitably workable in this vicinity, although most of them have been diligently prospected. The exposure nearly 800 feet below the Twin bed appears to have been followed b}' a drift for some distance above the wagon road on the east side of the gap, while another coal, about 400 feet below the Twin, has been somewhat extensively tested, not only farther north in the same gap, but at two or more levels in the gap at West- wood. The consideration of the approximate and comparative age of some of these coals, with reference to the pi-oductive coals toward the western end of the field, will be continued in connection with the dis- cussion of the fossil plants of the various horizons. Plant collections have l)een made from eleven different horizons, marked A-N in the section, as well as from the roof of the Twin coal, marked O. Other published sections excellently illustrating the lithology of the Pottsville formation in the Southern Anthracite field are those at Hacklebarney,^ Nescjuehoning Gap,- and Locust Gap at Tamacjua,^ in the region east of Pottsville. The character of the sedimentation in the region north of Pottsville is shown by the records of the diamond-drill bore holes near the Altamont colleries, throughout a dis- tance of 5 or t> miles along Proad Mountain.* 'Phc composition of the 1 Atlas Southern Anthracite Field, Pt. I, mine sheet i, cross-section sheet i, columnar-section sheet i, section 4. 2 Idem, Pt. I, mine sheet ii, colummir-section sheet ii, cross-section sheet ii. 3 Idem, Pt. I, mine sheets iii and iv, columnar-section sheet ii, cross-section sheet iii, section 39, profile 12. *ldem, Pt. IV, columnar-section sheet ix, sections l-(j. 766 FLORAL ZONES OF THE POTTSYILLE FORMATION. beds in the western part of the field is well shown in the sections at Lincoln/ Kalmia," and Likens/ Portions of the first two named are reproduced in Pis. CLXXXIII and CLXXXTV of this report, in illustration of the formation in the Lincoln mininy- reuion. CONGLOMERATIC NATURE OF THE COAL MEASURES. It may l)e remarked in this place that throughout the Southern Anthracite field the Lower Coal Measures also are largelj^ conglom- eratic. Frequently these conglomerates rival in size and rigidity individual beds of the Pottsville formation itself. Illustrations of the proportions of conglomeratic material, which in some cases constitutes nearly a third or more of the section, are found in columnar-section sheets i and vi, in Pts. I and II, respectively, of the Atlas of the Southern Anthracite Field; or in the sections at the tunnels in Wood's colliery and Dundas No. 6 colliery, at the north base of Sharp Moun- tain, between Pottsville and Tremont, shown in columnar-section sheet viii, Pt. IV of the Atlas. The same character is still better presented in the regions north and west of Tremont, the sections of which are given in columnar-section sheet x, Pt. IV B of the Atlas. THE LYKENS OR POTTSVILLE COALS. It will be observed that in the section of the Pottsville formation at the gap south of Pottsville a number of thin coals are present, several of them ha^^ng been prospected in the vicinity of the typical localit3^ Coals are to be found in varying numbers in every complete section of the formation, though in the neighborhood of the type section the^^ have not proved to be of profitable thickness. However, to the north of Pottsville. on Broad Mountain, and to the west, throughout the Southern field, coals occur in greater development, especially locally, and have been extensively mined. Reference to several detached or somewhat isolated mines in those coals will again be made in connec- tion with the consideration of the distribution of the fossil plarjts and the correlation of the coals. These coals of the Pottsville formation, which are commercially known as the "Lykens" coals, and which comprise the "Lower lied Ash" groups of the Southern field, appear to be best developed or most advantageously exploited in the districts west of Tremont. including the Lincoln region and the Wiconisco Basin. In the anthracite fields, as well as in other coal fields of the Appa- lachian trough, the combustible of the Pottsville formation is generally >Atlns Southern Anthracite Field, Pt. IH, mine sheet xvii; Pt. IV B, columnar-section sheet xi; Pt. V, cros.s-scftii>n sheet .wii, section 24. 2Idem, Pt. in, mine sheets xxi and xxii; Pt. IV B, columnar-section sheet xi, columnar sections 10, 11, and 12; Pt. VI, cross-section .sheets xxi, section 29. 3Idem, Pt. Ill, mine sheet xx; Pt. IV, columnar-section .sheet vii, columnar sections 9, 10, 11, and 12; Pt. VI, cross-section sheet xx. section 28. WHITE.] THE LYKENS OR POTTSVILLE COALS. 7(57 the most valuable of the entire series of Carboniferous coals; for, while as individual beds the Pottsville coals may be inferior in thick- ness and areal extent, their superior qualities create for them the high- est demand and encourage their production even under conditions entirely unfavorable for the exploitation of other and thicker beds. To this formation belong- the Sharon coal of northern Ohio and north- western Pennsylvania; the Pocahontas and New River coals of Virginia and West Virginia, celebrated as steam and coking- coals; the chief coal horizons of eastern Tennessee; the coals of Georgia; and the prin- cipal furnace and steam coals of Alabama, The special fitness for domestic use of the rather fi-ee-burning Lj^kens coals, which wins for them an advance of from 25 cents to $1.25 per ton over the prices of other coals of the anthracite series, has resulted in the establishment in the Lincoln-Lykens region of several of the largest raining plants in the anthracite helds, the capacitj^ of the Lincoln and Brookside collieries,^ which are exclusively occupied with the Lykens coals, being 2,900 tons a day of ten hours. For a long time it was supposed that the Lykens coals were of the age of the Productive Coal Measures, the supra Pottsville series, but later and more svstematic stratigraphic work has shown them to be distributed through the Pottsville formation itself. It requires but a glance at the plant fossils of these coals to detect their antiquity as compared with those of the coals of the higher formation. Like the other members of the formation, the coals are exceedingly variable in thickness, often attaining a remarkable local development, though east of the Lincoln region they seldom reach a workable thick- ness except in isolated and restricted areas. Nevertheless, one of the lower coals appears to extend over a considerable territory in Broad Mountain, where it has been worked at a number of points, and whence it ma}' be traced over the narrow arch into the Shamokin region of the Western Middle Anthracite field. The coals have been tested at many points to the eastward. One of the beds is still worked in a mine operated by Mr. Isaac Christ on the east side of Locust Gap, at Tamaqua, while the fossils obtained from a drift lately opened near the head of the incline on Mount Pisgah, at Mauch Chunk, show the coal to lie relatively high in the Pottsville. In the Dauphin Basin, west- ward from Rausch Gap and the Lincoln region, the Lykens coals are not worked at present. The basin, the central portion of which was extensivel}^ prospected in the early half of the century, has long been abandoned, for the reason that in passing westward the coals opened were found to be soft, crushed, semibituminous, and of generally inferior qualit}". 1 Analyses of the West Brookside coals made by Dr. Cresson in 1879 show : Volatile matter, 5.4 per cent ; ash, 8.78 per cent ; .sulphur, 0.36 per cent ; phosphorus, none ; fixed (carbon, 85.636 per cent. 768 FLORAL ZONES OF THE POTTSVILLE FORMATION. It will bo shown in this report that the exploitations and provings on which was leased the conclusion that the Lykens coals were of inferior quality or worthless in the Dauphin Basin were, in fact, contined to the softer and inferior coals of the Productive Coal Measures, in the interior of the basin. These coals are not in the Pottsville formation. All the coals mapped by the State geologist as ''Lykens" throughout the greater part of the southern limb of that basin, including practically all the early developments east of the Big Flats, arc, in fact, within the Productive Coal Measures. The entire Pottsville formation, with its scarcely prospected Lykens coals, not only lies to the south of the supposed approximate boundary of the "lowest Lykens coal," but a large portion of its steeph^ inclined terranes, including the lowest Lykens coals, outcrops for nearh' a score of miles along a zone rep- resented as red shale on the mine sheets. NOMENCLATURE OF THE COALS. It is uncertain how many of the Lykens (Pottsville) coals are at one place or anothei- worka])le, since some of them are evidently too thin for protitable mining in each of the mine sections. Certain of the sections may show as many as a dozen or more thin coals or coaly partings, but it is not pro))able that more than eight or nine at most I are anywhere worked, and it is only in a few cases that as many as five coals in this formation can be productively worked at one locality. Usuallv not more than three are profitably mined at one point. The number of the principal workable coals and their relative posi- tions are best revealed at the Lincoln mines and in the Brookside- Lykens district, which, is essentially continuous to the westward of the former. At the Lincoln mines, Avhere the upper Lyk(>ns coals are best displayed, six coals are or have been worked. The colunmar sections, earlier mine maps, and profiles of these mines are shown in mine sheets xvii to xxi. Atlas Southern Anthracite Field, Pt. Ill; columnar-section sheets vii and xi. Atlas Southern Anthracite Field, Pt. IV; and cross-section sheets xvii to xxi. Atlas Southern Anthracite Field, Pt. V. In several of the sections the Lykens No. 1 coal is shown at about 210 feet below the coal identified by the State geol- ogist as the Buck ^Mountain l)ed, as at Good Spring, or at about 250 feet below a bed presumablv the same, as at the Lincoln mine. (See PI. CLXXXIII.) Lvkcns coid No. li, formerly worked at tht^- New Lincoln mine, is platted at approximately 240 feet below Lykens coal No. 1; while Lykens coal No 2 in the same mine is but 78 feet below the lattei- in tiie second lift tunnel. At this ])()int Lykens coal No. 3, which at other points may di\ crgi' asmuch as :^.(> feet ormon' from No. 2, is sepai'ated from the lattx'r by only 3 inches of dirt. Lykens coal No. 4, locally known as "White's bed," is about 245 feet below No. 3 at Lincoln. Lvkens coal No. 5, th«> "Lykens Vallev" bed, or. as it is also locally ^^HiTE] NOMENCLATURE OF THE COALS. 769 called, the '"Big- bed,"' is about 115 feet below coal No. 3 in the Lincoln tvinnels, though the interval is 140 feet in the large tunnel at Williams- town. A thinner and less extensive coal, Lvkens No. 6, or the "little bed," lies about 48 feet below No. 5 in the vicinity of the Lincoln mine, though at Williamstown the interval is over 65 feet. None of the upper Lvkens coals, Nos. 1-3. are worked in the Wiconisco Basin, unless it be at Kohler's Gap, north of Brookside, where a coal, supposed to be L3^kens bed No. 3, is dug- for local use. At the Brookside, Williamstown, and Short Mountain collieries oidy the lower coals (Nos. 4, 5, and 6) are worked, except at Williams- town, where a rather thin coal, the Zero bed, the thickness of which is g-iven as 3 feet in the pu])lished section (columnar-section sheet vii. Atlas Southern Anthracite Field, Ft. IV), has been developed at 37 feet 7 inches below coal No. 6. This bed, if present at all in the Kalmia section, must be represented by only 0 inches of coal, l)ut 5 feet 3 inches below the Lykens coal No. 0. The nomenclature of the coals giA'en aliove is that employed by the Fhiladelphia and Reading- Coal and Iron Company and adopted by the late State geological sur- ve}'. In the vicinity of Lykens the coals were formerly numbered from the base upward in an opposite direction, and thev are so designated in Rogers's discussion of this lield in the Geology of Fennsylvania. The above statements of the intervals between the coals are given as typical, without reference to the remarkable variation in the inter- vals as ascertained })v the underground connections from mine to mine. The matter of this variation, as well as the stratigraphic position of the coals on Broad Mountain and in other portions of the field, will be touched upon when considering the thickness of the Fottsville for- mation and the evidence of the plants as to the correlation of the beds. For information as to the thickness of the coals the reader is referred to the State publications cited above, or to th(^ tvpical section repro- duced in Fls. CLXXXIII and CLXXXIV. TIIP] TYPE PAIiEOBOTAXIC 8ECTIOX OF THE POTT8- TILF^E FORMATIOX. FOSSIL-PLANT COLLECTIONS. In the Southern Anthracite field fossil plants have been collected at 41 localities, from tlu> Fottsville formation, or from the roof of the coal supposed to form the dividing line bet\v(>en the Fottsville forma- tion and the overlying Froductive Coal ^Measures. These may ]>e grouped as follows: 1. Lower Lykens coals at Miller's drifts,' Big Run,' Wiconisco,'' Big 1 PI. CLXXX, station 14. Atlas Southern Anthracite Field, Ft. Ill, mine sheet xx. - I'l. CLXXX. station K5. Atlas Southern Anthracite Field. Pt. HI, mine sheet xx. * PI. CLXXX, station 12. Atlas .Southern Anthracite Field, Pt. Ill, mine sheet xx. 20 CKOL. PT 2 4St (To FL(»KAL ZONES OF THE P0TT8VILLE FORMATION. Lick.' Williainstowii.- Brookside,'' and East Brookside in the "NVicon- isc-o Basin: and from Kalmia/ Lincoln.' Kaiiscb Gap (PI. CLXXXV) Schuylkill County." and Swatara Gap' of Sharp ^Mountain, in the Lin- coln district. -• ^"pp^^i" Lykens coals at Lincoln colliery. Ne\v Lincoln.^ the North r>rookside slope near Good Spring,'' and the lower Eureka tunnel north of the old Colket mine.'" 3. A third cateoory. including detached i)oints. or beds whose rela- tions to the individual Lykens coals are subject to doul)t, embraces col- l(>(tioiis from Kohlers Gap,'' a shaft near the North Brookside slope, the upper P^ureka tuimel, a prospect shaft near the mouth of the latter, tlu^ Kemble drift.'" A-ltamont colliery No. I,''' two levels in the gap at Westwood,'* the drift in Mount Pisgah,'' and 12 levels in the type sec- tion at the Pottsville Gap.'*' 4. The collections from various levels at Lorberry Gap,'' Fishing- Creek,'* Black Spring Gap (Mount Eagle),'' Gold Mine Gap,-'' Rausch Gap (Lebanon Count}^),^' Yellow Springs Gap,-' Rattling Run Gap,'* iPl. CLXXX, station 11. Atlas Southern Anthracite Field, Pt. Ill, mine sheet xix; Ft. VI. cross- section sheet XX. 2pi. CLXXX, station 10. Atlas Southern Anthracite Field, Pt. Ill, mine sheet xix; Pt. IV, columnar-section sheet vii, section 8; Pt. VI, cross-section sheet xx, section 27. 3 PI. CLXXX, station 9. Atlas Southern Anthracite Field, Pt. Ill, mine sheet xviii; Pt. VI, cross- section sheet xix, section 26. ■•PI. CLXXX, station 41. Atlas Southern Anthracite Field, Pt. Ill, mine sheets xxi and xxii; Pt. IV B. columnar-section sheet xi, sections 10, 11, and 12; Pt. VI, cross-section sheet -xxi, section 2it. s PI. CLXXX, .station 5. Atlas Southern Anthracite Field, Pt. Ill, mine sheets xvii and xxi; Pt. IV B. columnar-section sheet xi, sections 8-9; Pt. VI, cro.ss-section sheet xvii, cross section 24. «P1. CLXXX. station 4. Atlas Southern Anthracite Field, Pt. Ill, mine sheet xvi. ■ PI. CLXXX, station 3. Atlas Southern Anthracite Field, Pt. Ill, mine sheet xvi. 8 PI. CLXXX, station (5. Atlas Southern Anthracite Field, Pt. Ill, mine sheet xvii; Pt. IV, cohniinar- section sheet vii, section 4; Pt. V, cro.ss-section sheet xviii, section 24. '•'PI. CLXXX, station 7. Atlas Southern Anthracite Field, Pt. Ill, mine sheet xvii; Pt. IV B. columnar-section sheet x, section 8; Pt. VI, cro.ss-section sheet xix, section 25. 10 PI. CLXXX, station 33. Atlas Southern Anthracite Field, Pt. Ill, mine sheet xvi: Pt. IV B, columnar-section sheet x, section C; Pt. VI, cros.s-section sheet xvii, .section 23. " PI. CLXXX, station 15. Atlas Southern Anthracite Field, Pt. Ill, mine sheet xviii; Pt. VI, cros.s- .section sheet xix, section 20. I'-Pl. CLXXX, station 16. Atlas Southern Anthracite Field, Pt. II, mine .sheet xiii; Pt. V, cross- section sheet xviii, section 23. "PI. CLXXX, station 36. Atlas Southern Anthracite Field. Pt. II, mine sheet vii; Pt. IV, columnar- section slieet ix, section 1; Pt. V, cross-section sheet v. sections 16 and 17. '< PI. CLXXX, station 2. Atlas Southern Anthracite Field, Pt. II. mine sheet xiv; Pt. IV B, columnar- scctioi) sheet xi, section 5; Pt. VI, cross-section sheet xii, section 19. i^Pl. CLXXX, station 40. Atlas Southern Anthracite Field, Pt. I, mine .sheet i, cross-section sheet i, section 1. "'•PI. CLXXX, station 1. Atlas Southern Anthracite Field, Pt. IIIB, mine sheets xiv and xiva; Pt. IV, columnar-section .sheet viii, section 3; Pt. V, cross-.section sheet viii, section 17. '• PI. CLXXX, .station 17; PI. CLXXXV, Fig. 1. Atlas Southern Anthracite Field, Pt. Ill, mine sheet xxi; Pt. VI, cro.s.s-section sheet xvi, section 24. '* PI. CLXXX, station 18; PI. CLXXXVI, Fig. 1. .A.tlas Southern Anthracite Field. Pt. Ill, mine sheet xxi. 'M'l. CLXXX, station 19; I'l. CLXXXVI, Fig. 2. Atlas .Southern Anthracite Field, Pt. Ill, mine .sheet xxi. ••"'PI. cLXXX, .station 20; PI. CLXXXVI. Fig. 3; PI. CLXXXVII, Fig. 1. Atlas Southern Anthracite Field, Pt. HI, mine sheet xxii; Pt. IV, columnar-seetiou sheet viii, section 7; Pt. VI, cross-section slieet xxi. section 29. •-' PI. CLXXX, station 21; PI. CLXXXVII, Fig. 2. Atlas Southern Anthracite Field, Pt. Ill, mine sheet xxiii; Pt. IV, columnar-section shevt viii, section 9; Pt. VI, cross-section .sheet xxi, section 30. ■-■■-PI. CLXXX. station 23. Atlas Southern .Vnthrucite Field, Pt. Ill, mine sheet xxiv; Pt.IV, cross- section sheet xxi. section 31. "'PI. CLXXX. station 24. Atlas .Southern Anthracite Field, Pt. Ill, mine sheet xxv; Pt. IV, columnar-.section sheet viii, section 10. WHITE.] SPECIES WITHIN THE FORMATION AND FIELD. 771 Big Flats ^ north of Watertank Station, Fort Lookout,' and a number of the old drifts ^to the westward, made in the early part of the cen- tury, will be considered in connection with the special discussion of the Dauphin Basin. 5. The lifth category includes plants from the "Buck Mountain" coal or a coal (the Twin coal) supposed to ))e its equivalent at Swatara Gap at Middle Creek,* Ebony colliery' north of Newcastle, Altamont colliery No. 2,'' Locust Mountain and Sharp Mountain gaps," near Tamaqua. and at the Pottsville Gap. SPECIES AND THEIR OBSERVED DISTRIBUTION WITHIN THE FORMATION AND FIELD. In order to avoid the repetition of names which would result from an enumeration of the species from each locality or bed, the plants from the Pottsville formation in the region of Pottsville and west- Avard in the Southern Anthracite lield, exclusive of the Dauphin Basin, will be combined in ojie list, with a table showing their dis- tribution so far as yet ol)served in that foi-mation. Since the eco- nomic interest of the problem of stratigraphic paleontology centers primarily about the Lykens coals, the localities affording plants from the roof shales directly in connection with these coals, as detinitely correlated between the large mines, are placed tirst. From an eco- nomic standpoint thev constitute a typical paleontologic representa- tion of the productive coal-bearing horizons, just as the Pottsville Gap section affords a typical paleontologic section of the formation as a avIk^Ic. Since, also, it is at once clear that the species commonly in association with the lower Lykens coals are largely different from those over the upper Lykens coals, the principal coals of the mining- region are naturally divided paleobotanically into tA\o groups: An upper group, including coals 1-3, and a lower group, containing Lykens coal No. i and the remaining lower portion of the formation. For the sake of easier comparison, the plant-bearing horizons A-M,* in ascending order, in the Pottsville Gap section, are placed next. To the right of these are a number of columns representing isolated devel- opments of supposed Lower Lj^kens age; and beyond these are a few 1 PI. CLXXX, .station 26. Atlas Southern Anthracite Field, Pt. Ill, mine sheet xxvi. -PI. CLXXX, station 27. Atla.s Southern Anthracite Field, Pt. Ill, mine sheet xxvi; I't. VI, columnar-section sheet viii, section 11; Pt. Ill, mine .sheet xxvi. ^Pl. CLXXX, stations 28-32. Atlas Southern Anthracite Field, Pt. Ill, mine sheets xxvi and xxvii. matm'ial from the' rock dumps, either detinitely or api)roximately. to their oi'iginal sources. Such references, made with great caution, are indicated in the respective t-olunms by numbers referring to the coal from whose roof the specific fossil came. A discussion of the significance of the composition of the tlora and the range of the species will receive attention in coniuH-tion with the subject of the age and e(|ui\alents of the formation. Hcononiy of space forbids the description of the species in this report. Many of them are new. while many otliers have been the subject of careful revision. A few only will here recei\(' any systematic biologic treat- ment. The descriptions of all the material in hand are now complete, and will form part of a monograph, in process of preparation, on the flora of the Pottsville formation in the Appalachian i)rovince. Their ])resent publication \voul rcfcral»lt> to, and proba))ly lies within, the time interval marked hy the UpptM' Lykens group of coals. We shall, accordingly, in f urtherdiscussions, treat this group as belonging to an Upper Lj^kens division. Of the \'2'> species of plants found in this division, l)ut 13 or 14, including the gynniosperms, are conunon to the Lower Lykens division, while 1».5 are, so far as ol)served. confined to t\w Upper Lykens division. There remain two Ncrticaily restricted portions of the type section <(4 FLORAL ZOXES OF THE POTTSVILLE FORMATION. for further eon^iclonitioii. Thv lower of i\\o>o two. (■nii)r;K'ing beds E-G, from 570 to 640 feet below the Twin eoal. has furnished a flora of from IT to 19 speeies, 10 of which are common to the Lower L^'kens division, and C> or 8 to the Upper Lykens division. The lowest of these beds, (UO feet below the Twin coal, is bound to the Lower Lykens division by the presence of XraropterU Pocahonfas var. incB- qiHil'is^ which is not, 1 believe, present at any point in the Upper Lykens division. Eremopteris Cheathaml belongs in the upper por- tion of the formation, or in the Upper Lykens division, as, by its gen- eral distribution in other regions, does also AJ^'fliopt&rh grandifoUa. Ti'igonocarjmni Heleme is, in general, rare in the Upper Pottsville of other regions, it being largely characteristic of beds of nearly the age of Lykens coal No. 4. Similarly l)ed F, 50 feet higher, is bound by Aldliopter'i^ protaqu'iJ'inti and Xeui'opterh PocalionfaH to the Lower Lykens division, and perhaps more closely by Ercinop)teris dec'ip'ieiu^ Xi-nroptivlx ttnnesKeeanii, and CaWipterklium alJeghaniense to the Upper Lykens division. The Eremopterids and Mariopterids are largely characteristic of the upper Pottsville, while Callijyteridium aJl<'ghanl(n\-^r generally occupies a lower place in the sections in other regions. Considering the mixed composition of the floras of these two 1 )eds, it seems most expedient to regard them at present as belong- ing to the interval between the floras of L^'kens coals No. 4 and No. 3. lied G, which is l)ut :20 feet higher than F, is temporarily placed in the same rubric — the Lower Intermediate division — between the I'pper Lykens division and the Lower Lykens division, on account of ignorance of its flora. Should additional material come to light in this l)ed, which as yet has furnished but one fern species, Xtn(roj>teris ucvfoniijiifii.na, it will pro])ably l)e found referable to the Upper Lykens division. The remaining uppermost plant beds, ]M and N, in the type section at 245 feet and 2l<) feet, respectively, below the Twin coal, have yielded as yet but 8 species, none of which occur in either the Lower Lykens division or the Lower Intermediate division. Four of the species are, however, common to the Upper Lykens division. Of the 8 species, viz, P$.eudopecop>teris cf. s])t('ris species appear to lie close below the base of the Lower Coal ]\Ieasuresat "Campbell Ledge'' in the Northern Anthracite held. The phase of SpltenopliyUnni cunei- fal! 11,11 found in these beds is that common near the T)ase of, but within, the Coai Measures. In l»rief. it is evident that, while several of the species from these lieds are conniion to the Ui)per Lykens division. WHITE.] PALEONTOLOGIC DIVISIONS OF THE FORMATION. ( ( 5 the flora as a whole is perhaps more closely united with that of the Buck Mountain coal and the succeeding- Lower Coal Measures. These two beds, which are obviously younger than the Upper Lykens division, yet are not less than 200 feet below the Twin coal, will, therefore, be treated as representative of an Upper Intermediate division, which, as will appear later, in the discussion of the floras, seems to be transi- tional to the Lower Coal Measures as that formation was defined })v Rogers and has since been commonly recognized. The combined distribution of the species between the four divisions of the Pottsville formation somewhat temporarily proposed above is condensed in the four columns at the right-hand border of the table. It may be sununarized as follows (p. 790) : 776 FLORAL ZONES OF THP: POTTSVILLK FORMATION. Table i>}i(nfiiifj the ohtffrred fjeogmphir awl xlrnlifinipJiic |\,,Ti-— Till- liv]iln-ns }ii't\veen the numbers of the coals indicate that the fossils were collected the species associated on a slab or rock fragment to identify the horizon Lincoln-Lykens mining developments. Lower Lykens Coal group. Upper Lykens Coal group. Name* of species, strntigraph- ically arranfjed in • two groups. 6 be i 6 a 33 1 s "3 c o e 5 T O £ 1 in 1 6 o 1 -c ■55 1 6 .a 0 2 1 8 a c 3 ■f "3 a a "3 3 CO 1 u c c 3 "3 5 ■V 'x 0 3 ~ "3 1 'A 1 Groip Ko. l—Fmui Lower Ly- Arw.*! scelimi. 4 1 1 X X X? 5 1 ilariopteris eremopte r o i d e .s D W X X? X 5 '? X 5 4 5 or 6 Mariopterispottsvillea 1", W... ■- X X •? X Sphenopteris asplenioides .Stb. .Sphenoi)teris Monahani D.W.. .Sphcnopteris paten t i s s i m a •- X X X 4-5 X X X X X X X X X >: 'a or .=1 4 or 5 4or5 4or5 4or5 6? X Sphcnopteris novalincf)lniana Sphenopteris Lutherianii r>.W. ■■ ■'~ X 4? l*o('opteris sernilata H a r 1 1 -1? 4? X X .... ■ X X 1 . X X Neiiroi)teris Pocahontas IJ. W.. Keiiropteris Pocahontas var. X X X X '} 5? .5 or 6 ■5 X 5? 5 X .... 5 4 or 5 4 4 + ■- .... -- Neuropteris Pocahontas var. X 4 X 4 X X X 4 X 4 Asterocalamites scrobiculatus {.""chloth ) Zeill Calamites Kocnieri Goepp Astcropliyllites parvulus Dn.. Calamostachvs cf. lanceolata Lx.? : X X X X X 5? X - .... X >' RANGE OF PLANTS IN THE TYPE REGION. t i i niii'je of plants vitli'm tJ/r Paltxrilh' fnnnntion in the ti/pr region. from rock dumps of iiiiiiKleil luatfrial irom those coals. Whenever it is practicable, by means of or coal, the number of the hitter is placed opposite the species in question.] Beds n. tde finit ely ( orrelated with Lylvens g •on J) <■ o cc a .S p. 5 o o ft g 6 3 >. 1-1 tH o X X X X X X X X X X X X X X X X X X X X X X X X X X X 0 '\ 1 a ft g 50 1 ■o a '^ s> X X X X X X s i ft £: 1 ft O ft be "3 o O 3 ft ft X Pottsville Gap— type section. Sharp Moun- tain. Broad Mountain. M oT P. o "3 ■t, o 'C o 3 3 r* 1) ft a 3 t< 50 -2 ■3 1 1 i o Q o 1 o 1 m « c a o d o c "3 o ■o C o 5 in &^ 1 SB O o _o i ■Jl c o _C '^ H o "3 i 1 C S "3 o iC 1 C _o 3 a. a i CO J o o C % o 3 Si a> a) 1 c 'S o 3 % ,a> o 'C S 2 3 2 X 5" 4^ "3 s a 1 1 d c c 3 3^ X 3 1 3 a 1 'A 5" 1 X 1 o X 1 ,, X X X X X 1 1 ^, 1 1 X X X 1 . ! ?. X X X X - X 1 1 , 1 1 1 i X 1 X X 1 - X? X X ? 1.... X i !X X i X 1 X X X X * X 1 V X X X X 1 ' 778 FLORAL Z()N>:s oK THE POTTSVILLE FORMATION. Tdhir xliDiriiHj iJir nJiscrnd (jfoani]>Jii<' (nid slriitlfir(ij>}i!r rniu/r of Lincoln-Lykons mining developments. Lower Lykeiis Coal group. L'pper Lykens ■ Coal group. Names of species, stratiprapli- ically arranged in two groups. s i 50 J, q s s i > 5C c c g "3 o c "3 2 c "c o 13 g o c B 3 1 s 1 o o •ES 1 "«j5 o 1 o s m 1 c o o c 13 1 3 ci. "3 o u c "o c 3 "g O •C X g « I.. O o ■c o S 1 s Z; 3 3 GRorp No. l—From Lou-n- Lijkenn section— ConV A. X X 4? 4 4or8 X ? ? ■ Lepiiliidriiilnin alaliiiiu en se ? 5? X LepifloflendronVeltheimiiStb 4? X X 4 4 or 5 X '.. 5? Lepiflustrobus iiennsvlvaiii- eus I) \V X X 2or3 2or3 .. Leiiidostrobus cf. ornatus L. it H X X X Lef)iflr>phyllnni miinninion- X X 4or5 1 Lefiidophylliim laneeolatum X -• 5? X 5or6 X X Sit'illariH ichtynlepis d'resl) ■■■■f" Sipillarinstnilais ? iiieertus 1». X X X .... X var X X X var X X X X Cardiocarjion e 1 o n g a t u ni Tr i g n n o c a r p u m ampull»- X X X X -■ var. 5? var X var 5? X X X var X var X var X lor 2 4or5 var X var Trigonooarpiini Helena- 1). W . Trigonocnrpnm Dawsfjnian- nni I> W Rhalidocarpos speeiosiis D. W. X X X X X X X X X X X X X X X X Carpolitlies oriza-forniis Lx. MSS ^ X GRorp No. 2 — From 1'pprr Lykcnsgectioii. Eremopteris snbelegans 1). W . Ereniopteris sp. No. •> .. WHITE.] RANGE OF PLANTS IN THE TYPE REGION. 779 jjlautit irlthhi tJie PottRviUi- foinnatlon in the type region — Continued. Beds not definitely correlated with Lykens groups. _5 a a a ft >, d ft 2 bD a &: O o ft i cS O c § a u a ^ o >A X X X X X X X X X X X X X X X X X X X X X X X s a a ft a ft 1 a 1 s t 1 • 5 1 *^ ft &£ t; ft 5: .a ? X >: X X Pottsville Gnp — type section. Sharp Moun- tain. Broad Mountain. .5 2 ft s a ft o ft o 'C m a a o a o a p Si a ,3 a a 3 5" 1 0-. < i O % a a 1 o 1 1 d S c '^ _o "3 O Q s 3 o o 1 o S o o o pq "3 o d 3 i) c '^ "3 o g s 1 a - o "3 o m 1 o ■a c _c £i o ►4 ■d w pq "3 C c % o "3 O) 1 c o o a o g 03 a 1 a OS O 1 a' 2 1 2 1 "3 a 1 g m o a ft p .a a P d ;z; c S a a o S 1 ft "^ a ft ft ft O a X X X X X X X X X X X X X X ? X X .... X X X X var var X var X X X var var 7 X var XX XX ...x 1 •- X X ■ X X X X X X X X X X X X X X X x X X X X X X X . X .... X .... .. 780 FLORAL ZONES OF THE POTTSVILLE FORMATION. Table slioiriiif/ the obserivd fjeoynijiJiii- nud stratlfjrajjJiic ranye of Lincoln-Lykens mining developments. Lower Lykens Coiil group. L'pper Lykens Coal groui). Name of species, stratigraph- ically airangt'd in two groups. i 'r. i c 1 "3 1 o o o" 9 a t iC 1 ■ 0 — .id o g c I i c c a "3 o (3 8 c 3 ! i 3 'x im' c o 3 S GRorp No. 2.— From I'nper Li/keiis section— Cont'd. Eremopteris di.«secta D. W Eremopteris liiicolniana D. W. 2 2or3 2 or 3 ■? 2 or 3 2 Eremopteris Cheathami Lx . .. Eremopteris deeipiens (Lx) x 2 or 3 ? Eremopteris Aldriehi D. W Mariopteris Phillip.^i D. W Mariopteris Phillipsi var. intermedia Marioi)teris i)vgma;a D. W 2 1? 2 or 3 2 or 3? 2 ■- 2 or 3 X Mariopteris nervosit (Brongn.) Zeill. var. lincolniana Mariopteris tennesseeana D.W. X Mariopteris tennesseeana var. liirsuta Mariopteris ef. acuta (Brongn.) Zeill rse>idopeeopteris obtusiloba (Stb.) Lx. var. mariopter- oides Pseudopecnpteriscf. squamosa (Lx.) D.W Sphenopteris Lehmanni D. W. . X i>phenf)pteris Kaercheri D. W.. 2 Sphcnoiiteris sinnilans I). \V. . . Sphenopteris s] > X Sphenopteris d i v a r i e a t a (Goepp.) Geiii. i\; Gutb 2? 2 or 3 2? •• Sphenopteris niierocarpa Lx. var. dissecta ■ 1 Sphenopteris Harttii I)n ....!.. Sphenopteris subpinnati Ada D. W ' Sphenopteri.s furcata Brongn.. X 2or3 X Sphenopteris Rovi Lx Sphenopteris novalincolniana D.W 1 X V •• Sphenopteri.s p a 1 m a t i 1 o b a D.W 1 wHiTK.] RANGE OF PLANTS IN THE TYPE REGION. plants u'iihtn tlw Pottsville formation in the type region — Continued. 781 Beds not definitely correlated with Lykens groups. 0 a> ft a t^ tao 0 0 ^< 0 ft 3 s be 1 3 J 0 5 >-. 0 .c ft be 2 0 1 5 OJ a) ft ft Eic I g. ft bi. c 0 ft X X X X X X X X X X X X X Pottsville Gap — type section. Sharp Monn- I Broad Mountain, tain. ; « c a c3 Q oT ft ft =e 0 ho _c ft OS "3 a "3 o a v. ffl o 0 J i «■ w =3 0 0 c 0 X! •2 0 0 0 0 a "3 0 0 'O m 0 1 fa s pq 0 c 0 0 0 1 0 s s s 0 "3 0) iC ■■0 "3 0 '^ H "3 '> "3 1 m "3 ■£ CM D 05 '$ i-' "3 0 m ft 33 *S "x ft 0 1 3 X 0 1 s P3 "3 S 0 g CO c ft -a is 0 d c 0 S 0 S. a £ ■ft ft .ill ft ci a; 0 0 i ■3 g a 5* .... .... X •> X .. 7 X X -- X X X X X X X X X X X X X X X X X X X X X •? X X X X X X X X X X X X X X •? X X « ? 1 1 ? ....' 782 FLORAL ZONES OF THE POTTSVILLE FORMATION. lable shoiciinj the ohxerved geographic and slratigrapJiic range of Lincoln-Lykens mining developments. Lower Lykciis Coal group. Upper Lykens Coal group. Names of species, stratigraph- ically arranged in two groups. o" a a c D 5 c o 1 1 5 o aT 2 c S5 1 1 6 s" o 3 5 o .s c £ 6 o 2 i !2 c 2 M o u s o o o o ;5 6 c I c CO 1 o o 5 o « c a ■a i ■A + Groip No. 2.— From Upper Lijkats scciion—QoxiVa, Sphenopteris palmatiloba var. 1? X 2or3 lor2 X 1 X o A let h o p t e r i s 1 0 n c h i t i c a 1-3? 2or3 2or3? 2 or 3 X Ak'thdptcris lonohitica var. \ X 1 1 X •) •■ X Alcthopteris Serlii (Brongn.) X X .... 2 or 3 2or3? 2 or 3 Alothopteris Evansii var. X 2or3? Callipteridium alleghanicnse D. \V x CMllipteridium snspectum I) W . Callipteridium pottsvillense 1> \\ ' 1 • 2 or 3 X X .... 2 Neuropteris acutomintana D. \V 1 2or3 X ■? WHITE.] KANGE OF PLANTS IN THE TYPE REGION. plants irithln llie Podsvlllc funnation in (lie iijpe rcr/ion — Continued. 783 Beds not definitely correlated with Lykens groups. 1 ft >i 5 o tj ft o Eh ho "3 c o Eh a) ft ft bi C ft X X X X X X X X X V Pottsville Gap— type section. Sharp Moun- tain. Broad Mountain. i « d' a o ft cS o ho .s ft CO t^ X a o a d ft o So V "3 a Eh O u z> ft ft ■=) .S E-i s o o ■< o M o o 2 o ■X 1 "> H o •2 o 1^ -c o M 1 o o 0) o s o S .a "S o o g c3 O c & o '3 l_ d © o o H o 1 o o o i -^ -a 0) o3 c o c H o "3 1 ■a 0) cq o "3 i ►4 o X "3 c c X2 m o "3 o o eq S" o '3 03 d -d § o O d S o .a 1 "3 a; 0) o S o 3 ^3 "3 u d c o a < a c a s 0 X X X X X V X X X X X v X X X X ? X X X X X V X X X X X X X X X X X X X X X X X ? X X < X 784 FLORAL ZONES OF THE POTTSVILLE FORMATION. Tithle )t]ioir'ni(i Oie ohserved geo[irhic mul ."Irnlii/raplilr nint/r of Lineohi-Lyktn.s mining dcv<-lopinents. LowiT Lykens Coal group. ^cFm\ i^o^^^ Nauu's ni spocies. stratiprrapli- ically arranged in two groups. c .A o o s as c 11 1 i ci" g j2 13 s § 1 o o 9 =3 O u 'to 2 6 1 3 1 p "3 3 X 5 3 c-i 1 5 1. .iS o v: o 3 •§ GroipNo. 2— /Vom Upper Lijkcns secdoJir—ConVa. Nenropteris tennesseeana Lx. 2 or 3 .. 1 .. Nenropteris teniiifolia (Sehloth.) Brongn. var. Nenropteris aff. helerophyllii 2? var. X var. .. 1^ 2 .... 2 .. 1 X X 2or3 .. 1 1 1 Calamites approximates i L2or3 2or3 Asterophvllites arkansanns D. ' Asterophvllites pennsylvaiii- 1 Asterojihyllites cf. rigidu-^ Animlarirt platiradiata Lx. M^S 1 II.... 1 A 11 imlaria latifolia (Dn. 1 X 1 Calaniostachvs Kno\vUoniaii:i D W 1 ■■■■i Palaostaehya alabainensi.>i I > i 1 1 Splienoiihyllum tenerrininm 9 .S]illc!ii^|i:;\ rum liil'Mrraliii: 1 WHITE] KANGE OF PLANTS IN THE TYPE EEGION. plants iiillihi (lie PoHsvillc fonimtion in the type region — Continued. 785 Beds not definitely correlated with Lykens groups. o •a c O & &: o o .. ft y 2 60 1 o 3 o 3 o c fti he: £ 5 c o u (D ft >. Sc ft 1 X X X X V. X Pottsville Gap) — type section. Sharp Moun- tain. Broad Mountain. 1 i « 3' c _o "x a. O bx) •B 1' X 3 a i) a ft D 2 ■J -3 s. ft "3 o c % .n o in o m "3 8 o o > 1 o d o 1 o o i> o o o c '3 o o o ■c m o o a '% .a o g M ■? o ID O in d o M 1 o "3 o i K w -o a pq X o C '^ "3 X! a -ill 1-5 •a « X o "3 in 5 3 n 1 35 3 3 Si ID (D 1 c s "3 .2 o 1^ 'O S X £ o X 1 o 1 1 1 13 8 o i-, m '? § o "x O 2 1 o 1 O 1 o "o o 3 "3 o c o B 3 i = 1 Cardiocarpon late-alatum Lx . Cardiocarpon disculum 1). W . Cardiocarpon orbicularo Ktt . . 1 X 1 1.2or3 1 Cardiocarpon cornutum I)n ... 1 X Cardiocarpon e 1 o n g a t u m Newb. var. intermedium Cardiocarpon annul a turn Newb Cardiocarpon Wilco.\i I). W ... 1 1 Cardioearpon Girtvi I). W i . 1 lor2 Cardioearpon obliijuuin l)n ... .. 1 X X WHITE.] RANGE OF PLANTS IN THE TYPE REGION. plants within the Pottsville formation in the type region — Continued. 787 Beds not definitely correlated with Lykens groups. 5 0. 1 60 O o u o . d, 3 s bo M a o >^ ij o 5 CO p. x; 3 o. 3 o ti V > 3 5 1 ft ft 1 be '-J ft 5: % o X X Pottsville Gap — type section. Sharp Moun- tain. Bread Mountain. a 3 Q K o S" o so u "3 p s o o O o c 'S H & o "o i o M o o ; o d a> m o o o o m o o o ■o c4 o o 'S "3 .o o o o ft o o i d n 5 5 o o s M -a c m s "3 XI >* o H o s 1 m m C '% 0 i CO -a « X cS c o CP X o "3 o ■a o M ft S 0) S o O X! o 1 1 "3 t o S 2 « o 3 1 X! S O 3 1 oj O C E g. ■a .2 w o ■ft 3 o io g i: X X X .^ •? X Y X X •? V 1 ■? X X X .> ' ■? X X X X? X ■ - X X X X X X X X X X X X X X X .... Y X X X X X X X X .... X X NEW FLORAL ZONES OF THE POTTSVILLE FORMATION. Table .•ihoiriiHj Ihe oitscrred gi-of/rdjihic uml slr^ ft s tie 0 t.( 0 ft 1 60 6 a 5 c 0 0 ft >> i) ft 2 to 2 B c 3 5 i >, c ft i ft & 3 ft X X X X X X X X X Pott.sville Gap— type section. Sharp Moun- tain. Broad Mountain. 1 m d ft 3 4 V ft 0 ft s! bo a 'C ft 02 0 "3 _s 0 0 ft 3 2 1 a &.< 0) •— t t.1 ft 3 "3 o V a o "3 < o c "3 X! ■2 i m PQ "3 •2 d o P3 c c; "3 o o £^ P 1 a o o ■a o M o C 'S "3 (D o i o o o "3 o 6 o C % Eh ."3 'X! o 0) o s s o m "i .s a) CO -a 0) c o i o in o m o "3 ,v 0) W X 1 C o 1 o o3 O o C & ■3 .Q 0 <2 0 pq g ■33 2 s si 0 1 a 6 3 1 f m 0 "sir a. ft a, ■c ci CD 3 w .2 "c 0 d c S 0 3 3 0 a ce 0) C ■p. 01 ft ft 3, ■o .2 D ;.! 3 d 2 0 X X X X X 790 FLOKAL ZONES OF THE POTTSVILLK FORMATION. Analiisif! of till.' (llt plants over the "zero"" 1)0(1. the lowest of the Lykens coals, no precisti information is at hand. At Willianistown.' the onh" point at which the l)ed has l)een exploited, but little mining was ever done in the coal, and the bed was so long ago abandoned that it has not been practicable to obtain any fossils therefrom. It appears prol)alde, however, on account of its proximity to the succeeding coals (38 feet below No. 6), that little difference will be observed in its flora, fragments of which may possibly have been gathered from the rock dump at the Williamstown mine. Likewise the mingling of the roof shales from the Lykens coals No. 0, or the "Little l)ed,'"' and coal No. 5, in th(> rock dumps, as at Big Lick, AVilliamstown, and the Brookside mines, renders it, for the most i)art, impossible to discrimiiiat(> l)et\veen the fossils from tliese coals at this point, although it has fortunately been possible to procure distinct colkn-tions from the higher coal. A small collection from the roof of Lykens coal No. (t, within the mine 'Atlas Southern Anthracite Field, Pt. Ill, mine sheet xix; Pt. VI, cross-section sheet xx: Pt. IV, columnar-section sheet vii, section 8. WHITE] FLORAS OF THE LOWER LYKENS DIVISION. 791 at Brookside, was, however, obtained through the cooperation of the Philadelphia and Heading Coal and Iron Conipan3\ Excluding- AMi Ittleseya Camjjbelll^ which in some form is nearly everywhere present in the Pottsville formation, and Sphenopteids patent Iss hn a ^ which is more abundant in the neighborhood of coals No. 5 and 4, there remain yeuropterh l^tcahontas var. pentlas. and Alethopteris coinpos'ita^ the latter of which has not been found in any other bed. The variety of Xeuropteris Pocaliontas has, perhaps, not been seen from a higher level than the roof of No. 5. Its presence, accordingh', in a coal but 50 feet lower is quite natural. It must therefore be understood that the shales attending coal No. G have not yet revealed any floral characters of value, though the default may be due to lack of specimens known to have come from this level, rather than entirely to its propinquity to coal No. .5. FLORA OF LVKENS COAL NO. 5. The collective flora from the roof of Lykens coal No. 5, also known as the ''Big bed,'' or the "Lykens Valley bed,"'^ may readily be com- piled from the flrst section in the table of distribution. As will be noted in glancing at the vertical range of the species in the latter, a large portion of the plants are common to the flora of coal No. 4, while others, perhaps exclusively from this bed, are represented by specimens whose mingling in the rock dump with material from No. 6 deprives them of any present stratigraphic trustworthiness. Among the specimens definitely known to come from the roof of this coal the most important species are: Mariop>terls eremopteroides^ Splienopteris a.p)lenloides, S. patentlssiraa^ Neuropteris Pocahontas and its two varieties, Calaniites Roemeri, Asteroj)kyllites parvulus^ Lejj'idophyllum quinnimontanum, Lejjidoj^hylhmi Janceolatum var. virginianum^ and Sigillai'ia l-almiana. Eremopteris sp. No. 1, Calamostachys cf. lanceo- lata^ Lep'idodendron. alahamense^ and a S'lglllaria which I refer, Avith a little doubt, to .S', h-htyolepis^ are species of restricted range, but the circumstances attending the collection of some of the fossils make it uncertain whether their source is exclusively in the region of coal No. 5. The most a])undant and characteristic species is Neuropteris Pocahontax val'. pmtias^ which rarely fails to be present in large num- bers, even in a small collection. The variety ina'pialis, which is more connuon in the roof of Lykens coal No. 4, is also present. Another form, which throughout th(; Wiconisco Basin seems to bo confined to the same stage, is the ]K"diit\hi\J/arlopter(s ere/j/opteroidrs illustrated in PI. CLXXXIX. Sp/ienopteris a.splenimdes here, as in other regions, exhibits the diminutive round-lobed or Dickaonioides type common in the ' "Xo. 2" in the iiomeiiclature employed for tliis vicinity in Rogers, (Jeology of Pennsylvania, Vol. II, Pt. II, 1858, p. 192. 792 FLORAL ZONES OF THE POTTSYILLE FORMATION. lowor portion of the Pottsville formation. Sj>/u'?)oj)te?'i.s jx/fe)7tissima appear.s to he present at this level, though more common in and charac- teristic of the roof of coal No. 4. Calamostac/ti/s- cf . lanceolata^ collected from the East Brookside rock dump, ma}' ])c assumed to have come from No, 4, its occurrence in other regions being as high as the flora of No. 4, or even hig-her. The same is true of Lepidodendron aldbmnense. Le2)idoj)hi/Uiun hniceolatinn var. virg'nviaimm and Lej^f'dojyJiylhnn qui)i)il))iontanu7n are essentially characteristic of the lower Pottsville in all regions, but the former is usually predominant in beds lower than those in which the latter is common. From an examination of the table, as Avell as from the inspection of the collections, we may conclude that the almost invariable appearance of 2[nr'wj>terisi eremopteroides or the lax form, var. incequah's, of JTeu- 7'oj)tens Pocahontas^ as well as the occasional presence of a number of the ferns slightly more characteristic of No. 4, serves to distinguish the stage of coal No. 5 from the basal portion of the Pottsville formation, while the species first mentioned, together Avith the always abundant Neuroi^tei'is Pocahontas \?a\ pentias^ Neiiropteris Pocahontas ^ q.i\^ Lejri- dophyJlum lanceolatum var. virginiannm^ likewise assist in discriminat- ing between the floras of coals No. 5 and No, 4. In the latter differ- entiation the absence of those plants which are characteristic of the roof shales of No. 4 is perhaps an (equally valuable criterion, though negatiA'e in its nature. l'AL?:()NTOLO(Ut' FEATURES OF THE KOOF SHALES OF COAL XO. 4. The flora of the roof shales of Lykens coal No. 4 is perhaps the most readily recognizable among- those attending the series of the Lykens coals. While it contains a large percentage of species that are also found in the roof of No. 5, as ma}' be noted by reference to the ta))le, and while its facies is distinctly hnct-i' Pottsville as compared with floras of coals 2 or 3, it almost invarial)ly contains several species Avhich arc not only restricted to nearly this level in the Southern Anthracite held, l)ut which are also observed, similarly associated, and restricted as to vertical range, in other regions. These are Aneimites 2Jottf<%u7h'nsis, Jfarioj/te/'is jMfftsviNea, Aleth( pterin protdqiiUina^ Neu- mptti'ls Snilthsii^ YoIl'iiKintiia ci'dMsa, and Sphmoj/hi/Z/i/ni tenue. Additional species, hitherto found only at tiiis level in the Southern Anthracite fleld, are Sjylnnopterls lioralhicolniana^ S. Lutlu-rhina^ PJiiijtdoniipos sj/t'C/osus^ and a form of Sporangites. Of the species which are found at other horizons, Sphenopteins pateniissima^ the small form of .V. a.^phnnoldes^ AJoiopterisgeorglana^ Ltpidodoidron (da- Jxniiensi'^ and Trujonocarpum Ildtmcv. are more especially common in this stage, where they are generally associated with the species flrst enumerated. In fact, Aneiiintes j'ottsvillensis^ Mar'wpterls pottsvillea^ WHITE.] FLOKAS OF THE LOWER LYKENS DIVISION. 793 Keui'o2)teris Smithsli, Sj)henoj)hi/Uum tenue, Lepidodendron alabametue^ and Triq&mcarjpum Ilelence are among the characteristic species of this stage of the Pottsville formation throughout the Appalachian region. The most common, and therefore the .most useful of these in the Southern Anthracite field are 21ar!opteris, Xeuroj^teris, and Spheno- 2)hyUuiH. In fact, if we were to employ a paleontologic term for this portion of the section, the latter migiht appropriately be designated the Marlopterk poUsviUea zone. COMPARISOX OF THE LOWER PORTIOM OF THE TYPE SECTION WITH REFERENCE TO THE LYKENS COAL HORIZONS. Owing to the densely conglomeratic constitution of the basal portion of the f oi-mation at the type locality, comparatively few fossils have been obtained from beds A and B, which clearly belong to the Lower Lykens division: but while the materials from bed B of the section are quite insufficient to form a basis for horizontal comparisons, we find that the presence of 2fariop)terls sp. No. 1, which is perhaps insepara])le from Mariopteris eremopteroides, Neuropterh Pocahontas, and Spltenop- teris pafentlssima in bed C, 770 feet below the Twin coal, is entirely compatible with a stratigraphic position not far from Lykens coal No. 5. At the same time, the presence in bed D, 710 feet below the Twin coal, of Mariop)teris jyottsvillea, Sphenopterh dadeana, Aloiopterisgeor- glana, Neuropteris Pocahontas var. inmqualis, Muropterls Smithsii, and SphenojyhijIJuni tenue strongly points to a place -near the level of Lykens coal No. 4. Although several species are known to occur in beds above No. 4, they are rare in higher horizons, while the more important percentage of species which appear to be largely character- istic of the Lower Lykens division more than counterbalances them. Of greater correlative value, however, are such species as Mariopteris pottsvillea, Neuropteris Smithsii, and Sphenophyllmn teriue, which are in general characteristic of the horizon of No. 4 coal, and whose evi- dence is strengthened by the accompanying species enumerated above. In short, the plants of "bed D indicate a horizon approximate to that of the Lykens coal No. 4 (White\s bed), about 710 feet below the Twin coal in the type section. It is probably not higher; it may be slightly lower. Bed C, on the other hand, 770 feet below the Twin coal, is probably lower than the No. 4 coal, and may have been deposited at the same time as the Lykens coal No. 5. I do not wish to be understood as regarding the coals adjacent to beds C and D in the type section as unquestionably identical with Lykens coals 5 and 4, respectively. The obvious variation in the beds of the Pottsville formation, especially as regards the number of the coals, as shown in sections located but a few miles distant, and the frequentlv observed entire disappearance of the principal coals of 794 FLORAL ZONES OF THE POTTSVILLE FORMATION. the tield, as iwoak'cl hv l)oi'iii*4S not far from the LiiK-oln region.' eun not fail as convincing arguments against the free application, in the Pottsvilh' formation of the Southern Anthracite field, of the correl- ative methods employed by geologists working in the interior of the Appalachian trough, where, in the several l)ituminous ])asins. the beds are relatively uniform and clearly persistent over great art^as. That one of the thin coals occurring in the Pottsville Gap section nearly 700 feet below the Twin coal is contemporaneous with and equivalent in point of time to a portion of the Xo. -i coal at Lincoln or Williams- town is perhaps not improbable, since the favorable conditions for exclusivelv carbonaceous deposition luay have 1)een synchronous at both points, and the testimony of the fossils points toward the latter. It is lilvcwise possible that the 1-foot coal accompanying tlie 2 feet of dark shales at bed C may represent Lykens coal No. .'). It is. how- ever, extremely improl)able that either of these coals extends in a con- tinuous carbonaceous terrane from the type section at Pottsville to the very valuable deposit in the Lincoln district. The stratigraphic position of l)ed A in the topmost stratum of red shale at the ])ase of the section probably justifies the assumption that it is older than any of the Lykens coals. Its very small flora, of Lower Carbon if (>rous facies, appears to warrant this assumption, altliough it is too mcniger to serve as a foundation for satisfactory comparison. FLORA OF THE LOWER INTERMEDIATE DIVISION. milKF EXISTENCE OF A TRANSITION EEOKA. The relative distinctness, from a stratigraphic stanflpoint. of the floras of the Upper Lykens division, as compared with those of the Lower Lykens division, has already lieen remarked in connection with the proposed subdivision of the Potts\ille formation according to the concomitant grouping of the economic coals and the fossils. It may be noted at this point that the paleontologic difl'erenct^ l>etween the lower and the upper groups, which, excluding th(»gynmosperms and certain vertically widely distributed Lj'copodiales. have comparativeh' few species in common, is probably due in part at least to the interval 1 (('tween Lykens coals Nos. 4 and 3. which is about 250 feet in the Lin- coln region. This interval, of wiiicli we have from the Lincoln region no paleontologic re})r('sentation in the collections, and which is there- fore not assigned to either the upper or the lower division, still rcMuains accordingly a paleontologically unknown (piantity. Yet. notwithstand- ing the inferential conclusion that it contains a transitional mingling of U})per and Lower Lykens floral characters, such as occurs in the inter- \Til (Lower Intermediate division) l)etween .570 and 700 feet below the Twin coal in the type section, the vertical distance involved is 1 See the records of diamond-drill lx)re holes on Broad Mountain, platted in groat detail on colnninar- section sheet ix, Atlas Southern Anthraeile Field, I't. IV. WHITE] FLORA OF THE LOWER INTERMEDIATE DIVISIOlSr. 795 comparativeh' so little, when we at once take into view the entire sec- tion and the notable differences between the iiora above and that below, as to strongly emphasize the rapidity of the specific floral changes which it masks. As has previously been mentioned, the plants of bed G of the Pottsville Gap section are probably referable to the Upper Lykens division. Should plants be collected from several })eds between coals Xo. 3 and No. 4 in the Lincoln-Lykens mining districts, it is not improbable that some of the terxanes will show closer paleontologic connections with one division or the other; but the plants in the inter- val in the tvpe section, though few in number, indicate that within certain limits of a relatively thin zone of the sections the boundary, if drawn as between the Lower Lykens division and the Upper Lykens division, will be largely arbitrary- through beds with a mixed flora. The case in hand well illustrates the rapidly changing fades of the floras of the Appalachian region during Pottsville time. FLORAS OF THE UPPER LYKENS DIVISION. FLORA OF LYKENS COALS NOS. 3 AND 2. Fossil plants have been collected from the roof shales of Xos. 1, 2, and 3 of the upper Lykens coals in the vicinity of Lincoln. It has been impossible to make a collection from coal No. li, since the New Lincoln collier}^ where it was formerly slightly worked, has for a number of years been abandoned, the mineral from the other beds on the property being brought to light at the Lincoln mine. Accordingly, while it is not impossil)le that stray specimens from this bed may still have been accessible in the rock dump, it is probable from the very small extent of the workings that few, if any, were collected. At least it has not been possible to recognize such, and the specimens, if pres- ent, are presumably included in the column of the table devoted to the stratigraphically undifferentiated material from coals 1 to 3, inclusive, at the New Lincoln mine. Of the floras derived from the upper Lykens coals, by far the most interesting are those associated with the neighboring coals, Nos. 2 and 3. The proximity of these beds, which are separated b}' but 3 inches of dirt at the New Lincoln mine and by strata probably nowhere far exceeding 30 feet in the Lincoln workings, results generally in the removal of both coals at once and the mingling of the roof shales of No. 2 with the parting between Nos. 2 and 3. Separate collections Aviu'e. however, obtained from both, that exclusively from the parting, which may be regarded as the roof of No. 3, being procured at the Lincoln mine, while specimens from the cover of coal No. 2 were gathered at the North Brookside slope ^ in that bed and from the lower 1 Atlas Southern Aqthracite Field, Pt. Ill, mine sheet xvii: Pt. IV B, columnar-section sheet x, sec- tion 8; Pt. VI, cross-section sheet xix. section 2,'>. 796 FLORAL ZONES OF THE POTTSVILLE FORMATION. Eureka drift/ Of the entire 125 species so far discovered in the shales over the coals of the Upper Lvkens group, not more than li or 15, inchiding the vertical!}' widely distributed gymnosperms, are found in the Lower Lvkens group. By reference to the preceding table of dis- tribution, it will at once be seen that of the fern Hora l)ut 2 species, SjyJirnoptt'i'ls asj)]enioides and Peeopterh serrulata^ present in the Lower Lvkens division, are also apparent!}^ present in the roof shales of the mined upper Lvkens coals, or in the beds of the type section, which, on the paleontologic evidence, I refer to the same division as the upper Lvkens coals. CaJamltcs Roemerl^ AHterophyUites jKirrvlus^ Lepidodendron clypeatum^ Lepidostrobus pennsylvcmicus^ and the live gymnosperms, which occur in the Lower Lykens division, have a rela- tively wide distril)ution in the formation. Sp/ieNopteris asplenioides appears to l)e extremely rare in this division of the Southern Anthra- cite tield. though it occurs as a large form in beds of the same age in the southern Appalachian region. Pecopterissei'ridata is usually com- mon in beds of this age. Sphenopliyllum tenue is, on the other hand, extremely rare at so high a level; it, like Trigonocarpuni Helence, being usuallv characteristic of the zone of No. 4 coal, or lower. The zone of coals Nos. 2 and 3 is, in general, especially characterized by the presence of broad- or round-pinnuled forms of Ereinojjteris^ In' forms of 2far!(>pteris approaching the original rtiurlcata type; by the large number of Sphenopterids, especially of the Hymenophvllous group, as Avell as by a Pecopteroid form; by the presence of the large, lax. and distant-nerved Alethopterids of the types of A. discrepans and A. grandlfolia,' by the Jlegalopieris typea ', b}'^ the £^lrodi a,nd (//'(/heuopJiylIa; the numerous gj'mnosperms, including Cordaites and the broad-leafed Whittleseyce; as well as l)y a great a])undance and variety of fruits. As more peculiar to this zone, specitic mention should l)e made of^ — Ereniopteris Cheathaiiii. ]Mari()j>teris pygmtea. Mariopteris tennesseeana. Sphenopteris Lehmanni. SphenopteriH Kpprcheri. Sphenopteris divaricata. SithenopteriH Harttii. Sphenopteri^i Royi. Sphenopteris pahiiatil()l)a. Alethopteris Lac(jei. Alethopteris grandifulia. Alethopteris P^vansii. Callipteridium alleghaniense. Neuropteris Elrodi. Nenropteris tennesseeana. Asterophyllites arkansanus. Sphenophylhini tenerrimnni var. clon- gatum. Sphenophylhiiu l)ifurcatuni. Stigniariopsis Harveyi. Cardi( x-arpc >u Cuyahogfe. Cardiocarpon minus. Carpolithes transsectus. Whittleseya niicrophylla. Whittleseva elegans var. niinijr. 1 station 33, PI. CLXXX. Atlas Southern Anthracite Field, Pt. Ill, mine sheet xvi: Pt. IVB, columnar-section sheet x, section 6: Pt. VI, cross-section sheet xvii, section 23. WHITE.] FLORAS OF THE UPPER LYKENS DIVISION. 797 In addition to the plants specially mentioned above, there remain a nmnber of new species which are as yet unknown outside of the field, and which, as may be observed in the table, occur only in this zone. The species enumerated above, which paleobotanists will at once recognize as preponderantly conmion to the flora accompanying the Sewanee coal in Tennessee,^ are essentially characteristic of this zone of the Upper Lykens division. Many of them, such as Eremopteris Cheatham.i, Marwpteris fennesseeana, Sphetwpterls Boyi, Sphenopteris 'pilosa, Alethopterk E^xinsii^ Neurojjteristennesseeana, CarpoUthes trans- sectus, and WJdttleseya microphylla, have, so far as I know, never yet been found at any considerable distance from this zone in the Appa- lachian trough. The flora in the roof of Lykens coal No. 2 reveals, as compared with that in the roof of No. 3, a slight dift'erence, consisting of the presence of a few species of usually slightly higher occurrence and several forms which, in the Southern Anthracite held, I have found at no other hori- zon. As referable to the former category the following may be mentioned: Eremopterisi decipiens. Eremopteris dissecta. Mariopteris pygm?ea. Alethopteris Lacoei. Alethopteris inagnifolia. Neuro pterin Elrodi. Neuropteris gigantea var. The peculiar elements which characterize the flora of Lykens coal No. 2 at every locality from which a considerable collection of specimens has been obtained, and by which it would seem that, in the western portion of the Southern Anthracite held, the horizon may almost invariably be recognized, include Mariopteris pygmcea^ Sphenopteris Lth man nl Alethopterk Lacoei, and Neuropteris Elrodi. To this group of species may also be added Sphenophyllum tenerrimum var. elongatum^ although in other flelds this species has a somewhat higher distribu- tion, and such, we may anticipate, will be the case outside of a restricted area in the western portion of the Southern Anthracite field. It may b(i noted that even where drifted, at a point about 550 feet below the Twin coal, above the wagon road on the east side of the gap ])elow Pottsville, and at a point along an abandoned tramway near the apex of the mountain on the west side of Westwood Gap, this horizon reveals the same association of species in their identical forms. These species appear to attend Lykens coal No. 2 in the Southern Anthracite field, just as Eremopteris Cheathami, Spheiwpteris Boyi, S. palmatiloba, S. pilom, Mariopterix temiesseeana, AletJiopteris Evansii and Neurop- teris tennesseeana usually occur in the roof of Lykens coal No. 3. 1 Coal Flora, Vol. Ill, p. 853. Asterophyllites arkansanus. Sphenophyllum bifurcatuni. Sphenophylluin tenerrieium var. elon- gatuiii. Cardiocarpon Cuyahogse. Whittleseya elegans. 798 FLORAL ZONES OF THE POTTSVILLE FORMATION. FLOKA OF LYKKNS COAL NO. 1. Lykt'ii.s coal Xo. 1, it will be remembered, occurs at about 825 to 360 feet above coal No. 2, and about 250 feet below the "Buck Moun- tain" coal. The plant association in the roof shale of this, the highest of the Lvkens coals worked in this region, is marked, as compared with the Hora of the Lvkens coal No. 2, by the disappearance of species known to l)e present in the latter, as well as bv the introduction of new forms rapidly approaching the Coal Measures facies. Among the more interesting of the survivors are CordaUes Rohbtl^ Trixjonocarpimi amj)uIhpfor)iu\ WhitUeseya Cainj)hdl(., and Carj)ollthe8orizceforiH{s from the Lower Lykens division, and a form of Aletho2)teris Evansll and SphenophyUiim Vifurcatum from the Upper LN'kens division. An examination of additional collections will no doubt largely increase this number, since it is possible that representatives of all the ante- cedent Coal Measures types, such as Alethojyteris lonchitica^ JSeurop- teris aff. lu'tei'opliylliu and K. glyantea^ present in the zone of No. 2 coal, will eventually come to light at the horizon of Lykens coal No. 1. At the same time, how^ever, it is to be expected that the number of new forms w'ill be correspondingly increased. The forms.w'hich have not yet been found below the horizon of No. 1 include Sphenopteris jxdmatiloba var. squarrosa^ Neitropteris lunata^ Anmddi-iti cH-wtdata^ X. tenw folia var. humi/is^ and JS\ fimhriata. The lirst three of these appear to be characteristic of this zone, while the two last named continue into the Lower Coal Measures. Anmdarla cmjdddfd is most pro])u])ly the precursor of Anmdaria spjhenophylloid^s, which appears earl}- in the Allegheny series. UrrER LYKENS ZONES IN THE TYPE SECTION. It needs ])ut a glance at the names of the species recorded in the column re|)resenting the two approximate beds, H and I, about 550 feet below the Twin coal in the Pottsville Gap, to detect the floral characteristics of the zone of Lykens coals Nos. 2 and 3; while to pale()l)()tanists who are acquainted with the Sewanee flora, whose nearly identical composition has already been noted, the preponderance of common features will at once indicate approximately the same age. The greater ])ortion of the plants recorded from coals Nos. 3 and 2 are also found in the collective material from beds H and I, which have l)een somewhat thoroughly searched. It will also be observed that besides several new species, such as Sphenopterls shmdans, Odlipter- idniiii siixjxctuin, and W Ji!ftleseya Lesciiriana^ not found elsewhere, we ha\'e ('(n'diiltcx (DujiisfifolluK and Ctirdux'arpon in'tnux, which are not reported from the vicinity of Lincoln. A})ove the trolley road, on the east side of the gaj) at Pottsville, about -it'..") feet Ixdow the Twin coal, a drift has t)een dri\-en some WHITE] FLORAS OF THE UPPER LYKENS DIVISIOIST. 799 distance along a thin coal, and in dark coaly shales Avhich contain Mar!opfepi-'< pygim^^a^ AJethoptei'is Lacoei^ and Neuropteris Elrod'i in the facies and association characteristic of the horizon of Lykens coal No. 2 in the Lincoln district, and 1 have little hesitation in suggesting the prol)able approximate contemporaneity, if not equivalence, of the two beds. As previouslv mentioned, the same horizon appears also to have been touched in a trial shaft on the west slope of Westwood Gap. Among the more interesting or important additional species in bed J. the probable equivalent of this horizon in the type section in the rail- road cut at Pottsville, which has been more thoroughly searched for fossils, are Eremopterls Aldric/u', Sphenopferis palmafiloha, S, pllosa, Pecopteris serrulata^ AlethopterU Evansii^ Callipteridhim pottsviUe)ise^ and Keio'opteris hirsidina. Of these, the first three are usually rather more conunon at a horizon a little higher than that of coal No. 3 in other coal lields. Pecopteris serrulata^ which, if the specimen has not been misplaced, occurs in the shales over coal No. -i at Brookside, has hitherto been unknown at any distance below the zone of coals Nos. 2 and 3. Callipteridium pottsiyillense is very close to a species from the ''coal-bearing shales" of Washington County, Arkansas, where it is associated, as in bed J, with a dilated, thin type derived from Alethop)teris EvanslL The Neuropteris Mrsutina is a new species with slender, acute, long-pointed pinnules, strongly suggesting Weu- roj)teris Scheuchzerl^ to which it appears to sustain an ancestral relation. It is the earliest-known hirsute A^europterk. The rather small number of plants from bed K is hardly worthy of special consideration, since their source is only about 25 feet higher than J, with whose flora the}' are in general agreement. It is, how- ever, intei'esting to note the appearance at this level of an Erejnopjteris {E. ffuhelegans) close to E. eJegans^ and a Sphenopteris {S. mixtilis) pro])ably ancestral to the S. mixta of the Coal Measures. The flora of bed L, about 380 feet l)elow the Twin coal, like that of Lykens coal No. 1, is one of the most interesting in the type section on account of the antecedent Coal Measures forms mingled with typical Pottsville types. In Calamites Roemeri, WJt!ttJeseya CampjhMl^ and Carpolitlies orizceformis we seem to have survivors from the Lower Lykens division, though it is possiljle that the name Calamit'^s Suekoivii should be substituted for that flrst mentioned. Omitting the enumeration of other species recorded from the Upper Lykens horizons, at other localities, in the table, it may be observed that, of the species present in bed L, Eremopterls dlssecta^ Marlopterls PhilUpsi^ Annidaria Icdifolia,, Bothrodendvcm. arborescens^ Cordaian- fhn.s sp/ici(fus\ and Cdrdiocarpon annulatum^ characteristic of the Potts- ville formation in other regions as Avell, are unknown in the Lower Coal Measures of the bituminous or anthracite basins of the Northern 800 FLORAL ZONES OF THE POTTSVILLE FORMATION. States. Pseudopecopterls ohtusiloha var. iRarloj)teroides^ Sphmopteris suhpinnatijida^ Oligocarjna (yremdata^ and Neuropteris tenuifoUa var. kumilis seem to foreshadow as many Coal Measures types, while Sp/umopteri.s /areata^ frequently reported in the lower portion of the Lower Coal Measures, is, however, generally more common in the top- most beds of the Pottsville formation. As elsewhere remarked, I have not seen the typical form of Aht/(02:>teris lonchitica in the Lower Coal Measures of the Northern States ; the same may be said of Trigo- nocurpum Xocggerathi As to whether bed L represents approxi- mately the horizon of Lykens coal No. 1 in the type section, little that is detinite can be said. The fact is simply that the flora of each bears nearly the same relation to the older floras, and to those of the Lower Coal Measures, yet there are but few species common to the two. It is not unlikely, however, that the latter circumstance is largely due to the meagerness of the material from the roof of Lykens coal No. 1. The general biologic evidence, treating the subject from the stand- point of the composition, vertical range, individual relations, etc., of the species, would seem to indicate a similar stage for both. Reason- ing from the same evidence, we may conclude that the two beds are referable to horizons not far distant at most. It would also appear slightly more probable that the older terrane may be bed L in the type section. However, very little weight should be attached to so tentative a supposition, even though the latter is supported by the cir- cumstance that the interval between bed L and the Twin coal in the type section is over 375 feet, while Lykens coal No. 1, about 300 feet from the "Buck Mountain" coal, the supposed equivalent of the Twin at I^incoln, approaches within 225 feet of the same horizon at Good Spring. The known variability of the Pottsville terranes is too great to entitle a relative distance of that extent to any serious consideration when the localities are so far removed. • FLORA OF THE UPPER INTERMEDIATE DIVISION. PLANTS OF IJEDS M AND X IN THE TYPE SECTION. On passing to the consideration of the species in beds ^I and N of the type section, it is important to bear in mind that the floras of the roof of the upper Lykens coal No. 1, at the Lincoln mine, and of bed L, 380 feet below the Twin coal at Pottsville, are essentially very distinct specifically from the flora of the roof of the "Buck Moun- tain" (Twin) bed, as will be shown later. The phuits of Lykens coal No. 1 and of bed L, which we have tentatively assumed were nearly contemporaneous, are, in fact, characteristic of a zone in the upper part of the Pottsville formation, and are closely ])ound to the flora of the preceding Lykens coals Nos. 2 and 3, or of beds H, I, and J, WHITE] FLORA OF THE UPPER INTERMEDIATE DIVISION. 801 although htiving little in coinmon with the plant associations of the Lower Lykens division. The small plant collections from the phy- tiferous terranes in the remaining upper portion of the lithologic t3'pe section, which collectively were designated on an earlier page (7T5) the '"'■ Upper Intermediate division," will be found to contain a still larger proportion of Coal Measures species, though yet exhibiting manv forms which are common in the beds of supposed Pottsville age in other regions, and which are still unknown in the Lower Coal Meas- ures. The two beds in question are but 35 feet apart, or 2-i5 and 210 feet, respectivelv, below the Twin coal. They are both, as may be seen by reference to the section, Pl.CLXXXI, intercalated in the mas- sive conglomerates which succeed the great white, egg conglomerate that underlies the south portion of the railroad bridge at the north end of the gap. The conglomerate last mentioned is, on account of its hardness, light color, thickness, and the regularity of its coarse quartz pebbles, one of the most easily recognized beds, lithologically, of the Pottsville formation over a large portion of the Southern Anthracite field. By glancing at the columns of the table showing the species furnished in the small collections, obtained with some difficulty from the coal}-^ or shaW partings between the conglomerates, we find that Alethojjteris Se)'Ui, A. coxto7iiana^ Neuropteris ovata^ iV\ Desoriif^ tiplienophyUum cuneifolium, and Slgillaria cf. Icevigata have been obtained from l^ed M, while Pseudoj)ecopte7'i8 cf. squamosa^ Pecopteris sp., Neuropteris ovata^ Aletliopteris Serli'i^ Cardiocarpon elongatum var. intermediwn^ C. annxdatum^ and, perhaps, C. hicuspidatwn var. ohioense are present in bed N. The last identification is uncertain, since the speci- mens, which were obtained from coarse, conglomeratic sandstones, are very indistinct and fragmentary. The Pecopteris species comprises a villous type close to the ferns described by Professor Lesquereux as Pecopteris vestita from the Lower Coal Measures of ^Missouri, and as P. Buchlandii Brongn., from the Pottsville formation at Campbell Ledge, near Pittston, Pennsylvania. With, perhaps, the exception of the Alethopteris coxtoniana^ the fiora of l)ed N is apparently as ancient as that of M. The com- bined list from these two beds, which, on account of their stratigraphic proximity and their similar plant contents, ma}" for the present be treated as one flora, contains l)ut 11 species, yet these are of a highly interesting and suggestive character. But one fern species,^ Alethop- teris coxtoniana., is represented in the collections from the Lykens groups. The remaining ferns are either identical with the species of the Lower Coal Measures of the same region, though varying somewhat in minor details, such as size, or are very closely bound to typical Coal Measures species. Sphe7iopkyllu7n cuneifolium is repre- 1 yeuropteris ovala is represented by a variety in the roof of one of the upper Lykens coals at New Lincoln. 20 OEOL, PT 2 51 802 FLORAL ZONES OF THE POTTSVILLE FORMATION. sented ])y the more rigid, eotir.se-nerved, irregularly dissected, broad- toothed form more characteristic of the Lower Coal Measures ; not by the very narrow, lax-leafed type, Avith thin nerves, described by Lesqiiereiix^ as Sjj/icuoj>/ii/I/w/i sa:i'(fi'a(j(i:'f(>liiuii^ from beds of Upper Lykens age in Washington County, Arkansas. The Sigillaria Icevi- gata is at once suggestive of the Coal ]\Ieasures. The g^-mnosperms, on the other hand. l)elong to species Avhich have generalh' a relativel}^ wide range in the higher part of the Upper LA^kens division in other coal fields, and which are hardh' known from the Lower Coal Measures of the Northern States. These comprise species that are especially common in the upper portion of the formation, of which Caniiocarpon annuldtain. A\u\ C. hicusj)idatum Yiiv. oliioense appear to bo distinctly characteristic, the former being more restricted to the upper beds. From the foregoing it appears that in beds M and N we have a flora the pteridophytic elements of which are, on the whole, generally dis- tinct from those characteristic of the preceding zones of the Pottsvillo formation. The gA'mnosperms, on the other hand, are character- istic of the Upper Lj^kens division. Yet the ferns, though identical or closely related to those of the Lower Coal ]Measures, appear not to exhibit the forms and facies of the species found either in the roof of the Buck Mountain coal (Lower Coal Measures) or in the Brookville or Clarion coals of the Allegheny series, in the bituminous basins of the Northern States. The wide difference between the floras of the preceding- zones of the Upper Lykens division of the Pottsville for- mation, on the one hand, and those of the Lower Coal Measures, on the other hand, has already been indicated, and will be further shown on a later page. Between these two sections — between the flora of Lj'kens coal No. 1. or of bed L of the type section, and the roof of the , Buck ^Mountain coal, or base of the Lower Coal Measures in the anthracite fields — we have an interval of about 375 feet, within which occurs a very distinct, though perhaps gradual, change from the purely Pottsville plant life to the flora which, as we shall presentl}" see, is distinctly that of the Lower Productive Coal Measures, as that group is recognized in the coal fields of the Northern States. The small collections obtained from the partings, beds M and N, in the upi)er plexus of massive conglomerates, which occurs within the top of the Pottsville formation as generally defined on a lithologic basis, appar- ently constitute fragments in evidence of this floral transition. For the present, when speaking of the type section and region, and until the subject is treated in a broader light, in connection with the Potts- ville of the other portions of the Appalachian province, I shall continue to use the term '""Upper Litermediate division" in referring to this portion of the Pottsville formation. 1 Coal Flora, Vol. Ill, p. 720, pi. xciii, fig 9, 9a. WHITE.] FLORAL ZONES OF THE FOTTSVILLE FORMATION. 803 CORREI^ATIOIS^S. It is not within the scope of this paper, whose primary purpose is to present a combined stratigraphic and paleontologic type section and definition of the Pottsyille flora in the type region, to enter in detail into the subject of the correlation of the yarious terranes and groups in the Appalachian trough which haye been or should he regarded as equiyalent to the whole or a part of the Pottsyille formation as deyel- oped in the Southern Anthracite field. Such a treatment of these extensiye and complicated problems can be satisfactorih' accomplished only in connection with the consideration of the detailed paleonto- logic evidence of all the terranes concerned in the comparisons. In this report questions of contemporaneity will be confined to beds at isolated localities in the anthracite region, or to formations or groups in other regions whose floras are already more or less known, and which will be correlated onl}" in a broad sense. These cases will be divided into two groups: (1) Detached localities which are situated within the Southern Anthracite field itself and whose actual occurrence in the Pottsyille formation in the typical region renders this correlation more important as well as certain, while at the same time adding to our knowledge of the distribution and range of the species in the Pottsyille Basin. (2) Terranes or groups whose floras have been studied in other fields. In discussing the beds of the first category greater confidence will be reposed in the occurrence, in a given bed, of the particular group- ing or association of species which, in the beds or sections alread}^ discussed, appear to be characteristic of the several horizons, although the number of species from the locality in question may be small. On the other hand, in considering the relative age of formations geograph- ically more remote, greater stress will be laid on the composition of the entire flora, and on the vertical range of its elements as well as the pro- portion of its identical species. PALEONTOLOGIC RELATIONS OF COALS DEVELOPED AT ISO- LATED MINES IN THE SOUTHERN ANTHRACITE FIELD. The principal detached localities, within the limits of the Southern Anthracite field, from which fossil plants have been obtained are those inscribed to the right of the columns devoted to the type section in the table of distribution. All of these have at some time been the scenes of coal exploitatioii or prospecting. In most cases the beds have been either tentatively or definitely, and, as Avill ])e further shown, some- times erroneously, correlated with reference to the Lykens coals mined in the Lincoln-LyktMis region. Several of these localities are but a few miles from the mining developments of the latter region, and nearly all are east of the mines. The correlation of these beds, so far as it can 80-4 FLORAL ZONES OF THE POTTSVILLE FORMATION. be made with precision or close approxinmtion. lias an important l)ear- ing on the geographic extent and economic condition of the several coals. AVith few exceptions the beds discussed are located on the mine maps, while generally they will be found approximately if not exactly identitied in the cross-section and columnar-section sheets of the Atlas of the Southern Anthracite Field. Proceeding along the upturned edge of the coal field in Sharp Mountain, west of Pottsville, we shall consider: 1. Drifts! hi tlu' Loicer Lyl'ens division in Swcdara Gap. Station 3, PI. CLXXX. Two of the Pottsville coals have been drifted at a little above water level in this gap. The geographic positions of the open- ings are shown in mine sheet xvi. Atlas Southern Anthracite Field, Pt. III. The structure of this portion of the basin is illustrated in section 23, cross-section sheet xvi. Atlas Southern Anthracite Field, Pt. VI. From the upper of the coals, which is about \AS) feet below the "Buck Mountain "" (Twin) coal, as identified by the State survey in this gap. no fossil plants were obtained. The roof shales from the lower coal, mined to a slight extent on both sides of the gap. have fur- nished species as follows: Xeuropteris Smithsii. Whittleseya Campbelli. Mariopteris pottsvillea. Neuropteris Pocahontas. Xeuropteris Pocahontas var. iniiequalis. These species, though few. are always common in the roof shales of Lykens coal No, 4, of which the first and fourth named are especially characteristic. The inference that this coal, which was mapped by the late State survey as Lykens coal No. 6, is more probably the Lykens coal No. 4, as indicated by the fossils, is further supported by the thickness of the rock (about 600 feet) between it and the Buck Mountain coal. That this coal is as old as Lykens coal No. 6 seems very improljable. 2. Bausch Gap^ Schuylkill County. Station -I, PI. CLXXX. At Rausch Gap, 1 mile west of Swatara Gap, two of the Lykens coals have ] )een driven into for some distance. The district is shownon mine sheet xvi, Pt. Ill of the Atlas of the Southern Anthracite Field. The struc- ture, consisting of a slightly overturned (70^ dip) south liml) of the deep Coal Measures basin, is similar to that at Swatara Gap. The section at this point, as compiled from the incomplete conglomerate exposures in the gap. is shown in PI. CLXXXV, Fig. 1. From the lower of the two coals mentioned, about 975 feet below the representative of the Buck ^Mountain coal, at the opening on the east side of the gap, there have been gathered the following: Mariopteris eremopteroicles ? Lepi'lostrobiis inMiiisylvanicus. Neuropteris Pocahontas var. inaMiualis. \ Trigonocarpuni anipulkeforme. CORKELATIONS IN THE SOUTHERN FIELD. 805 The other coal, about TO feet higher, opened on the west side, has furnished fragments representing — Neuropteris Pocahontas var. pentias ? Neuropteris Pocahontas var. insequahs. Calaniites Eoemeri. Trigonocarpum ampullseforme. Trigonocarpum Helente ? Neither of these florulas is sufficient!}^ complete to form the basis for definite correlation. Nevertheless, not onh- is it clear that both belong to the Lower Lykens division of the Pottsville formation, ])ut it is also highly probable, from the absence of species characteristic of Lykens coal No. 4, as well as from the presence of Mariopteris erern- opten'oides^ and, probabh% of Neuropteris Pocahontas var. pentias^ that we have here to do with Lykens coal No. 5, or a still lower coal. As already stated in the discussion of the floral characters of the horizon of Lykens coal No. 6, owing, perhaps, to the scantiness of material in the collections, no delinite paleontologic distinctions can yet be drawn between it and coal No. 5. Taking into account the agreement of the florulas with Lykens coals Nos. 5 and 6. as well as the interval between the beds, it seems probable that the coal opened on the west side, which was mapped by the State geologists as Lykens coal No. 6, and corre- lated by them with the lower bed in the Swatara Gap, is really Lykens coal No. o, in which case we ma^^ assume that the other, lower coal drifted on the east side of the gap represents the Lykens coal No. 6. The interval between the coals, about TO feet, as well as the general distances of the latter from the "Buck ^Mountain" coal, corresponds well with the stratigraphic relations of Lykens coals Nos. 5 and 6 at the Lincoln mine, about 3 miles to the northwest. Though few in number, the plants in the Rausch Gap, which are distinctly characteristic of the Lower L^'kens division, are especially interesting as compared with those from coals that have hitherto been supposed to be of the same age in Lorberry Gap, a mile to the west. The latter will later be especialh' treated in connection with the Dauphin Basin. 3. Coal shaft northeast of the yorth Bhxjkslde slope. At a distance of a little more than 200 3"ards northeast of the North Brookside slope on Lykens coal No. 2 (Station T, PI. CLXXX), a trial shaft was, several years since, sunk on a coal which has l)een supposed ])y the local engineers to be the Lykens coal No. 4, though the isolated position of the proving, on the north side of the Wiconisco Basin, opposite Good Spring,^ left some doubt as to the accurac}' of the correlation. The presence of ^[ariopteris pottsr'dlea., ^^phenopteris p>atentissiina^ Neurop- teHs Pocahontas var. huepialh, and Xeuropteris Sniitltsii in the flora from the roof of the coal points clearly to its contemporaneity with the Lvkens coal No. -i. 1 Atlas Southern Anthracite Field, Pt. Ill, mine sheet xvii. Trigonocarpum ampul Iseforme. Carpolithes oriz?eformi.>^. Whittleseya Campbelli. 806 FLORAL ZONES OF THE POTTSVILLE FORMATION. 4. Thr Eurrhi drifU. Station 33, PI. CLXXX. The two Eureka drifts or tunnels are, as shown in mine sheet xvi. Atlas Southern Anthracite Field, Pt. III. located on the slope of Broad Mountain, nearly 1^ miles northwest of Tremont. In both the mine map and the cross-section sheet, a portion of which is repeated, with a description, in the Summary Final Keport ' of the State geological survey, the upper Eureka tunnel is represented as starting from near the outcrop of the beds mined in the lower tunnel and traversing a thin relict of the Middle Creek anticline and a narrow basin beyond, so that, at a hori- zontal distance across the measures of about 375 feet from the coals mined by the lower drift or tunnel, the same coals were again reached and mined on nearh' the same south dip (30^-38"^). That this interpre- tation of the structure is almost certainly erroneous will at once be seen on referring to the fossils derived from the two long-abandoned mines. The plants from the lower tunnel comprise the following species: Mariopterii- pygma?a. C'ardiocarpoii Cuyahogse. Aletliopteris Lacoei. Callipteridium alleghaniense. Nenropteris acutimontana ? Calamites approximatus. This flora, as may be seen bv an examination of the chart, is typical of the zone of Lykens coals Nos. 2 and 3, with which the beds in this drift have been correlated bv the State geologists. The identitv of No. 2 is indicated especially stronglv bv the presence of the three species first enumerated. "When, however, we examine the roof shales brought from the other (upper) tunnel we find — Mariopteris pottsvillea. j Neuropteris Smithsii. Xenropterig Pocahonta?' var. ina}quali.«. Trigonocarpum Helenpe. This flora, though small, is characteristic of the Lower Lj'kens divi- sion, to which all but Trigonocarpum Ilelence exclusivel}^ pertain. Fur- thermore. Mwioptcrix potUviUiHi and Xetirojderis Smithsii are, in the Southern Anthracite field, so far as known, exclusively in or near the horizon of coal No. 4. in which th(> variet}^ i/icequalis of Xeiirojjte)'is Pomhnnta-s is at home, while the Trigonocarpum is most common at. and essentially typical of. the same level. I have, therefore, little hesita- tion in referring tin- horizon of the shales, which are undoubtedly of Lower Lykens age, to the horizon of Lykens coal No. 4. The shal«\>< from a prospect shaft a short distance to the east of the mouth of the upper drift have furnishcHl — Eremopteri.s lincolniana. EremopterLs decipiens. Alethopteris grandifolia var. obtusa. Calliptmdium alleghaniense. Xeun )j)teri8 acut(,)niontana. Neuropteris tennesseeana. 1 Vol. HI, Pt. I, p. 2120, i.l. 384. WHITE.] COREELATIONS IN THE SOUTHERN FIELD, 807 The species here associated are all, in g-eneral, typical of the zone of Lvkens coals Nos. 2 and 3. If the horizon of the shales is on either side of this zone it is perhaps slightly higher. It appears most prob- abje. however, that it is near the outcrop of the neighboring coals Nos. '2 and 3, in agreement with the mapping of the latter on mine sheet xvi of the Anthracite Atlas. The correlation of the coal mined in the upper Eureka tunnel with Lvkens coal No. -i necessitates a very different structural interpre- tation of the beds. It strongly suggests a strict and regular paral- lelism in the same monocline of the coals in both drifts, in which case the interval between coals 3 and -i, about 250 feet, would be entirely in harmony with the corresponding interval, 245 feet, in the Lincoln mine, about 3 miles to the southwest. The Middle Creek anticline seems to have been either erroneously interpreted on the State mine maps as extending too far westward, or, as is quite possible, wrongly platted to the south of the coals in the upper tunnel instead of to the north. Neither the mine map nor the protile appears to contain evi- dence of importance in contradiction to either alternative. 5. Valley View colliery ^Kolders Gap. Station 15, PI. CLXXX. At the gap in Bear Mountain, 2 miles north of Brookside, several of the Lvkens coals have been located, one of the upper Lvkens coals being now worked for local use at the Valley View colliery. The position of the developments and the stratigraphic relations of the beds in the north side of the AViconisco Basin in this region are shown in mine sheet xviii. Atlas Southern Anthracite Field, Pt. Ill, and in section 26, cross-section sheet xix. Atlas, Pt. VI. The columnar section was described by H. D. Rogers.^ The species from the heav3% «^iidy roof shales of the coal, which is mapped as Lvkens coal No. 2 in the State mine maps, include among others — Eremopteris decipiens. Cordaites grandifolius? Eremopteris Aldrichi. Cardiocarpon elongatum var. antholith- Sphenophyllum bit'urcatum. oides. Cordaites Phillipt^i. Cardiocarpon ol)liquuni. The flora is unlike the floras found in the roof shales of the lower Lj'kens coals. The distribution of its species is essentially in the Upper Lvkens division, and prevailingly in the zone of Lykens coals Nos. 2 and 3, though it appears to lack tlie species specially characteristic of either coal. 6. Kemhle drift. Station !♦>, PI. CLXXX. The Kemble drift is sit- uated near the western spoon of the Peaked Mountain Basin, on the flrst of the large, shallow, synclinal, westward-projecting lobes of Broad Mountain. Its position and the general geologic environment are shown in mine sheet xiii. Atlas Southern Anthracite Field, Pt. II, and 1 Geology of Pennsylvania, 1858, Vol. II, Pt. I, p. 190, PI. VIII; Lykens coals correlated, by A. DW. Smith, in Summary Final Report, Second Geological Survey of Pennsylvania, Vol. Ill, Pt. I, j). 2130. 808 FLORAL ZONES OF THE POTTSVILLE FORMATION. in section 23. cross-section sheet xviii. Atlas. Pt. \\. republished on a small scale, with a brief description, in the Sunniiary Final Report of the late State geological survey.' From the rot-k dump at the mine, which has for uany years been operated for country use. the following species were codected : Whittleseya Campbell!. Cardiocarpf)n di?ciilnni. Cardiocarpou ol)li(]Uuiii. Trigonocarpum aiiipulheforme. Trigonocarpuiii ainpulLcfDrnie var. s2-)ec- taVjile. Trigonocarpum Helena'. Trigonocarpum Dawgonianum. Carpolithes orizreformi?;. ^lariopteris pottsvillea. Sl)henopteris Lutheriana. Alethopteri:* grandifolia. Neuropteris Pocahontas var. in;equalis. Xeuropteriis gigantea var. c-lavata. Caiamites Roemeri. Asterophyllites cf. rigidus. Calamostachys cf. lanceolata. Si>iienophyllum tenue. Whittleseya Lescuriana. By far the greater portion of the above names are familiar in the di.scu.ssions of the floras of the Lower Lykens division. Nearly all the species occur in the shales of that group, while in Marioptens potiKviUea^ Sphenoptrrk Lutheriana, and ^phenophylhim temiev^Q seem to have species specially characteristic of Lykens coal No. -i, of which NeuTOpteris Pocahontas var. incequalis and Trigonocaipum Helenas also are largely typical. In fact, in view of the general agreement of the flora as a whole with that of Lykens coal No. 4. and of the presence of several of the species supposed to be typical of that horizon, we may consider the palcontologie evidence as pointing very strongly toward the assignment of the coal at the Kemble drift to an approxi- mate level. As tending, however, to impair the strength of the evi- dence of these fossils, mention should be made of certain minor diflfer- ences in the forms of the species. Thus the form of Marlopt&rk jMts- villea present at this mine is a rather lax type with somewhat dilated pinnules, while the form of Xeurop>terts Pocahontas var. iniequalis is ))oth elongated and robust. Furthermore, Alethopteris grandifolia, Whit- tleseya Lescuriana, and Cardiocarpon disculurn. appear in other portions of the basin to Ije confined to the zone of Lykens coals Nos. 2 and 3, while AsU^rophyUitt's cf. rigidus is more at home in the upper divi- sion of the Pottsville formation. In view of the presence of these elements of generally later age. we may. I l)elieve, safely conclude that the coal mined at the Kemble drift, which is mai)ped by the State geologists as Lykens No. 5,' can not be any older than Lykens coal No. 4. In the Summary Final Report,^ the correlation of this coal, w'hich is there described as "clo-sely overlain" by coal No. 4, 4 feet thick, is expres.sed as uncertain. It seems possible that the low^er coal is a new one occurring near Lykens coal No. 4, if it 1 Vol. in, I't. I, p. 2n9, pi. 3M. = AUas Southern Aiithnu'ite Field, Pt. H. miin- slu*et xiii. 3 Vol. HI. Pt. I. p. 2n9. wHiTE.j CORRELATIONS IN THE SOUTHERN" FIELD. 809 is not. in fact, identical with the latter as mined in the Wiconisco Basin. That the Lvkens coal No. 4 probably extends farther to the east on Broad Mountain will be suggested by the ilora obtained at the old Altamont collier}' No. 1. 7. Altamont colliery No. i, near Frackville. Station 36, PL CLXXX. This colliery is situated on the south margin of the New Boston-Gor- don Basin at the north Ijorder of Broad IVIountain, on the northern limit of the Southern Anthracite lields. The areal geology is shown on mine sheet vii. Atlas Southern Anthracite Field, Pt. II. The struc- ture is illustrated in section IT, cross-section sheet v. Atlas. Pt. V, the rock sequence being shown in columnar-section sheet ix. Part IV. As might be inferred from the columnar sections,^ compiled from diamond-drill bore holes along the basin, the coal was found to be very "" faulty," soon pinching too much for profitable mining. The plants collected at this mine are: Lepidodendron clypeatum. Lepidostrobua cf. ornatus. Lepidophylluin quinnimontanum. Cordaites grandifolius? Trigonocarpuin ampullajforme. Aneimites pottsvilleiisis. Mariopteris pottsvillea. Sphenopteris asplenioides. Sphenopteris patentissima. Alethopteris lonchitica? Neuropteris Pocahontas var. iiifeqiialis. , Carpolithes orizajformis. Neuropteris Smithsii (form) . ! This flora, even more clearly than that from the Kemble drift, shows a composition distinct!}' characteristic of the Lower L^'kens division, to which is added a small element of younger species. The two floras are of nearly the same general composition and significance, and repre- sent, I believe, approximately, if not identically, the same stage. There- fore, notwithstanding the reported presence of a 4-foot coal, supposed to be Lykens coal No. 4, closely overlying the Kemble drift coal (l)ut one coal appears in this portion of the diamond-drill borings in this vicinity on Broad Mountain), I am strongly disposed to regard the coal here as even slightly younger than, if not really contemporaneous with the Lykens coal No. 4. It is probable that the same coal, which appears, as will next be shown, also to extend along the southern limb of the Western Middle field to the west of Frackville, is opened at the Gordon incline slope, Moser's drif t,^ and other points on the western lobes of Broad Mountain. 8. Mount Plsgah, near Ifaiich Chunl. Station 40, PI. CLXXX. From a small drift recently driven on the north slope of Mount Pis- gah, not far from the head of the incline of the Switchback Railway, the following species were collected: Pecopteris serrulata. Neuropteris sp. indet. Lepidodendron clypeatum. Cardiocarpon cornutum ? Cardiocarpon bicuspidatum var. ohioense. 1 See also Summary Final Report. Vol. Ill, Pt. I, p. 2080, pi. 367 2 Atlas Southern Anthraeito Field, I't. II, mine sheet xiii. 810 FLORAL ZONES ()F THE POTTSVILLE FOKMATIOX. Of these species, Pecopterix ■'^errulata and the two Cardiocarpa are, in oeneval. charac-teristic of and ahiiost exc-hisivoly contined to the Upper Lykens division in the Southern Anthracite field and to the upper portion of the Pottsvillc formation in the })ituminous basins. Lt-phJndendi'iDi clypcKtuut, also, is not o-onerally foiuid in beds below the same division except in the region under consideration. Its typical phase is developed near th(> l)ase of the Coal Measures. .The remain- ing species of the florula. although rt^presented by material too frag- mentary for satisfactory identihcation, appears to be allied to one of the forms of Neiiropterk in the upper divisions of the Pottsville, rather than with the small, Callipteridioid, narrow-pinnuled types of the Lower Lykens division. Thus, from the composition and distri- bution of the flora, it seems probable, notwithstanding the small number of species, that the coal, which is here over 5 feet in thick- ness, is situated in the Upper Lykens division of the formation. The geolog}' of this portion of the field is shown on mine sheet i, Atlas Southern Anthracite Field, t*t. I. The structure of the east end of the region is illustrated in section 1, cross-section sheet i of the same atlas. The columnar section of the upper portion of the formation as measured in the Hacklebarney tunnel, al)out .3 miles dis- tant, shown on columnar-section sheet ii, will, perhaps, serve in a general wa}' to indicate the sequence of terranes at the top of Mount Pisgah. although the coal in question is not identified. If this coal is referable to the Upper Lykens division of the formation, as the fos- sils seems to indicate, we may conclude either that the Lower Lj^kens division at the eastern point of the field is much thinner than elsewhere or that the basin Avhich, as mapped by the State survey, extends nearly to the level of the Central Railroad of New Jersey' along the Lehigh River is deeper near its extremity than has generally l)een supposed. HORIZON OF THE LOWER LYKENS VALLEY COAL IN THE WEST- ERN MIDDLE ANTHRACITE FIELD BETWEEN FRACKVILLE AND SHAMOKIN. About I mile east of the Altamont No. 1 colliery the Pottsville for- mation ])ridges the axis which forms the line of separation ])etw(HMi the Southern Anthracite field and the "Western ]\Iiddlt> Anthracite field, and plunges into the steep Mahanoy Basin east of jNIahanoy Plane. A. For a long distance to the west of Frackville no coal of the Pottsville formation has been worked to any extent, but at the old Gordon (Franklin) mine, in the "Western Middle Anthracite Held, about 4 miles northwest of the old slope at the Gordon plane, which is within the north border of the Southern field, a coal designated on the mine sheets of the riekU and in the Summary Final Report of the State lAtlns Western Middle Anthracite Field, I't. H. mine sheet v; columnnr-scction sheet ii; I't. Ill, cross- section sheets v, vi, section 12. WHITE] CORRELATIONS IN THE WESTERN MIDDLE FIELD. 811 survey' the Lower Lykens Vulley coal was formerly worked to a con- siderable extent. From the rock dump at this mine were obtained the following species, which, though few, appear in the Southern Anthra- cite field to be either characteristic of or most common in the horizon of Lvkens coal No. 4: Aneiniiteg pottsvillensis. Neuropteris Pocahontas. Xeuropteris Smithgii. Lepidodendron clypeatum. Lepidostrobus pennsylvanicus. Trigonocarpum Helenfe. B. At the abandoned Helphenstein colliery,"^ 2 miles south of Locust Gap, there were collected a few species, as follows: Neuropteris Pocahontas. ; Sigillaria sp. cf. dentata. Neiiropteris Pocahontas var. msequaUs. 1 Trigonocarpum ampuUeeforme. Lepidostrobus pennsylvanicus. I C. About '2 miles farther west, along the southern border of the same field, a coal mapped as the same as that wrought at the Gordon and Helphenstein collieries was formerly worked rather extensively by Messrs. Doutv and Baumgartner.^ The species collected from the rock dump are — Eremopteris sp. Mariopteris pottsvillea. Sphenopteris asplenioides. Neuropteris Pocahontas var. injequaUs. Neuropteris Smithsii. Trigonocarpum ampullseforme. Trigonocarpum Helena?. Carpolithes orizseformis. D. Along the ravine extending up the mountain side above the site of the old Enterprise colliery, a coal supposed to be the same Lower Lykens Valley coal was formerh^ worked at the jMount Franklin and the Margie Franklin collieries,* between which lies a narrow anticline. From the mine last mentioned the specimens collected include — Eremopteris sp. Mariopteris pottsvillea. !Mariopteris cf. tennesseeana. Sphenopteris asplenioides. Sphenopteris microcarpa. Alethopteris prota(juilina. Neuropteris Pocahontas var. inaequalis. Neuropteris Smithsii. Lepidodenron alabamense. Whittleseya Campbelli. Trigonocarpum ampullaeforme. Carpolithes orizasformis. A review of the species collected from these mines along the south- ern border of the Western Middle Anthracite field shows that we have to do with the same flora as that present at Altamont collier}^ No. 1, within the Southern Anthracite field. In all the collections we find representatives of Mariopter is pottsvillea^ Xeiiropteris Pocaliontas var. liKpfjualls, Neuropteris Smithsii. and Lepidodendron cdabamense^ charac- teristic in the Southern field of the zone of Lykens coal No. 4. The 1 Vol. Ill, Pt. I, pp. 2058-2060. 2Atlas Western Middle Anthracite Field Pt. II, mine .sheet v. 3 Atlas Western Middle Anthracite Field Pt. II, mine .sheet vi. ■•Atlas Western Middle Anthracite Field, Pt. II, mine sheet vi; Pt. Ill, cross-section sheets v, vl, section 14. 812 FLORAL ZONES OF THE POTTS VILLE FORMATION. evidence offered b}' these species is further strengthened by the pres- ence in the combined flora of the rarer species A^ieimites pottsviUensis and AlethojyUi'is protiuiuUinn^ also typical of that level, as well as l)y Sphenopteris asplenioides and Trigonocarpum HelenKB^ usually found associated with the former in the larger collections. In short, the tes- timony of the (•oml)incd flora is so strongly indicative of the con- temporaneity of the Lower Lvkens Valley coal in this portion of the Western Middle field with Lvkens coal No. 4 of the Southern field as to leave little I'oubt of its approximate synchronism or correlation therewith. The similarity in the stratigraphic position of the coal worked in the Shamokin Gap and at several other points in the west- ern portion of the Western Middle field makes it seem probable that most of the small mines in the Pottsville formation in that region are developed in this coal. No fossils are at hand from the openings in this field on the "upper Lvkens Valley'' coal, which appears to be thin and unstable. If the correlations more or less definitely proposed above are accu- rate, the Lvkens eoal No. -i has a relatively wide distribution, not only in the Southern Anthracite field, Init also in the Western ]\Iiddle Anthracite field, and has the greatest extent in workable thickness of all the Lvkens coals, though its thickness is generally less than that attained by Lvkens coal No. 5 in the Lmcoln-Lykens mining district. ZONES OF THE POTTSVILLE FLORAS IN OTHER REGIONS OF THE APPALACHIAN PROVINCE. Li discussing the distribution of the floras of the several divisions of the Pottsville formation in other basins of the Appalachian province, I shall assume that the dispeTsion and migration of the species along the shore of the interior Carl)oniferous sea were, under the favoring con- ditions of a continuous, broad, Imse-level coastal- plain shore and cur- rcMits l)oth strong and varying, so uniform and so rai)id as compared with the geologic time required for the sedimentation of the terranes that the similar associations of ideiitical species occurring at difl'erent points along the coast are to be regarded as ai)])i'oximately contem- poraneous. Li other words, when regarding the sui-cc^ssion of terranes along the eastern l)order of the great Appalachian l)asin, in which we ha\e in difl'ei'ent districts th(> same regular succession of floras, we are justilied in considering that beds, along a contiiuious and unifoi'in coast, contaiidiigthe same flora are, geologically speaking, synch ront)us. rather than tiiat we ha\ e to do with homotaxy without contemporaneity. As aii-eady remarked, in correlating l>eds in regions more distant from the locality of the type section, great weight is attached to the com- position of the iloi-as and tiie \-ertical range of their elements, as Avell as the ))roportion of identical species In other liasins. Referring to the WHITE.] POTTSVILLE FLORAS IN OTHER REGIONS. 813 paleontologie features of the different coal horizons of the Pottsville formation, as outlined in an earlier part of this report (p. 773), it will be recalled that on the basis of the vertical distribution of the contained floras, so far as they have been brought to light, the formation was divided primarily into (1) a Lower Lykens division, including the roof shales of No. 4 in the mining region and bed D of the type section, and extending downward to the red shale; and (2) an Upper Lykens divi- sion, including Lykens coals Nos. 3 and 2 and the roof of Lykens coal No. 1, or beds H to L, inclusive, in the type section. In beds E and F (PI. CLXXXI) of the type section, comprising what 1 have designated the Lower Litermediate division, there appears to be some inter- mingling of the Lower and Upper Lykens species, while in beds about 225 feet below the conventional base of the Lower Coal ^Measures, or perhaps nearly 100 feet below the paleontologie base, we find another flora of somewhat mixed composition, suggesting the term "Upper Intermediate division.'' The two intermediate divisions are thin as compared with the whole formation. Of the two zones in the Lower Lykens division, the lower, including the horizon of Lykens coal No. 5, is characterized by a relatively simple flora. This, as indicated in the discussion of the floras of the several coals, consists principal!}' of Xeuropteris. Pocahontas^ which is always present and overwhelmingly abundant, its variety j?«?/?i/t/.y being peculiar to this zone. It is also marked b}- the absence of the forms characteristic of and conflned to the upper zone of the division, as the latter is represented in the roof shale of coal No. 1. The plants below coal No. 5, in the basal portion of the formation, are not suffi- ciently known for the discovery of any special zonal types. No attempt will therefore be made to determine in other regions any equivalents of this basal portion of the section, although certain inferences are unavoidable. The continued study of the Paleozoic floras along the eastern margin of the Appalachian trough fully confirms the conclusion I stated some years ago,' that it is only in the lower portions of the very thick sec- tions of the Pottsville formation in this province that the oldest floras are to be found, and that in general the very thin sections (e. g., along the northern and northwestern margins of the trough) correspond only to the upper portions of the sections of great thickness on the eastern and southeastern shores of the basin. The correlations suggested below will incidentally serve to illustrate this fact. However, since in this report, which is })ut preliminary to a monograph of the Pottsville flora of the Appalachian province, the correlative significance of the floras will be treated in ])riefest possi])le form, without enumeration of the characteristic species or the full presentation of the paleontologie, 'Bull. Geol. Soc. America, Vol. VI, 1895, pp. 319-320. 814 FLORAL ZONES OF THE POTTSVILLE FORMATION. descriptive, aiul illustrative data, (luestions of equivalence or con- temporaneity will ))e restricted to a few of the most important and best-known floras. Iii the later publication it is my intention to diseuss the character, range, and sequence of the floras .somewhat in detail, and to suggest such correlations of the numerous terranes of the formations, known often under different names in different regions and States, as are indicated by the very voluminous collcH'tions in hand, covering both the vertical range and the greater part of the areal extent of the formation in the Appalachian province. CLARK FORMATION. In passing from the Southern Anthracite field southward by way of the thinner developments of the Pottsville formation in the Broad Top and Potomac regions, we do not, so far as is at present known, meet with so low a phytiferous horizon as that of Lykens coal No. 5 until we approach the basin of the New Eiver in south-central West Wv- ginia, where within the rapidly deepening sections the Pocahontas and Clark formations appear in the basal portion of the Pottsville forma- tion. Paleontologically, one of the most interesting of the floral zones in this region is that represented by the plants in the roof shales of the Pocahontas coal (360 feet above the Mauch Chunk formation) in the Great Flat Top region of Virginia and West Virginia. In these shales, which comprise the basal portion of the Clark formation, we find a flora containing the greater number of the species found over Lykens coal No. 5 and presenting the precise f acies of the latter, including the invari- ably and almost exclusively abundant Neuro}}terh Pocahontas. The preponderance of identical species, the composition of the flora, and the relations of the latter to the succeeding floras render it certain that the horizon of the flora of Lykens coal No. 5 is, in the great Flat Top region, not far from the Pocahontas coal. The question whether its more precise horizon is above or below the latter coal will be discussed in the monograph, when all the evidence is presented. It may, how- ever, be here stated that it can not be far above the Pocahontas coal, nor is it likely to be over 200 feet below it. The zone of this flora, which has ])een identified through the Tazewell, Pocahontas, Oceana, and Raleigh (quadrangles, might ap})ropriately be designated the Xturop- teris Pocahontas zone, though in the Virginia region, as in the Southern Anthracite field, varieties of this species are found in higher terranes of the Pottsville formation. This zone includes the basal portions of the Welch formation in the Tazewell quadrangle;^ the Clark forma- tion in the Pocahontas^ and Oceana quadrangh^s; and a portion, above the middle, of a unit Avhich Mr. ]\I. R. Campbell in the manuscript folio relating to the Raleigh quadrangle has named the Thurmond formation. 1 Geologic Atlas of the United States, folio 44. 2 Op. cit., folio 20. WHITE.] PUTTS VILLE FLORAS IN OTHER REGIONS. 815 QUINNIMONT FORMATION. The upper portion of the Clark formation, which is about 375 feet thick in the Pocahontas quadrangle, is marked by the enrichment of both the Sphenopterid and Neuropterid groups in the flora. The variety imequalis of JVeuropter^'s Pocahontas survives, and in passing upward into the base of the Quinnimont formation, or toward the middle of the Welch formation, the partial contemporary of the Quin- nimont, we And it associated with the identical forms of Aneimites j^otts- villensis, JIa/'iojjfe/'is potts villea^ Sjjhenopterisjxctentissima^jVetiropteris Smithsii, 8phenopliyllim%tenue^ Lepidodendronalabametise, and Trigono- carpum Helenw^ so common in and characteristic of the horizon of the roof of Ljivens coal No. i in the Southern Anthracite field. In fact, the flora becomes practically identical with that in the anthracite region. To this zone, for which I have already suggested the term Jfariopteris pottsvillea^ on account of the common occurrence and very easy recog- nition of the latter therein, belong the fossils from the Dade coal in the Ringgold, Stevenson, and Chattanooga quadrangles^ and the lower coal mined at Dayton in the Pikeville quadrangle," all in the Tennessee- Alabama region. In fact, to the Mariopteris pottsinllea zone, giving the latter a broad interpretation so as to include a series of closely connected modifications of the tj'pes, belongs the entire suc- ceeding Quinnimont formation, 300 feet thick in the type region repre- sented in the Pocahontas quadrangle, and present in the Raleigh and Kanawha Falls ^ quadrangles; the upper portion of the Welch formation in the Tazewell quadrangle; a part of the Lookout sandstone, including the vicinity of the Dade coal, in the Chattanooga, Stevenson, Ringgold, Pikeville, and Kingston* quadrangles in the southern Appalachian region; and probabl}' a portion at least of the Lee formation in the Estillville, Briceville, Wartburg, and, perhaps, also in the London quadrangles'' m the northern Tennessee-Kentucky region. The flora of the Hindustan whetstone beds of Orange County, Lidiana, is also referable to this zone, and indicates the contemporaneity of those beds with at least some portion of the Quinnimont formation. The extent of the zones and the more definite relations and equiva- lents of the formations in the several quadrangles will be discussed in the later, monographic, treatment of the floras. 1 Geologic Atlas of the United States, folios 2, 19, and 0, respectively. 2 Op. eit , folio 21. 3 Op. cit. These folios have not yet been published or numbered. 4 Op. cit., folio 4. 6 Op. cit., folios 12, 33, 40, and 47. 81«5 FLORAL ZONES OF THE POTTSVILLE FORMATION. SEWELL FORMATION. The next higher general flora of marked characters and distinction in either the bituminous or the anthracite regions is that which I have indicated as characteristic of the proximate horizons of coals Xos. 3 and 2 in the Southern Anthracite held. The zone of this flora is characterized b}' the development of both the Rhacopteroid and the broad-lobed types of Eremopteris ; by small, round, and inflated- pinnuled species of ]\lariopteris ; by triangular Alethopterids : by small, palmate-lobed, and Pecopteroid Sphenopterids : by narrow Alethop- teroid forms of Xeuropteris, such as JT. SchUliani ; by MegalopUrh species and the ]Megalopteroid types ; by the broad-leaved Whittlesti/a', and by the introduction of the dentate Pecopterids, as well as a great diversity- of gymnospermous fruits. The more explicitly distinctive species of the zone of Lvkens coals Nos. 3 and 2 — Eremopteris Cheat- Juiiiil and E. decip)'ie)u^ Mainopteris py(jmi^a of the M. inflata group and J/, tennesseeana^ Sphenopteris jnlosa and S. palmatiloha^ Alethop- teris Evcmsil^ the Callipteridioid t3'pes, Neiiropteris acutimontana and J\\ tennesseeana^ SpliemplnjUuni cunelfolium {saxifragcefolium form), the Whittleseyas, and many of the fruits — are present in identical forms and associations in the shales over the Sewanee and Sewell coals. In fact, the elements of the flora from Lvkens coal No. 3 are so preponderantly identical with those in the roof of the Sewanee coal in Tennessee and the Sewell coal in southern West Virginia that these coals can only be regarded as practically contemporaneous.^ The paleontologic evidence for the identitication of the horizon of the Sewell-Sewanee coals presents the most complete and convincing as well as the most interesting case that has yet come within my observation. Marlopti^'is piigiiHi'H and the identical forms of Alethopteris Lacoei and Axt&ropliylllUs arkansanus^ which are especiallv t3'pical of the roof of Lvkens coal No. 2 in the Southern Anthracite field, have general!}' a somewhat higher occurrence and range in the bituminous fields. The zone of the plants of Lvkens coals Nos. 2 and 3, which, in recog- nition of the long-known flora of the Sewanee coal, at Sewanee and Tracy City, in Tennessee, I have in a previous paper"" called the Sewanee flora, may be termed the Sewanee zone. The distinctly Sewanee flora is present above the Sewanee coal in the lower part of the Walden for- mation in the Sewanee,^ Kingston, Pikeville, and Chattanooga quad- rangles of the Tennessee-Alabama region, and over the Sewell coal in the Raleigh, Kanawha Falls, and Hinton quadrangles in West Virginia. 1 The contemporaneity of the Sewell coal and the Sewanee coal, as well as the similarity of their stratigraphic relations in the Tennessee and Virginia sections, was pointed out in the description of the Pottsville section along New River, West Virginia : Bull. Geol. Soc. America, Vol. VI, p. 316. - Loc. cit. ^Geologic Atlas of the Unite>l Stales, folio s. WHITE.] POTTSVILLE FLORAS IN OTHER REGIONS. 817 \'erv nearl}' contemporaneous with the same flora is that in the roof of the Sharon coal in northwestern Penns^dvauia and northern Ohio, par- tially described by Dr. Newberry in 1872. ^ The * ' coal-bearing- shales " of Washington Count}', Arkansas, whose species, ranging usually a little higher than those of the Sewanee coal, were described by Pro fessor Lesquereux in his great work on the Coal Flora," can not be of very much later date than the Sewanee coal, and are undoulitedly representative of the Sew^^U formation. The lower portions of (a) the Sewell formation in the Pocahontas, Raleigh, and Kanawha Falls quadrangles, of (/y) the Dismal formation in the Tazewell quadrangle, and of (c) the Norton formation in the Estillville and Bristol quadrangles, are included in the zone of the Sewanee flora, to which in its broader sense are also referable certain plant-bearing beds of the Briceville formation in the Briceville and AVartburg quadrangles,^ in Tennessee. It also appears probable that large parts, perhaps the greater portions, of the Pickens and Black- water formations in the Buckhannon and Piedmont quadrangles,* respective!}', in West Virginia, are referable to the Sewell formation and are included within the Sewanee zone. However, the question of the existence of the lower horizons of the Pottsville in the relatively thinner sections in the Potomac region will receive particular atten- tion in the later and more complete report. LOOKOUT FORMATION. In those earlier published folios of the Geologic Atlas of the United States that relate to the Carlson if erous formations of the southern Appalachian region the coal-bearing terranes included in the quadran- gles are grouped in but two formations, the Lookout (lower) and the Walden (upper). The Lookout extends from the Bangor limestone (Mississippian) to the top of the great Sewanee conglomerate of Saffcrd.^ The oldest plants I have yet seen from this formation on the east side of the coal field, where it is thickest, are closely related to those from the roof of the Dade coal in northwestern Georgia, and are clearly referahle to the Mar iopteris pottsvillea zone. Whether the basal ter- ranes of the formation in this region are as old as or older than the Pocahontas coal or the Lykens coal No. 5 is still uncertain. It is, how- ever, highly probable that in the region included in the Kingston, Pikeville, Chattanooga, and Ringgold quadrangles the lowest beds of the Lookout are not older than the zone of the Lykens coal No. 5. 1 Kept. Geol. Survey Ohio, 1873, Vol. I, Pt. 11. 2 Coal Flora, Second Geol. Survey of Pennsylvania, Report P, :5 vols, and atlas, Ilarrisburg, 1S79-1H84. •■! Geologic Atla.s of the United States, folios 33 and 40. ■• Idem, folios 31 and 28. •'■Geology of Tonne.s.see, 1869. p. 3(i0. 2U GEOL, I'T 2 52 818 FLOKAL ZONES OF THE POTTSVILLE FORMATION. The study of the higher floras of the formation. iiR-ludino- tliat of the Dade coal above referred to in these quadrangles, shows the higher .shales of the Lookout, up to, or nearly to, the base of the Sewanee conglomerate, to ])e referable to the Mar'iojjterU jmtUvlUea zone in its broad sense, and to the Quinnimont formation in the Poca- hontas quadrangle. The fossils at the base of the Sewanee conglom erate in Tennessee, and of the Kaleigh sandstone, about 100 feet in thickness, in southern West Virginia, show a slight mingling of Sewa nee zone species, the flora being comparable to that of beds E and F of the Lower Intei-mediate division of the Pottsville (lap section. The Sewanee-Sewell or Upper Lvkens flora appears immediately above the great conglomerates which comprise the Raleigh sandstone in the Pocahontas, Raleigh, and Kanawha Falls quadrangles in West Vir- ginia, and which complete the Lookout sandstone in the Chattanooga, Pikeville. McMinnville.' Kingston, and Sewanee quadrangles in Ten- nessee and Ala])ama, The Lookout sandstone of Hayes seems, there- fore, to essentially represent both the Lower Lykens division and the Lower Intermediate division of the type section in Pennsylvania, although I am slightly disposed to doubt the presence in the Lookout of beds as old as the lowest at Pottsville. The Sewanee conglonierate at the top of the Lookout formation of Tennessee and its contemporary, the Raleigh sandstone in AVest Virginia, appear to stand in the same relative position paleontologically to the Lower Lykens division as does the Lower Intermediate division in the Southern Anthracite field, and each similarly seems to till the time break between the ^[ariojitLfh pottsvillea zone in its l)road (Quinnimont) sense and the Sewanee (Sewell-Lykens coal No. 3) zone. FAYETTE SANDSTONE. The characteristic species so well marked in the lower portion of the Sewanee zone become modified in the later portion, while at the same time new forms are introduced, so that in the southern Appala- chian region no sharp line will perhaps be drawn between the more resti'icted zone of the flora and that of the Lykens coal No. 1 or })ed L of the section at Pottsville, into which the species in the upper portion of the Sewanee formation, about 400 feet thick in the Kanawha Falls quadrangle, gradually merge. In this quadrangle the horizon of coal No. 1 is certainly close beneath, if not actually within, the Fayette formation, a group of massive sandstones and shales which succeeds the Sewell formation in the vicinity of New River, and completes the Pottsville, as the latter naturally would be and has been defined on the lithologic basis. A flora possibly con- temporaneous with that of Lykens coal No. 1 or bed L seems to be 1 Geologic Atlas of the United States, folio '22. WHITE.] POTTSYILLE FLORAS IN OTHER REGIONS. 819 present near the Gladeville sandstones in the Estillville quadrangle,^ Virginia-Tennessee region; also in the Breathitt formation in the London quadrangle, Kentucky; and in the Fayette sandstone itself near Zela, in the Nicholas quadrangle in ^^'est Virginia. The Mercer coals of northwestern Pennsylvania are possibly near the same horizon. CAMPBELL LEDGE. NORTHERN ANTHRACITE FIELD. The dark plant-bearing shales which lie within a few feet~ of the supposed representative of the Mauch Chunk (No. XI) in the very thin section (56 feet, more or less) of the Pottsville formation at Campbell Ledge, near Pittston, Pennsylvania, contain a large flora,'' which can not be older than Lykens coal No. 1 or bed L, and which has so much in common with l>eds M and N (Upper Intermediate division) of the type section at Pottsville as to strongly argue for a reference to the same time interval. The question of the equivalence in the southern and central Appa- lachian regions of the Upper Intermediate division, including the upper- most 3(»0 feet of the type section, involves a both complicated and diflicult problem, which, on account of the great expansion of the ter- ranes representing this period as we pass from Sutton in central West Virginia southward, cati not be appropriately discussed without a care- ful presentation of the accompanying paleontologic evidence. Omitting all the details relating to this question, which will receive special consideration in a later paper,* it may suffice to say that in the southern Virginia region the time interval, which is represented by 250 to 300 feet of beds, consisting chiefly of the upper plexus of conglomerates, between Lykens coal No. 1 or bed L and the Twin coal in the Southern Anthracite field, appears, as indicated by the fossil plants, to measure over 800 feet of strata on the Kanawha River. It includes the lower half or ''group" of the Kanawha formation. Within the lower por- tion of this interval, which is less argillaceous, there occurs a more gradual transition from the flora at the top of the Sewanee zone, or that of Lykens coal No. 1, to the typical flora of the Lower Kanawha group. The flora first recognized in the Brookville and Clarion hori zons of the Northern Bituminous basins, or coal A at Tamaqua and the Twin seam in the Southern Anthracite field, does not appear in the Kanawha region until later Kanawha time. In the lower portion of the Kanawha formation the flora agrees exactly in characters with that of the Lower Coal Measures of the British coal fields, or the greater portion of the Westphalian in continental Europe.^ In this 1 Geologic Atlas of the United States, folio 12. 2 1. C. White, Rept. Geol. Survey of Pennsylvania. <;?, ji. 143. »Lesquereux, Coal Flora, Vol. Ill, pp. 8.5.5-8.56. r portion of our Lower Coal Measures (Allegheny series), as, for example, at Mazon Creek. Illinois, or Henry County. Missouri, is gvnerally recog- nized as referal:>le to the Middle Coal Measures, or as near the base of the Upper Coal Measures of Europe. RELATIVE HORIZONS OF THE RASAL BEDS OF THE THIN SEC- TIONS AS COMPARED TO THE THICK EASTERN SECTIONS. So far as the examination of the phytiferous beds in the Pottsville formation has (extended, it appears that only the lowest beds of the formation in the Virginia region, where the formation attains a thick- ness of perhaps 2,500 feet, contain plants of greater antiquity than those of the lower beds in the Lower Lykens division of the type sec- tion. It is even possible that the oldest plants in the basal portion of the formation of Virginia are not of earlier date than those in the top- most red shale at the Pottsville and "Westwood gaps. Further study is needed to determine this question. The oldest plants that have yet been found in the Lookout sandstone seem to be of later age than the Pocahontas coal, and to be referable to the Mariopteris jwUsvillea zone. Beds possibly referal)le to the same zone appear to lie close to the red shale in the Pickens formation, in the Buckhannon quad- rangle, in central West Virginia; and the lowest beds of the Potts- ville, which is 270 feet in thickness,' at Hanging Rock, near Ironton. on the Ohio Kiver, at the western margin of the Appalachian trough, are perhaps referable to the uppermost part of the same zone. North of these points the sections of the Pottsville formation seldom exceed 400 feet, and throughout the bituminous sections of Pennsyl- vania they are in most cases less than 300 feet in thickness, while in some cases they are less than 150 feet. In none of these sections have I yet found plants older than the upper beds of the Quinnimont formation. No phytiferous ])eds so low as the Lykens coal No. 4, in the M(irhj2Jteru 2)(>itile, the boundary is uncertain. In general, and especially in the central portion of the field, including the vicinity of the type section, it is possible to trace the Twin coal, adopted by ' Atlas Southern Anthracite Field, Pt. IVB, columnar-section sheet x; also Pt. II, columnar-sfction sheet vi; and Pt. I, coluninai-sectior sheet i. 2Ge()l. Pennsylvania, Vol. II, Pt. I, laW, p. 17. 3 Ann. Rept. Geol. Survey Pennsylvania, 1886, Pt. Ill, p. 9:«; Summary Final Report, Vol. Ill, Pt. I, pp. 18.54, 1920. 82-1: FLORAL ZONES OF THE POTTSVILLE FORMATION. Roo-ors. for a long distance. It is probable that the errors, which un(lou])tedl3' exist, in the identitication of this horizon in the more distant portions of the field fall within a vertical distance of loo feet, except in tlie Dauphin Basin, which will be specially considered in a lat(>r section. As to the correlation of the A coal at Locust Gap, nortii of Taniacjua. there is still a difference of opinion; and again, in the western i)ortion of the Wiconisco Basin, the place of the Buck ISIountain coal is a matter of speculation. The })rol)al)le approximate position of the TavIm coal, wiiich has ))een agr(H>d upon by all the State geologists who have worked in this region as the equivalent of the Buck Mountain coal, along Sharp Mountain throughout the central portion of the field, to which for the present we shall confine ourselves, is shown in mine sheets xiv, xi\(/. XV. and xvi, in the third part of the Atlas of the Southern Anthracite Field. FLORA IN THE ROOF SHALES OF THE TWIN COAL. Since the fioor of the Buck ^Mountain coal, or its supposed equiva- lent, the Twin coal, forms the ar])itrary line between the Pottsville formation and the succtHnling Coal Measures, the flora in the roof shales of this coal maj^ be regarded as representing the basal horizon of the Lower Coal Measures.^ It therefore marks the upper limit of the Pottsvilh^, as that formation has, so far as I can learn, been defined, with reference to the type section, ])v the various geologists of the State. The localities whose plants are listed below are all mines or drifts, chosen from the central regions of the field, l)etween the Potts- ville^ Cap on the east and Tremont on the west. Over these the Twin coal has l)een probably correctly I'ecognized. it being of good body and (juality over nuich of this territory. (A) Pottsville Gap: Station 1. PI. CLXXX.' P'rom the Twin coal, whose position in the type section is illustrated in PI. CLXXXI. were ol)taiiied the following species:* IVcopteris ck'iitata Brongn. Pec'uptcriH arguta Sternh). Pecopteris viiiita Brongn. Pecopteris CandoUiana Brongn. PfcopterisorcopU'ridia (Schloth.) Sternb. Pfciipti'ris ^liltniii Artis. Neuroptoris Seheuclizeri Hoffni. Annularia s^teUata (Sohloth.) "Wood. Splienophylluni eniarginatnin Brongn. Lepidodendron np. indct. Bigillaria tessellata (iSteinh.) Urongn. (T.) Swatara Gap: Station 8, PI. CLXXX. Li this gap the Twin coal was formerh^ mined on l)oth sides, the more extensive operation 'Rogers, Geo). Pennsylvania, Vol. H, Pt. I, p. 140. -Atlu.'f Soull^eni Anthnicitu Field, Pt. II, mine sheet.s xiv, xiva. •'It should be distinctly understood that the identifications on which arc ba.sed the following lists of species from the Productive Coal Mea.sures in the Southern Anthracite field are merely temporary and provisional. As such they are subject to revisicm. Most of the names here given may be inter- preted as designating the sjime forms to which they were npi>lie!eiKlopecopteri8 squamosa (Lx.). Xeuropteris Sehenchzeri Hoffm. Annularia stellata (Schloth.) Wood. Splieiiophylluiu cuneifoliuni (Sternb.) Sphenophyllum emarginatum Brongn. Lepidodendron vestitnm Lx.? Lepidocystis (Sigillariostrobus?) quad- rangularis Lx.? The combined flora from these typical localities of the Twin or Buck Mountain coal is o'iven in the followins' list: Mariopteris sphfuopteroides (Lx.) Zeill. Mariopteris muricata (Schloth.) Zeill. var. nervosa (Brongn.) Kidst. Mariopteris cf. Sillimanni (Brongn.). Pseudopecopteris squamosa (Lx.). Pseudopecopteris obtusiloba (Brongn.) Lx. Sphenopteris pseudomurrayana Lx.? Sphenopteris nummularia Gutb.? Si)henopteris (n. sp.?) . Sphenopteris mixta Pchimp. Sphenopteris suspecta D. W.? Oliogocarpia cf. Brongniarti Stur. Pecopteris dentata Brongn. Pecopteris arguta Sternb. Pecopteris unita Brongn. Pecopteris villosa Brongn.? Pecopteris oreopteridia (Schloth.) Sternb. Pecfjpteris penmeformis [Lx.]. Alethopteris aquilina (Schloth.) Goepp. Alethopteris Serlii (Brongn.) Goepp. Callii)teridium Grandini (Brongn.) Lx. Xeuropteris rarinervis Bunb. Neuropteris ovata Hoffm. Neuropteris plicata [Lx.]. Neuropteris capitata Lx. Neuropteris vermicularis I.,x. Neuropteris fiml)riata Lx. Neuropteris Scheuchzeri Hoffm. 0teris ohtusiloha^ and Pseudo^yecojyterwci. squamosa are found onl}' in new varieties or doubtful forms in the Pottsville, while, as has previously been remarked, Sphenoj)hyUum cuneifolium is represented bv the broader, more rigid-leaved, and thick-nerved phase character- istic of the Lower Coal Measures. In short, a comparison of the lists shows that less than one-tenth of the ferns are represented b}" identical forms, both in the basal horizon of the Lower Coal Measures and in the upper phytiferous beds of the Pottsville formation. The species common to both formations are largely present in the upper 300 feet of the type section, i. e., the Upper Intermediate division. It must be evident even to one who is not familiar with Paleozoic fossil plants that the fern flora at the base of the Lower Coal Measures in the anthra- cite and other northern coal basins is almost totally different from that of the Upper Lykens division of the Pottsville. That the paleontologic characterization of the Buck Mountain coal is similar in the adjoining territory of the Western Middle Anthracite fleld is indicated b}' the following combined list of the floras obtained at the Mahanoy Plane (M); at the Vulcan colliery, 1 mile west of Buck Mountain station (V); at the Buck Mountain mine (B), and from the Big Mine Run colliery, north of Ashland (A), in the latter field. Neuropteris ovata Hoffni. (M B A). Neuropteris plicata [Lx.] (B A). Neuropteris vermicularis Lx. (A) . Neuropteris fimbriata Lx. (B) . Pseudopecopteris squamosa (Lx.) (V). Pseudopecopteris obtusiloba (Brongn.) Lx.(B). Pecopteris pennreformis [Lx.] (M B) . Pecopteris oreopteridia (Schloth.) Sternb. (VB). Pecopteris villosa Brongn.? (A) . Alethopteris Serlii (Brongn.) Goepp. (AB). Callipteridium cf. Grandini (Brongn.) Lx. (B.). Neuropteris Scheuchzeri Hoffm. (A M) . Annularia stellata (Schloth.) Wood (B). Sphenophyllum emarginatum Brongn. (VB). Lepidodendron aculeatum Sternb. (B) . Lepidophvllum lanceolatum L. & H. ? (Bk (F) Toward the eastern end of the Southern Anthracite field, in the Sharp Mountain Gap, south of Tamaqua:^ Station 39, PL CLXXX. lAtlas Southern Anthracite Field, Pt. I, mine sheet Hi ; Pt. II, mine sheet iv ; Pt. I, cross-section sheet iii, section 12; Pt. I, columnar-section sheet ii, columnar section 49; Summary Final Report, Vol. in, Pt. I, p. 2095. 82S FLOKAL ZONES OF THE POTTSVILLE FORMATION". Here the Twin coul has l)oon apparently (lefiiiitely identified at about 830 feet above the red shale, and is now ])eing mined. From the roof shales over this coal was obtained a flora whose character will at once be recog- nized as similar to that contained in the Pottsville-Tremont district: ERMnopteris! ci. arteiuisiajfolia (Stern b.) : Callipteridium Grandini (Brongn.) Lx. Srhimp. Pt^cnidopefopteris Hquaniosa (Lx.). Sphenoj)teris nuininularia Gutb. ? Sphenopteris (n. sp.?). Pecopteris dentata Brongn. Pecopterisoreopteridia (Schloth.) Sternb.? Alethopteris Serlii (Brongn.) Goepp. Neuropteris ovata Hoffm. Calamites Suckowii Brongn. Calamitcs Cistii Brongn. Annularia stellata (Hchloth.) Wood. Sphenophyllum emarginatnni Brongn. Lepidophyllum affine Lx. ? Lepidocystis quadrangularis Lx.' As has already been remarked, some doubt remains as to the indentity of the Twin or Buck Mountain coal in the Locust Gap, in the opposite side of the basin, al)out a mile and a half north of the last locality. At this point occur two coals, A and B, one or the other of which is sup- posed to represent the Buck Mountain coal. Coal A (16 feet in thick- ness), as shown in PI. CLXXXV, Fig. 2, lies about 750 feet above the red shale. Coal B is reported as 202 feet above A on the west side of the gap, and 260 feet on the east side.^ From the roof shales of coal A a small flora was obtained, as follows: Mariopterismuricata(Schlotli.) Zeill. var. nervosa (Brongn.) Kid.st. Sphenopteris pseudomnrrayana Lx.? Neuropteris capitata Lx. Neuropteri.s ovata Hoffni. Pecopteris dentata Brongn. Alethopteris aquiUna (Schloth. Goepp. It is obvious that this, like the floras previously enumerated, is a distinct!}' Coal Measures flora; and I have no hesitation in concluding that the A coal, which has generally been regarded b}^ the geolo- gists who have worked in this region as l)elonging within the Potts- ville formation, is referable rather to the Lower Coal Measures. It is important to mention in this connection that a flora, probably of no earlier age, is also present in the roof of a thin coal about 72 feet below the Twin coal in the type .section at Pottsville. The shales at this horizon in the latter locality have not ])een systematically searched. They are crowded with great numbers of Neuropteris Scheuchzeri^ among which are ]}ve>ie\\i Neuro2)teris rarl rurius^'AWiX small fragments of AxtrropliijUitrx cf. eiiamdifonnlxi^ thus indicating a Lower Coal IVIeas- ures age for this ]»ed as well. It has been suggested by Mr. Smith" that the A coal, which, at the Nesquehoning railwav tunnel, al)out ;> miles ea.st of Locust (lap, has a thickness of ])ut 1 foot, and which, 1 Probably a Sigillariostrobus. 2Rept. Geol. Survey I'CMinsylvania, AA, p. IWi; Summary Fitial Report, Vol. HI, Pt. I, \>. 20'J5. The interval is also j?iven as but 115 feet by Ashburiior in Report A, p. 80. sSummary Final Report, Vol. HI, Pt. I, p. 209G. WHITE] UPPER LIMIT OF THE POTTS VILLE FORMATION. 829 in one of the collieries a short distance west of the latter gap, was not discovered in a tunnel extending 400 feet below coal B, is in reality- only a bottom split of the Buck Mountain coal. THE PALEONTOLOGIC UPPER LIMIT OE THE POTTSVILLE. The important fact embodied in the preceding lists of plants from the roof shales of the Twin coal, in both the Southern and the West- ern Middle Anthracite fields, is that they represent a typical and dis- tinctive Coal Measures flora. The small element that this flora has in conunon with that of the Potts ville formation comprises species whose precursors, for the most part difi'ering in their forms and phases from the normal types, have made their appearance only toward the close of Pottsville time. Compared, as a whole, with the flora of Lykens coal No. 1. about 300 feet below the Twin bed in the Lincoln region, or bed L, 380 feet below the Twin in the type section, the species of the latter coal are so different as even to suggest the existence of a time break between the intervening beds. As an argument against such a supposition, 1 have, however, only to cite the plants from Ijeds N and M, whose floras, not less than 210 feet below the Twin coal at Pottsville, clearly presage the development of the Lower Coal Meas- ures plant life by the introduction of a number of Coal Measures types of ferns, notwithstanding the generally stronger paleontologic bond which attaches these beds to the Upper Lykens division of the Potts- ville formation. The transition already indicated in beds N and M seems to have been entirely completed within the time represented by the succeeding 200 feet of the section. It is necessar}" in this connection to note the relations of the flora of the Twin coal to those of the Lower Coal Measures in other regions. If we compare that flora with those accompanying the lower coals of the Allegheny series ^ in the Northern States, we find that its composi- tion, range, and development point definitely to a level as high as that of the well-known plant beds at Mazon Creek, Illinois; and that the horizon of the Twin coal should be nearly as high as that of the plants described from Henry County, Missouri,^ and Cannelton, Penn- sylvania. Compared with the better -known floras of the anthracite fields, the plants from the coal in question appear to indicate a level certainly not lower than that of coal C in the -Northern Anthracite field, while it is perhaps safe to say that they are nearly as young as coal D (the Marcy coal) in the vicinity of Pittston. According to the evidence of the plants, the beds of the basal por- tion of the Alleghen}^ series, between the top of the Homewood sand- stone and the Morris coal at Mazon Creek — an interval, probably including the Brookville coal, between the same sandstone and a level probably as high as the Clarion coal in the bituminous basins of western 1 Bull. Geol. Soc. America, Vol. XI, p. 149. SMon. U. S. Geol. Survey, Vol. XXXVII. 830 FLORAL ZONES OF THE POTTSVILLE FORMATION. Pennsylvania — and tho terranes extending from coal A to at least as far as coal C in the Northern Anthracite iield, were laid down prior to the deposition of the roof shales of the Twin coal. The examination of the plants from the above-mentioned basal beds of the Allegheny series shows that they constitnte a typical Coal Meas- ures flora, with but a small proportion of Potts ville forms, though the}' lack so high a development of the Pecopteroid group, as well as several other more advanced types which appear in the Kittanning and Buck Mountain coals. The same is true of the flora of coal C in the North- ern Anthracite field. In fact, like the flora of coal A at Tamaqua, or like that indicated in the coal 72 feet below the Twin bed in the Potts- ville Gap, and that of the lower horizons of the measures in the North- ern Anthracite field, the floras of the basal beds of the Allegheny series, which are above the lithologic Pottsville in the bituminous basins, are clearly referable, on paleontologic grounds, to the Lower Coal Meas- ures. In short, the comparative paleontology of the terranes shows {<() that, as related to the Coal ]Measures of other regions of the Appa- lachian province, or other basins of the world, the flora of the roof of the Twin coal, which has been made the dividing line between the Pottsville formation and the Lower Coal Measures in the Southern Anthracite field, is of a pure and well-advanced Coal Measures type; and (/>•) that its horizon is distincth' above beds, generally of no great thickness in the northern basins, containing floras characteristic of not so high a level, but nevertheless having a composition which is distinct from that of the floras of the Pottsville formation and which is too thoroughly identical with the plant life of the Lower Coal Measures to permit of any other reference. From the foregoing it appears: First, that the conventional upper limit of the Pottsville formation, in the Southern Anthracite field, is higher than the paleontologic upper limit. Second, that its horizon is also considerablv above that of the same boundary as drawn., not only in the ])ituminous basins, but also in the Northera Anthracite fields. Third, that the paleontologic limit appears, so far as evidence has been ol)tained, to lie below, though perhaps very near to, the coal at 72 feet below the Twin coal in the type section, and probably above the plant beds 210 feet below the Twin coal. In other words, it also appears that in the type section the paleontologic upper limit of the Pottsville formation lies close within the upper outskirts of the great plexus of conglomerates in which the formation culminates. Thus, the paleontologic limit falls within and near the natural or lithologic limit. It hardly need be repeated that the A coal at Tamaqua should, according to the evidence of the fossil plants, ])c included in the Lo win- Coal Measures, as should, also, the thin bed next below the Twin coal in the type section. It is not impossible that one or the other of these lower coals is the representative of the Scott Steel coal occurring WHITE.] LOWER LIMIT OF THE POTTSVILLE FORMATION. 831 at Mill Creek Gap, on the north side of the Hecksctierville Valley Basin, in which case the latter should be of Coal Measures age rather than Pottsville age, as has generall}^ been assumed. The reference of several of the thin coals not far below the Twin or Buck Mountain horizon to the Lower Coal Measures is not discordant with the opinion that the}" are but splits from the Buck Mountain bed,^ though I do not so regard them. It appears more probable, however, that they are distinct and earlier beds, whose geographic extent may not be great, and whose individual correlation, in any event, is uncertain. The difference between Ihe positions of the conventional formation limit in the Southern as compared with that in the Northern Anthra- cite held is no doubt due to the continued deposition, with exceed- ingly slight intermissions, of heavy conglomerates above the main plexus of egg conglomerates in the Southern Anthracite field, which, in turn, is the result of the nearness of the latter to the abundant and rapid suppl}^ of coarse sediments. LOWER LIMIT OF THE F0RMATI0:N^. Owing to the transitional character of the passage beds from the typical red shale of the Mauch Chunk to the typical gray conglomerate phase of the Pottsville, as illustrated in PL CLXXXll, the discover}^ of a constant and recognizable boundarj^ is a much more difficult matter than would at first appear, if indeed it is not an impossibility. Having observed that the upper beds of the red shale are, like the interca- lated conglomerates, irregular, unstable, and subject to disappearance by wedging or pinching out,^ the practice of selecting some arbitrarj^ boundary in the conglomeratic upper portion of the red shale was inaugurated l)y the first State geological survey, and has been followed b}" the geologists of the second surve}'. As the result of this usage in 'the type region, where the transition is the most gradual, "the fixing of a precise limit between the two formations becomes, in man}^ instances, a matter of individual preference and judgment,"^ and it fol- lows, not only that the thickness assigned to the sections varies with the geologists, but that it is often necessary to hold the published colunmar section in hand in order to find the arbitrary boundar3\ To this ele- ment of uncertainty is probably due a portion of the apparent varia- tions in the recorded thickness of the formation at certain points, as compared to that at other places, since in some sections several hun- dreds of feet below the topmost red shale have been included in the Pottsville formation, while in other regions the line has been drawn at or near the last stratum of red shale. Fortunately, toward the west- ern end, and along the northern border of the Southern Anthracite 1 Summary Final Report, Geol. Survey Pennsylvania, Vol. Ill, Pt. I, p. 2083. 2 Rogers, Geol. Pennsylvania, Vol. II, Pt. I, pp. 22, 25. 3 Smith, Summary Final Report, Vol. Ill, Pt. I, p. 1921. 832 FLORAL ZONES OF THE POTTSVILLE FORMATION. field, as well as in the other anthracite coal fields, the contact of the Pottsville formation with the Mauch Chunk is very much more sharply defined. As perhaps the less of two evils, I have adopted the topmost bed of normal red shale in each section as the lower boundary of the Potts- ville formation, thus applying- the method used under more favora))le circumstances b_v the geoloo-ists who have Avorked in the bituiuinous regions of Penns3"lvania. Though arbitrary and variable, since the topmost beds are possibly in certain cases mere washes or redepositions of the true Mauch Chunk, such a boundary line possesses at least the merit that when once seen it may usually be readily recognized by subsequent visitors to the locality. The difl'erences in the estimates of thickness, which are largely due to the lack of uniformity in selecting a basal boundary for the Potts- ville, will be illustrated in the discussion of the thickness of the formation. THE POTTSVILIiE FORMATION IX THE 13AITPHIK BASIX. Dauphin Basin and Schuylkill-Dauphin Basin are terms applied to the entire south prong of the '''fish-tail" in the western portion of the Southern Anthracite field. It is a long, narrow trough extending about 80 miles west from Lorberr^- Gap to within 1^ miles of the Sus- quehanna River at Dauphin. From a width of about 1^ miles near the eastern end and of nearh' a mile at Rattling Run, over half way toward the western end, it tapers to a narrow, relatively acute apex. Structurally the basin is essentially a simple close fold. The hard conglomerates of the Pottsville formation, which constitute the floor of the field, rise as rim walls on either side of the basin, forming the axes of Stony Mountain, the northern limb of the syncline, and Sharp Mountain, the southern limb. The profound erosion of the thick forma- tion of soft red shales on either side of the basin causes the coal field to stand out topographically as an elevated trough. The beds of the north limb, Stony Mountani, whose crest dips southeast about 70", are not so steep as those of Sharp Mountain, which, from a nearly A'ertical attitude in the region of Black Spring Gap, become overturned at Lorberry Gap to a dip of 78'^ N. AVestward from Black Spring Gap the Pottsville wall (l(>clines with gradually lessening dip as Ave approach the apex of the field. Stony Mountain presents a regular and unbroken crest. Sharp Mountain, on the other hand, is cut by six V-shaped gaps to the west of Lorberry Gn]). The softer interior Coal Measures have been gently eroded to form a generall}- shallow, rounded interior valley in the trough, which is locally more deeply cut, but still broadly rounded, by the work of the small streams whicii escape through the jaws of Pottsville conglomerate in the gaps. At Big Flats, over 8 miles from the west- WHITE.] AGE OF THE COALS IN LORBERRY GAP. 833 erii apex of the field, Sharp and Stony mountains become confluent by the convex contents of the ])asin, and from that point westward, as the successive terranes ''spoon out,'' the basin forms a single crest, known as Short Mountain. The topography of the basin is repre- sented on the Pine Grove. Lykens. Hummelstown, and Harrisburg sheets of the Topographic Atlas of the United States. The areal geology of the region, as shown in mine sheets xxi-xxvii, inclusive, of Pt. Ill of the Atlas of the Southern Anthracite Field, is essentialh^ correct except as to the mapping of the Pottsville forma- tion, which, as will presently ])e shown, is, together with the Lykens groups, located quite to the north of the Lykens groups as they actually exist in Sharp Mountain throughout the greater part of the length of the basin. Cross sections are shown on sheet xxi. Atlas Southern Anthracite Field, Pt, V; and several columnar sections, based principally on the work of R. C Tajdor and the first survey, are given on columnar-section sheet viii, Pt. IV B. Descriptions, usually of a fragmentary character, and based chiefly on Taylor's ^ reports of the explorations and developments of the field prior to IS-iO, are giA^en by Rogers,' and in more complete form, with additions derived in part f rom Ta^dor's notes, b}- A. DW. Smith, -^ both of whom reproduce in modified form the sections published by Dr. Taylor. It is to the southern or Sharp Mountain liml) of the basin that the discussion in this paper will, for the most part, be confined. The key to the stratigraphic problem in hand is the Lorberry Gap section, at the eastern extremity of the basin, and this will first be considered. AGE OF THE COALS IN LORBERRY GAP. SECTION AT LORBERRY GAP. (Station 17, Pi,. CLXXX.) On entering upon the study of the plants from the Pottsville forma- tion in the type region, when endeavoring systematically to obtain fossils from the greater number of mines or more important drifts located in diflerent parts of the field, Lorberry Gap, -i miles south of Tremont and about 2 miles south of the Lincoln mine, was selected as a favora))le locality, since in the latest published anthracite mine sheets several of the L3^kens coals are represented as mined there. This point was also thought to be especially important, because it appeared to ort'er good facilities for securing fossils directly from drifts in Lykens coal No. ti, concerning which it is still desirable to obtain more paleontologic data. 1 Two Reports on the Stony Creek Estate, 1840. See also: Report on the Swatara Mining District. Pennsylvania State legislature, 1839, p. l(i, diagram. 2Geol. Pennsylvania, Vol. II, Pt. I, pp. 19:5-198. sSummary Final Report, Vol. Ill, Pt. I, p. 2141. 20 GEOL, PT 2 53 834 FLORAL ZONES oK THE POTTSVILLE FORMATION. The section ut L()i'l)eiTv (iiip is ilesi-ritx'd by Kouer.s and tiie geolo- gists of the second State survey, on whose mine sheets (xvi and xxi) the coals have ostensibly biHMi traced from Rausch Gap westward. The cross section is given in cross-section sheet xvi, Pt. VI of the Southern Anthracite Field Atlas. These mine sheets are valuable as showing tlie |)osition and extent of the mine workings, and the dip as Avell as the tlirection of the strike of the coals. The topographic f(>:itur("s of the district are shown on the Pin(> drove sheet. Topo- graphic Atlas of the United States. Great was my surprise when, on inspecting a collection of plants from the roof shales of the southernmost coal opened in the gap— the bed mapped as Lykens coal No. 6, and apparently lying in the position of that coal with reference to the top of the red shale — I discovered the presence of a .distinctively and unmistaka])ly l^roductive Coal Measures flora, comprising species as follows: Neuropteris verinicnlaris Lx. Neurojateris limbriata Lx. Neuropteris Clarksoni Lx. Mariopteris muricata (Schloth.) Zeill. var. nervosa (Brongn.) Kid.st. Pseudopecopteris squamosa (Lx.) . Sphenopteris .susiK'cta D. W.? Pecopteris aspidioides Sternb. Pecopteris unita Brongn. Pecopteris oreoptcridia (Schloth.) Pternl). Pecopteris ^liUoni Artis? Alethopteris pennsybanira Lx.? Neuropteris ovata llofl'in. Search was then made in the shales accompanying the next higher coal, mapped as Lykens coal No. 4. mined both at Molley's slope, within the south end of the gap. and at Yoder's drift. The plants obtained at the slope are : Neuropteris Scheuchzeri Hoffni. SphenophyUum emarginatum Brongn. Lepidodendron diehotonium Sternb. Lepidophylhnn o))longifoUuni Lx. Rhabdocari30s nuiltistriatus (Presl) Lx. Rhabdocarpos jacksonensis Lx.? Neuropteris ovata HdITih. Neuropteris Scheuchzeri Hoffiu. AsterophyUites equisetiforinis (Schloth. Brongn. Sphenophylhnn emarginatum Brongn. Mariopteris cf. cordato-ovatu (Weiss). Pec(Ji)teris emarginata (Joepp. Pecopteris jjolymorpha Brongn. Aletliopteris pennsylvanica Lx.? Callij)teridium (irandini (Brongn.) Lx.? 'i'bosc gathered at Voder's drift include: I'ecopteris emarginata (ioepp. Annularia ramosa Weiss. Pecoi)teris unita Brongn. Pecopteris lepidorhachis Brongn.? Pecopteris polymori)ha Brongn. Alethopteris iH-misyivanica Lx.? Neuropteris ovata Hoffm. Neuropteris Scheuclizcri lloffm. Asteroi>hyllites ('(|ui.>;ctiforiins (Schloth.) ]'>roiiL'n. Aniudaria stellata (Schlotli.) Wood. Sphenophyllum emarginatum Brongn. Lepidodendron cf. dichotomum Sterid). Lepidostrobus cf. vaiiahilis I,. i*< II. Lepidoi)hyllum hastatum l,\..' Lepidocystis vesicularis Lx. Sigiliaria cauiptotaMna W.jod. Sigillaria ef. alternans Sternb. The llora of tbis coal indicates a still highei- horizon in Ihe Goal Measure>. To this evidence may t)e added that of the fossils from the mine dum)) at the soutli end of the gap, though the latter are less WHITE.] AGE OF THE COALS IIST LORBEREY GAP. 835 trustworthy, on account of the liabilit}' of transportation along the outlet of the valle}'. The species are as follows: Neuropteris Scheuchzeri Hoffui. Neu^()pteri^^ Clarksoni Lx. Neuropteris intlata Lx. Neuropteris Desorii Lx.? Odontopteris sp. Linopteris obliqua (Bunb. ) . Annularia stellata (Schloth.) Wood. Annulariasphenophylloides (Zenk. ) Gutb. Sphenophyllum emarginatum Brongn. Sphenophyllum euneifolium (Stern b.) Zeill.? Lycopodites Erdmanni [Lx.]. Lepidodendron dichotomum Sternb.? Lepidodendron niodulatuui Lx. Lepidostrobus Geinitzii Schimp.? Lepidophyllum oblongifolium Lx. Lepidophyllum affine Lx.? Lepidophyllum majus Brongn.? Rhabdocarpos niultistriatus (Presl) Lx. Rhabdocarjios jaeksonensis Lx.? Mariopteris sp. Mariopteris ef. Sillimanni (Lx.). Sphenopteris sp. (nov.?). Sphenopteris cristata Brongn.? Sphenopteris of. flagellaris Lx.? Oligocarpia missouriensis D. W. Aloiopteris serrula (Lx.). Pecopteris dentata Brongn.? Pecopteris unita Brongn. Pecopteris notata Lx. Pecopteris cf. pusilla Lx. Pecopteris oreopteridia (Schloth.) Sternb. Pecopteris polymorpha Brongn. Alethopteris aquilina (Schloth.) Goepp. Alethopteris pennsylvanica Lx. Callipteridium cf. Mansfieldi Lx. Neuropteris minor Lx. Neuropteris ovata Hoffm. Neuropteris vermicularis Lx. Neuropteris timbriata Lx. Neuropteris cajiitata Lx. A somewhat crude representation of the terranes in the Lorberr}^ Gap including the coals referred to is shown on PI. CLXXXY , Fig 3. In this section the ''South" and "Peacock" coals are those mapped as Lykens coals Nos. 6 and 4, respectively, on the mine sheets. Since the proximity of the Lower Coal Measures to the red shale clearly indicated the disappearance of a part or the whole of the Potts- ville formation by faulting, a stratigraphic study was next made of Sharp Mountain in this district. The results of this examination will be stated only in brief form, since the local stratigraphic conditions are apparent when once the presence of a fault is recognized. THE FAULT IN SHARP MOUNTAIN. The section of the Pottsville formation at Rausch Gap, Schitylkill County, shown on PL CLXXXV, Fig. 1, has been found to be normal, the formation being about 1,200 feet thick, and the lower coals, discussed on an earlier page, also found to belong to the Lower Lykens group. Proceeding westward from Rausch Gap,^ which is a little over 1 mile from Lorberry (jap, the protruding edges of the nearly vertical hard conglomerates of the upper part of the formation may be readily traced for a short distance along a sharp knob. However, at al)out one-third of a mile from the gap the ledges b(K;ome crushed, and the knob, topographically shown on mine sheet xxi, and on the Pine Grove 1 This gap must not be confused with that of the same name farther west, along Sharp Mountain^ in Lebanon County, or with that in the north side of the Wifonisco Basin, formerly incorrectly designated Kliiigors (ia|>. ami in a later State report Kohlers Gap. 83(l FLORAL ZONES OF THE POTTSVILLE FORMATION. sheet of tlie Topographir Atlas of the United States, is somewhat abruptly sheared in a direetion apparently X. 15' E. Westward, instead of a dense talus of massiv(> eonulomerate bowlders, whieh never fails to mark the vieinity of the stiM'ply outeropping Pottsville con- glomerates, we tind a gently rounded, smooth, broad ridge nearly devoid of all talus of a coarse type. Furthermore, over this smooth plateau surface there are scattered a nmnber of prospect shafts, in one of whieh. less than 100 yards from the crushed ends of the conglom- erates, and nearly in the prol)able strike of the horizon of Lykens coal No, 1, I collected fragments of Annularia sphejiophylloides^ t ii(q>liiilhnii arc successively found, when traced from the gap a very short distance eastward, to be somewhat abruptly transformed ' State U'K^sliitnre of Pennsylviiiiiix, 1S39. 2 For more t-xact data as to the ititervals of the coals, the reader is referred to the published sections. WHITE.] THE FAULT IN SHARP MOUNTAIISr. 837 into crushed fraginentarv talus, and to disappear on approaching- the fault line, be^'ond which we lind the ordinary' red shale. Similarly, the South coal itself is cut off at a short distance beyond the point, less than 1,000 feet from the gap, where it was abandoned on account of its squeezed condition. The red shale of the Mauch Chunk is exposed 300 feet east of Yoder's drift (1,600 feet from the gap), which is in the next higher exposed coal (Peacock), mapped in the Anthracite Atlas as L^'kens No. 4; and it is probable that had the mine gangway, which, in the miners' vernacular, "ended in fault," been driven 250 feet farther, it would have penetrated olive-green and red shales. From the foregoing details it will be seen that the somewhat oblique fault crossing Sharp Mountain just west of the knob that abuts against Rausch Gap entirely cuts oft* the Pottsville formation and a portion, at least, of the Lower Coal Measures, so that Coal Measures, probably including the greater portion of the section shown on PI. CLXXXV, Fig. 3, are thrust past the truncated Pottsville formation, or the red shales, against which the lower coals, carrying fossils clearly typical of the Productive Coal Measures, are found to abut. The Pottsville, if any part of that formation is present in Lor])erry Gap, must lie to the west of that fault and south of the lower coal (South bed) drifted in the gap. The cause of the displacement of the formations between Lorberry and Rausch gaps may perhaps be ascribed to the close group of folds to the north, and more immediately to the pressure-thrust resulting from the Georges Head anticline. As interesting, as well as corroborative of the evidence of the plants of the lower coals, I may add that the fossils from near the coal mapped as the Buck Mountain bed at the north end of the gap comprise a Coal Measures flora containing Odontoi^teris of the type of Brardii and sevtn-al small Pecopterids indicative of a very high stage in the Coal Measures. That the strata on the east side of Lorberry Creek are continued on the west is pro\'ed 1)y the extension of the lower levels in the Peacock and Umbehauer beds beneath the creek and for some distance beyond, one of the gangways in the higher coal having been driven nearly 1,500 feet west of the creek. The strike of the coal is nearl}^ parallel to that of the crest of the mountain. The continuit}^ of the series on the west side is also shown in a general wa}' by the fossil plants. Thus the flora from the rock dump at the south end on the west side of the gap may also 1)0 cited in evidence: Mariopterirt cf. conlato-ovata (Weiss). Pec(>j)tt'ris unita liroiijin. Pecopteris eiiiargiiiata Goep}). Pecopteris polyniorpha Broiifin. Alcthopteris pennsylvanica Lx.? N('nr()j)teris ovata Hoffin. Neurupteris verinicularis Lx. Neuropteris fiml)riata Lx. Neuropteris ('larksoni Lx. Neuroptoris Schevichzeri Hoffin. Anmilaria Htcllata (Schloth.) Wood. Spheiiophylluni emarjj;inatuiu Brongn. Lepidodendron sp. iiidt't. Lei>idopliylluni oblongifoliuni Lx. 838 FLOKAL ZONES OF THP: POTTSVILLE FORMATION. To tho latter may also ho added the following species from a eoal, prohahly the Peacoek, on the west side, i. e.. the eoal mined at the Molley slope, and Yoder's drift on the east side of the ga]i. and mapped as Lvkens coal No. 4: Calamites Cistii Brongn. Sigillaria camptotit'iiia Wood. Pecopteris ptcroidcs Bronirii. Pecopteris polyniorplia Brongii. Nenropteris Sclieuchzeri Hoffin. The shales from the second coal, nearly 390 feet al)ove the Peacock coal, are filled with. Aniudaria stellata^ Pecopteris. xin'da^ and vast quan- tities of Pecopteris cthorescens. The plants in the shales over the individual coals of the Productive Coal Measures of the Southern Anthracite field have not yet been studied systematically and from a stratigraphic standpoint. No attempt will, therefore, at present be made to correlate the Lower Coal Measures in Lorberry Gap l)y means of fossils. The conglomerates in the lower part of Lorberry Gap may easily ))e traced through the greater part of the distance across to Fishing Creek Gap, 2i miles to the west, but owing to the subsidence of the crest in a broad and rounded ridge slope the individual beds can not be traced quite to the Fishing Creek Gap without the aid of a careful instru- mental survey. The late State geological survey has indicated the approximate outcrop of the lower coal of Lorberry Gap as gradually diverging slightly from the crest l)eyondapoint 1 mile west of Lorlierry Gap, and as passing just south of the little knob facing Fishing Creek Gap. The boundary of the red shale was evidently thought to be necessarily parallel with that of the coal, and we find it thus traced on the mine sheet. SECTION AT FISHING CKEEK GAP. (Station 18, Pi.. CLXXX.) No detailed descrii)tion of the section in Fishing Creek Gap appears to have l)een put)lished. The geographic position of the three coals opened toward the lower end of the gap is shown in mine sheet xxi. Atlas Southern Anthracite Field. Pt. III. In this map the lower coal is correlated with the lower coal (South) mined at Lorberry Gap. and the approximate outcrop of this supposed Lykens coal No. (S is traced between the drifts in the two gaps. The upper coal (Peacock) is likewise mapped as Lykens No. 4. just as in the eastern gap. The relations of these lower coals to th(» outcrop- pini:- sandstones in Fishing Creek Gap are shown in the section. PI. CLXXNN'l. Fig. 1. the position of the red shale being recorded as it appears, immediately on the east side of the gap, which, indeed, the section re])re.sents. the upi)er coal being projected from the Avest side. It will be ot)S('rved from both tiie topography, which is low on the east, and from the (•(•lunmar section, that the conglomerates are neither WHiTE.J THE FAULT IN SHAKP MOUNTAIN. 839 heavy nor numerous. The reason for this fact will appear upon an examination of the fossils obtained at the south drifts. The flora col- lected from the lower drift, the supposed Lj^kens No. ♦>, on the east side, includes: Sphenoi:»terif! cf. mixta Schiinp. Oligocarpia cf. Brongniarti Stur? Pecopteris pusilla Lx. Pecopteris unita Brongn. Pecopteris Miltoni Artis. Pecopteris polymorpha Brongn. Neuropteris ovata Hoffni. Annularia stellata (Schloth.) Wood. That from the drift, about 50 feet higher on the same side, reveals: Mariopteris cf. cordato-ovata (AVeiss) . I Neuropteris ovata Hoffni. Pecopteris oreopteridia (Schloth.) : Neuropteris plicata [Lx.]. Sternb. j Neuropteris C'larksoni Lx. Pecopteris arl)orescens (Schloth.) Brongn. | Neuropteris Schenclizeri Hoffm. Alethopteris pennsylvanica Lx. ? Calaniites Suckowii Brongn. Sjjhenopterls plnnatifida (Lx.), -S'. cf. GravenJujr.stll Brongn., and Cordaites serpens Lx. are additional species gathered from the rock dump, which mav contain shale from l)oth drifts. The similarity of this flora to the floras listed from the Twin coal (Buck Mountain bed), in the central portion of the Southern Anthracite field, or to the flora of the lower coal in Lorberry Gap, is at once apparent. It, like those considered, is a pure Coal Measures flora. WESTERN LIMIT OF THE FAULT. It will be observed that the red shale is shown in PI. CLXXXVI, Fig. 1, as but 247 feet below the lower coal on the east side of the gap. It is not wholly improbable that the Mauch Chunk formation approaches along the line to the east of the section, even to within 120 feet of the coal. Unless the red shales in the old clearing, about 3.50 feet to the east of the gap, have been transported by human agency, as seems rather improbable under the local conditions, the Mauch Chimk lies at that point within 100 feet of the lower coal drifted in the gap. .•Vs lending color of probability to the assumption of the existence of such a diagonal contact of the two formations, I will add that along the wagon road on the west side, below the gap, I have not seen the red. shale above a point about TOO feet below the lower coal mined in the gap; this point is about 155 feet south of the forks in the wagon road. I do not, however, question the correctness in this respect of the mine map on which the shale is platted as nearh^ 110 feet north of the same road forks. To the west, and for a short distance to the north of this exposure, a zone of dark-brown ferruginous soil, accom- panied by springs, extends in a direction apparently SSW. To the east of this zone no heavy conolomcrates are sn in place, though the brown tract is strewn with loose blocks from the hillside on the west. This zone 1 believe to ])e alone" or near the fault line. 840 FLORAL ZONES OF THE POTTSYILLE FORMATION. The most important I'tict, however, in connection with the trend of the fault at Fishino- Creek Gap is the existence, at a short distance west of this brown zone and alonu" the upper part of the mountain slope, of the characteristic shoulders, l)enches. and dense talus trains of fragments of the massive Pottsville conglomerates. The entire thickness of the formation seems to have appeared opposite the first knol). at l)ut a short disttuice west of the Fishing Creek Gap. The more northern outcropping of the red shale on the east of the ga"p is (|uite in harmony with the apparently diagonal trend of the fault. That the Lower Coal Measures are continuous acrws the gap in the vieinity of the mine drifts is shown not only by the alignment of the coal horiKons and sandstones, but also ])y the character of the fossils. Those colh^cted from the lower coal on the west side are: Neuropteris ovata Hoffiu. Neiiropteris Scheuchzeri Hoffm. Calaniites Cistii Brongn. PecopterifJ ixjlymorpha lirongii. Danjt'ites sp. Alethopteris pennsylvanica Lx.? It should be recalled that Pecopteris polymoiyha is not known in the lower coals of the Lower Coal Measures, either in the ])itumi- nous basins or in the Northern Anthracite field. The species from the upper coal, al)out 170 feet higher, mapped as Lvkens coal No. 4. on the west side of the gap are: Peeopterif; miita Brongn (cf. Xcwherryi F. Neuropteris fimljriata Lx. & W.). I Linopteris obliqua (Bunb.) Pot. Pecopteris oreopteridia(Schloth.)Sternl). i Annularia stellata (Schloth.) Wood. Neuropteris ovata Hoffni. Pecopteris unita is an exclusiveh' Coal Measures species, while Neuropterk ovata and Neuropteris jimhrlata are plants having a wide vertical distribution in the Lower Coal ^Measures. On tracing the conglomerate, situated about 37 feet above this upper coal, PI. CLXXXVI, Fig. 1, westward from the gap, it was found to pass along the side of the mountain about one-third of the way up the north slope. Roughly measured, it is about 610 feet from this upper conglomerate, over the second coal, or nearly 400 feet from the lower coal, to the upper massive benches of the Pottsville forma- tion, which, for most of the way from this point to Rattling Run, form the narrow "' backbone" or crest of Sharp Mountain. For the latter half of the distance from Fi.shing Creek to the BlacU Spring Gap, to which these u]iper ledges of the Pottsville may easily l)e traced, the second of the adjacent massive conglomerate plates projects vertically from the mountain crest to form an jihnost continuous series of highly pictures<|ue " s-.iw-teeth." which are often 5(» feet in height and nearly as broad. It is hardly necessary to cite the unquestionable identify of theso ledges, which are also found at the top of the normal thickness of the WHITE] vvestp:rn limit of sharp mountain fault. 841 Pottsville in the Black Spring Gap. in order to eonlirni the relation of the coals in Fishing- Creek Gap to the Pottsville formation, since an examination of the south slope of the mountain a half mile west of Fishing Creek Gap leaves little room to doubt that long ])efore we reach a point in Sharp Mountain opposite the Fishing Creek Gap in the Second ^Mountain the entire thickness of the Pottsville formation is present in normal sequence between the Mauch Chunk and the Lower Coal Measures. In this connection it is interesting to observe that the displacement involved in the reappearance of the Pottsville and the restoration in its normal attitude of the red shale is compensated ])y a marked offset of the Pocono (Vespertine) and Catskill in Second Mountain east of and at the Fishing Creek Gap in the latter. This feature is clearh^ brought out on the Pine Grove sheet of the Topographic Atlas of the United States.' In passing it is proper to observe that the position of the '•South" coal in the Fishing Creek Gap at 350 feet or more above the horizon of the Buck Mountain bed effectually precludes the existence of any considerable portion of the Pottsville formation 'at the south end of the section in the Lorberry Gap, provided the coi-relation of the lower coals in ])oth gaps by the Penns3"lvania geologists is well founded. For my ©wn part, I am slightl}^ disposed to consider the '■"South" bed at the latter gap as not very far above the base of the Lower Coal Measures. If we hypothetically treat the South bed as a possible representative of the Skidmore coal, farther to the east along Sharp Mountain, the Lorljerr}' Gap section may with great interest be better compared with that of the water-level tunnel at Dundas colliery No. 0, at the foot of Sharp ^Mountain, a few miles to the eastward, pul)lished in columnar-section sheet viii. Atlas Southei'n Anthracite Field, Pt. IV. If the Fishing Creek section, PI. CLXXXVI, Fig. 1, be also compared with the Dundas section, the stratigraphic sequence in the region of the lower coals in the former will l)e found to be highl}' suggestive of that in the vit-inity of the Homes and Primrose coals at Dundas. POSITION OF the pottsville FORMATION ALONG SHARP MOUNTAIN. T© ignorance of the striitigraphic displacement at Lorberry Gap and Fishing Creek Gap, and the consequent erroneous identification of the coals in those gaps as Lykens coals, is directlj^ due the omis- sion of the true Pottsville formation from the region to the west mapped as coal-l)earing. For, since, in tracing these Coal Measures coals westward, they were f oun d to lie wholly to the north of the 'Lorberry Gajj and the gaps occupied by Fishing Creek in both Sharp and Second mountains evidently owe their existence to the structural weakness near the displacements. 842 FLOKAL ZONES OF THE POTTSVILLE FOKMATION. crest of Sharp Mountain, it was concluded that the crest of the moun- tain had shifted to the lower conglomerates of the Pottsville forma- tion. Hence it was natural that, as we shall presently see, the approx- imate outcrop of the Buck Mountain bed, or its equivalent, should be mapped from Black Spring Gap to Rattling Run, a distance of over 12 miles, as the "lowest Lj'kens Valley' " bed. SECTION AT BLACK SPRING GAP. (Station 19, Pi.. CLXXX.) The section at Black Spring Gap has been described by both Taylor^ and Rogers." A somewhat complete ' representation of the coal beds above the Pottsville formation is given in section 7, columnar-section sheet viii, Atlas Southern Anthracite Field, Pt. IV. A cross section is given on cross-section sheet xxi, Pt. VI of the Atlas. Reference to these publications reveals the fact that the sections begin with the top bed of Pottsville conglomerates and extend upward in the Coal Measures. This, as was just remarked, is the natural result of the identification of the lower coals in the Lower Coal Measures with the Lykens coals. The section which I give in PI. CLXXXVI, Fig. 2, is an imperfect one, since only a portion of the thickness of the massive conglomerates of the Pottsville is visible. It will be observed, how- ever, that the upper ])ed of the red shale, which is fixed with a fair degree of precision in the section, is approximately 1,150 feet below the horizon of the coal I suppose to represent the Buck Mountain (Twin) b(Hl. It will be noted, also, that the composition of the forma- tion in the Dauphin Basin is essentialh^ the same as that along Sharp Mountain, in the region of Pottsville. Especially noticeable is the great group of conglomerates which occur at the top of the formation, and which form the crest of Sharp Mountain from Fishing Creek nearly to the Big Flats. No evidence of serious search for the Lykens coals appears in this region within the limits of the Pottsville formation. SECTION AT <;OLI) MINE (iAP. (Statio.n 'JO, PL. CLXXX. I The topography of Sharp Mountain in this district and the locations of the drifts at Gold Mine Gap are shown in mine sheet xxii. Atlas Southern Anthracite Field. Pt. IV. The topography may be seen on the Lykens sheet of the Topographic Atlas of the United States. Descriptions of the coals north of the crest of the mountain are given by Taylor,^ Rogers,* and Smith," the fornuM- of whom ])ublishe(l a 'Report on the Stony Creek Estate, Pt. II, p. 16, pi. 147, fig. (i. -Geol. Pennsylviiniii, Vol. II, Pt. I, p. 195, fig. 181. See also Taylor's Report on tlie Swuturu .Mining District, l.s:59, p. IH. ^ Report on the Stony Creek E.stttte, Pt. II. i>. 19, pi. 117, fig. 5. njeol. Pennsylvania. Vol. II, Pt. I, p. P.t.i. •'•Summary Final Kciiorl, Vol. III. Pt. I. p. 'Jill. De.scriptive notes are also cDiitained in the early Report on the .'rwatara Mine Distriet, l.s;i9. ]>. is. WHITE.] POTTSVILLE FORMATION ALONG SHARP MOUNTAIN. 843 cross section. A more complete columnar section of the Productive Coal ^Measures is presented on columnar-section sheet viii, Pt. IV of the Anthracite Atlas. In PI. CLXXXVI, Fig. 3, I give an imperfect section, which is extended in PI. CLXXXVII, Fig. 1, to include the Pottsville formation. In this, as well as in the gaps on either side of Gold Mine, the immense amount of coarse conglomerate blocks, largely furnished by the upper plexus of conglomerates which form the sharp crest of the mountain, usually conceals everything but por- tions of the most enduring ledges. It is, therefore, only by some effort that outcrops of more than the most prominent beds are to be found. For the same reason I have not attempted to definitely show the upper boundarj" of the red shales, which I am convinced can hardly be less than 1,130 feet below the horizon of the Buck Mountain (Twin) coal. It is interesting to note a certain degree of regularity in the group of conglomerates at the top of the Pottsville. This appears on a com- parison of the Gold ]Mine Gap section with the sections at Rausch Gap (PI. CLXXXVII, Fig. 2) and Black Spring Gap (PI. CLXXXVI, Fig. 2), to both of which these outcropping backbone ledges may easih' be traced. A similar development exists at the Rausch Gap in Schuylkill County (Station 4, PI. CLXXX), in which the Buck Moun- tain coal is similarly disposed. From the mine dump, consisting of material from probably more than one of the coals drifted on the east side of the gap, were obtained the following species, which evidently represent a characteristically Productive Coal Measures flora: Pecopteris unita Brongn. Pecopteris arguta Sternb.? Pecopteris arborescens (Schloth.) Brongn. Pecopteris squamosa Lx. Pecopteris oreopteridia (Schloth.) Sternb. Pecopteris cf. elHptica Bunb. ? Pecopteris polymorpha Brongn. Alethopteris pennsylvanica Lx.? Caniptericlium Grandini (Brongn.) Lx. ? Neuroi^teris ovata Hoffm. Neuropteris vermicularis Lx. Neuropteris fimbriata Lx. Neuropteris Clarksoni Lx. Neuropteris Scheuchzeri Hoffm. Neuropteris Rogersi Lx. LepidophyUum inajus Brongn. ? Cordaites sp. The following species were also collected on the west side in the gap: Pecopteris polymorpha Brongn. Pecopteris cf. Newberryi F. and W. Neuropteris ovata Hoffm. Neuropteris Scheuchzeri Hoffm. Annularia sphenophylloides (Zenk.) Gutb. Sphenophyllnm cf. tiliculnie Lx. SECTION AT RAUSCH GAP, LEBANON COUNTY. (Station 21, Pl. CLXXX.) The Lower Coal Pleasures and the upper beds of the Pottsville forma- tion at Rausch Gap, 3^ miles west of Gold Mine Gap, have been the 844 FLORAL ZONES OF THE POTTSVILLE FORMATION. .su))ject of fraginontary descriptions ])y Taylor/ Rogers," and Suiith.^ The cross section giveii by the first-named author is repeated in more comph^teform hy Kogers and the second geological survey', though in the pul)lications of the latter the boundary of the red shale appears to have been drawn where it might theoretical!}' lie if the Buck Mountain (Twin) coal were the lowest Lykens coal. The position and extent of the drifting in the vicinity of this gap are shown in mine sheet xxiii. From the section which I give in PL CLXXXVII, Fig. 2, it appears that the Pitch l)ed, the next coal below Bill's V)ed, pro})ably lies within the group of conglomeratic plates at the top of the Pottsville forma- tion. The roof shales of the coal, which I interpret as the probable rep- resentative of the Buck Mountain coal, furnish the following species: Marioj^teris marietta (Schloth.)Zeill. var. nervosa (Brongn.) Kidst. Mariopteris ocfidentali;^ T>. W. Pecopteri.s uiiita Brongn. Pecopteri.s pusilla Lx.? Alethopteris aquilina (Schloth.) Goepp. Neuropteris Desorii Lx. Neuropteris ovata Hoffni. Neuropteris vermieularis Lx. Neuropteris fiml^riata Lx. Neuropteris Scheuchzeri Hoffm. Linopteris obliqua{Bunb.) Pot. The composition of this fiora indicates a horizon in the basal portion of the Lower Coal ^Measures, or in the Allegheny series. A drift in the same horizon on the west side of the gap 3'ields large numbers of Aldltopterls Serlii (Brongn.) Goepp., Neurojjteris ScheuchzeriYLo^m.. ^ and Aster(>j>hijll/frs equl>< (Schloth.) Brongn. P>el<)w the Pitch l)ed there is but little, if any, evidence of search for coals in the Pottsville formation at Rausch Gap. Here, as well as in Gold ]\Iine and Black Spring gaps, effort was made to find the coals of the Lorberry and Fishing Creek gap sections, which were seen to pass, in the Dauphin Basin, to the north of the crest of Sharp Mountain instead of to the south of the latter. In his discussion of the region under consideration. Dr. Taylor,* whose reports and notes form the basis of the later publications relating to the district, remarks that " no examination for veins [coals] south of the backbone ledge of Sharp Mountain has taken place herea])Outs.'' Owing to the except ionall}^ favoiablc conditions at the southeast corner of the gap, the top of the red shale was located by me, with a prol)ably high degree of certainty, at about 1,175 feet l^elow Bill's coal, which I have assumed to be near the horizon of the Buck Mountain bed. The boundary can be hardly more than 75 feet higher. YELLOW SPRINGS GAP. (Station 'JS, Pi,. CLXXX.) At "^'('llow Spring.s, \\ miles west of Rausch Gap, there is a high gap in the mountain, through which a .small stream, draining a section about 1 Report on the Stony Creek Estate, Pt. II, p. 19, pi. 147, (ig. 4. 2Geol. Peiinsylviinia, Vol. II, Pt. I, p. 19.5. fig. 182. •■'Summary Final Report. Vol. Ill, Pt. I, p. •.'11-1. <0p. cit., p. 21). WHITE] POTTSVILLE FORMATION ALONG SHARP MOUNTAIN. 845 2 miles in leng'th of the Coal Measures valley, finds an escape. The topography of the vicinity is shown on the Hummelstown sheet of the Topographic Atias of the United States and mine sheet xxiv of the Southern Anthracite Atlas, the latter of which locates the points of exploitation. Descriptions of the coals are given by Tajdor,' and quoted l)y Rogers" and Smith.'' Crosss sections of the basin at this point are given by Taylor, and by the geological survey of the State in cross-section sheet xxi, Pt. VI of the Anthracite Atlas. All the information in the later reports relating to Sharp Mountain appears to have been derived from Taylor's report, printed in 1840. At this gap the Pottsville appears to present its ordinary characters and its full thickness. The conditions for the'discovery of the upper boundary of the Mauch Chunk shale are not favorable, but it is quite certain that the red shale is not present at 960 feet below the supposed horizon of the Buck ]Mountain coal. No prospecting appears to have been done in the Pottsville formation at this gap since the publication of the report by Taylor, who states that none of the coals on the south slope of Sharp ]\Iountain had been opened or sought. However, in some early explorations, carried on in 1824, a tunnel driven through the upper portion of the Pottsville formation in the gap penetrated a bed of good coal, which. Dr. Taylor adds, was "not fully proven." The rock dump taken from a shaft which appears to have been located on this bed contains the following species: Eremopteris decipieng. Mariopteris tennesseeaiia var. hirsuta. Sphenopteris palmatiloba var. squarrosa. Megalopteris plumosa. Neriopteritf lanceolata. Neuropteris tennesseeaiia. This flora will at once be recognized as clearly referable to the Sewanee zone; and it can hardly be lower than Lykens coal No. 3. The variety of Sjjhenopterls palmatiloha, the variety of Cardiocarpon hkmpklatam, and Bothrodendron arhorescens appear to bind the flora somewhat closely to that of the horizon of Lykens No. 1. Through Ernnopti'vk declpiens and Cardiocarpmi Wilcoxi the flora seems to be related to that of the Sharon coal of northwestern Pennsylvania, and it is worthy of mention that Nenopterh lanceolata, found at Yellow Springs, has hitherto been known only from the sandy shales at some distance above the Sharon coal, in northwestern Ohio. On the whole, I am slightly disposed to regard the horizon of the plants at Yellow Springs as higher than that of the Lykens coal No. 2. The stratigraphic distance of this horizon below that of the Buck 1 Reports on the Stony Greek Estate, Pt. I, p. 52. " Oeol. IVnnsylvania, Vol. II, Pt. I, p. 19(J, fig. 183. asuinmary Final Report, Vol. Ill, Pt. I, p. 2145. Calamostachys Knowltoniana. Bothrodendron arboreseeii!^. Cordaites Robbii. Cardiocarpon biscupidatuni var.oliioense. Cardiocarpon Wilcoxi. 846 FLORAL ZONES OF THE POTTSVILLE FORMATION. IMouiitain coal is about 2iU feet; and as the latter is prol)a])ly recog- nizable in this gap, this points toward the level of Lykens coal No. 1. The plants from the Backbone bed, which appears to correspond to the Buck ^Mountain coal at Yellow Springs Gap, constitute a flora agreeing well with that listed from what I believe to be the same horizon at Rausch Gap. The^' are: Sphenopteris sp. Pecopteris villoma Broiign.? Neuropteris ovata Hoffiu. NeuropterLs plicata [Lx.]. Neuropteris fimbriata Lx. Neuropteris Scheuchzeri Hoffm. Linopteris cf. squarrosa (Ett.). Sphenophyllmn emarginatuni P>rongn. Neuropteris capitata Lx. I Rhal)docarpos tenax Lx.'? Before passing farther toward the apex of the field, certain sugges- tions, resulting from a comparison of the columnar sections at Yellow Springs and the gaps to the eastward, deserve some attention, although it is not within the province of this paper to attempt the correlation of the coals of the Lower Coal Measures. K(>ferring to the section at Kausch Gap (PI. CLXXXVII, Fig. 2), we may ()t)serve that Bill's ])ed. the '* 3-foot" bed, the '•2-foot" bed, and pr()t)ably ''Dan's" bed, have been drifted in the Yellow Springs Gap, in wliich the intervals separating the coals are nearly the same as in Rausch Gap. A hig-her coal opened at Yellow Springs is probably e(|uivalent to that designated the ''Heister" in Rausch Gap, though it may possi- bly be the representative of the coal next below. We may safely conrhide. therefore, on a comparison of these two sections, that the coal in the Pottsville formation from which the Lykens plants enu- merated a))ove were ol)tain(Hl lies approximately 21<» feet below BilFs bed in Rausch Gap, 4i miles to the eastward. The section of the Lower Coal Measures at (xold ]Mine Gap exhibits, when compared with that at Rausch Gap, considerable \ariation. If the supposition that the coal which I have designated in the section (PI. CLXXXVl, Fig. 3) the Buck Mountain (Twin) l)ed may be the ec|uivalent of the Bill's bed at Rausch Gap is correct, then the })od known as ""-Ir-foot" bed in the (iold Mine Gap section would appear to l)e (Miuivalent to one or both benchi\sof "Dan's bed" at Rausch (rap; in which case the *' P(Micock" and the coal next above may be par.UleliziHl with coal "Xo. 1 "and the '"Heister" bed, respectively, in the Kausch Gap section, while the two coals next higher at Gold ]Mine may l)e consid ered as representatives of the "(iray" bed and "No. 2" bed at Rausch Gap. The ••No. 4" bed in Rausch Gap lies at the approximate level, stratigraphically, of the "3-foot" bed in Gold Mine Gap. It is obvious that, if these tentative correlations, especially as they relate to the "No. 1(0,'' the •• 1 leister." the "Gray," and the ''No. 2" beds at Rausch (iap, are not ei-roneous, the ••Ileister" bed, which, as designated in the published sections, is al»()ut 120 feet above the '•3-f()()t" bed at Gold Mine (iap. can not p()>sibly be the e(|ui\alen( of the bed bearing the >ame name in the Kausch (iap section, less than ;').l miles distant. WHITE] POTTSVILLE FORMATION ALONG SHARP MOUNTAIN. 847 In the former section the supposed Heister bed is nearly 870 feet above the horizon of Bill's bed, while in the latter section it is only aliout 420 feet, less than half as far. Similarly, the section at the Black Spring Gap, often cited as Mount Eagle Gap, when compared with that in either Gold Mine Gap or Rausch Creek Gap, presents a series of intervals and coals which suggest several tentative correlations. Thus the horizon which I have desig- nated in PI. CLXXXVl, Fig. 2, as the place of the Buck Mountain (Twin) coal is probably on the same stratigraphic level as BilFs bed in Rausch Gap, or the lower coal at Gold Mine Gap. Likewise the hori- zon higher in Black (Spring Gap, described by Taylor as "traces of a southern coal," would seem to correspond directly to the first coal above the Buck Mountain in Gold Mine Gap, while the "4-foot" beds in both gaps will, in that case, be on the same stratigraphic level; but if, as would seem naturalh' to follow, the "Peacock" coals in both sections are in reality equivalent, then we must conclude not only that the Pitch bed is not developed in the Gold Mine Gap, l)ut also that the rep- resentative of the next coal above the "Peacock" bed at Gold Mine is identical, if exposed at all at Black Spring Gap, with the Black Spring coal. The Mount Eagle coal, the next higher in the latter gap, and the second coal opened above the "Peacock" coal in Gold Mine Gap, both of which are coincidently situated at the same distance from the Twin coal, prol^ably represent the same bed. But, in order tt) illus- trate the ease with which sections containing a number of well-distrib- uted coals may l)e in dili'erent ways adjusted to one another, while at the same time pointing out certain other probable coincident similarities between the Black Spring section and that at Rausch Gap, let us assume that the horizon designated as "traces of a southern coal" in the former gap represents the Bill's bed in the latter. In that case, we shall find not only that the "4-foot" coal lies at essentially the same distance above the second coal in Gold Mine, but also that the "Peacock" coal is near the level of the "4-foot" bed at Gold Mine and the "Dan's bed" at Rausch Gap, in which case the Pitch coal at Black Spring Gap might, without too great a strain of the imagina- tion, be correlated with the "2-foot" coal at Rausch Gap, and the '"4-foot" coal at Black Spring might correspond to the "3-foot" coal at Rausch Gap. Continuing the same assumption as to the identit}^ of the horizon of Bill's bed or the Buck Mountain coal at Moiuit Eagle Gap, it is evident at a glance that the Black Spring and Mount Eagle coals near the top of the section at Black Spring correspond, so far as stratigraphic intervals are concerned, almost exactly with the "No. 1" and the '"Ileistcr" coals, respectively, in the Rausch Gap section. The quoted names of coals represent the local identifications or corre- lations by the State geologists. So far as the problems discussed in tliis report are concerned, the 848 FLORAL ZONES OF THE POTTSVILLE FORMATION. chief interest in tluM-orrelatioii of the coals of the Lower Coal Measures relates to th(^ e(iuivalence. in the western sections, of the coals mined in the Fishino- Creek and L()r])err3' gaps. My own observations of the terranes in the Pishing C^reek ^Gap (PI. CLXXXVI, Fig. 1), com- bined with measurements across Sharp Mountain, less than a mile west of the creek, where the Pottsville formation is present in its normal eoMstitution and thickness, lead me to the opinion that the horizons of the two lower coals mined in Fishing Creek Gap are comparable to those of the "Peacock" coal in the Black Spring and Gold Mine sections and the coal marked as " Heister " in the Rausch Gap section. In this case the upper coal at Fishing Creek ma}' represent the "Gray bed"' at Rausch Creek Gap, the second coal al)ove the sup- posed "Peacock" in Gold Mine Gap, and probably the Black Spring- coal at Black Spring Gap, the lowest coal at Fishing Creek being possibly comparal)le to the Pitch bed at Black Spring. Tlu> coals in the section at Lorberry Gap, PI. CLXXXV, Fig. 3, will appear to correspond most satisfactorily, with respect to the separating inter\'als, to the coals of the gaps to the westward, if we assume that the lowest bed (South coal) opened at Lorbej'ry occupies the horizon of the snpposed Buck Mountain, Backbone, and Bill's beds in the other gaps. If this hypothesis is correct, the })ed designated "Peacock" coal in Lor- berry, Black Spring, and Gold Mine gaps, which, in the first two sec- tions, is approximately the same distance from the Buck Mountain bed, and which is scarcely farther in the Gold Mine Gap, may per- haps safely be considered as one coal. Next, the ' ' Zimmerman " coal in Lorberry Gap would appear to deserve comparison with the Mount Eagle bed at Bhick Spring, with the second exposed coal above the " Peacock" })ed in Gold Mine Gap, and with either the "Gray" bed or "No. 2" bed. next higher, in Rausch Gap. On assuming that the South coal at Lorberry is at the horizon of the I)Utk Mountain (Twin) bed, it becomes probable that the developed coal> above the '•Zimmerman " bed are higher than those represented in the Rausch Gap section, ludess the "No. 4" bed in the latter section, whose distance above the Buck Mountain bed is a little greater than the corresponding interval below the next coaP above the "Zinnuerman" in the Lorberry (iap, is the representative of the latter unnamed coal. The next higher coals in the Lorberry section will then deserve compar- ison with the " Heister" and " Gray" beds in the Gold Mine Gap section. The rm])ehauer and the Furnace beds at the north end of Lorberry Gap probably represent higher horizons than have l)een developed to the westward, and it is possible that they may not ev(Mi b(> ]iresent ill the hasin at and to the westward of Rausch Gaj). It will be observed that in respect to the relative intervals Ix'tween 1 This appt-arsto Ix? identical with the bc-fl iiaineil Peacock coal in cros-s-sectioii sheet xvi, I't. VI of tlie Athis of the Southern Anthracite Field. WHITE] rOTTtSVILLE FOEMATION ALONG SHARP MOUNTAIN. 849 the coals in the sections, the Gold Mine section appears to present the highest degree of agreement with or similarity to the Lorberry Gap section; but, whether the comparison be with the Gold Mine Gap or the Mount Eagle Gap, or the Dundas No. 6 collier}^ tunnel,"^ a few miles to the east of Lorberry Gap, we must conclude that if the South coal represents the Buck Mountain l)ef1 the Lorberry Gap should, in addi- tion, contain a number of the intermediate coals which have been opened in the Black Spring and Rausch gaps. With no other evidence than the measurements of the intervals between the discovered coals on which to base correlations of the latter, an almost equall}^ satisfactory parallelization of the beds might be framed were we to assume that the ''Peacock" coal in the Lorberry Gap corresponds with the first coal above the Buck Mountain bed in the lounger sections to the west- ward. Such an assumption is, however, manifestly untenable, since it involves the reference of the South coal, whose fossils are fully as young as, if not younger than, those of the Buck Mountain bed, to the approximate horizon of the Lykens coal in the Yellow Springs Gap, the fossils from which are referable to the Sewanee zone of the Upper Lykens division of the Pottsville formation. 1 would have little reliance placed in the foregoing suggestions as to the equivalence of the various coals of the Productive Coal Meas- ures along Sharp Mountain. It needs but an examination of the columnar sections, showing the surprisingly great variation of the intervals between the coals as actually ascertained by direct connec- tions between the mines in the Southern tield," to convince one that correlation of these beds by no other means than the comparison of columnar sections is, in the Southern Anthracite tield, hardlv less than jugglery. RATTLING RUN GAP. (Station 2-1, Pl. CLXXX.) Rattling Run, the most westerly gap in Sharp Mountain, is 15^ miles from Fishing Creek Gap and about -i miles east of the point at which the Pottsville formation begins to spoon out, above Water- tank Station. The. description of this gap, which is about 3 miles west of Yellow Springs, is given with considerable detail in Dr. Taylor's report,* which is quoted in the later State reports.* The principal drifts in the Lower Coal Measures are platted on mine sheet 1 Section 6, columnar-section sheet viii, Atlas Southern Anthracite Field, Pt. IV. 2 The variability of the Coal Measures intervals, even between near localities, is well illustrated in the diagram prepared by Ashburner, and published on columnar-section sheet iii, Pt. I of the Atlas of the Southern Anthracite Field. It is also shown in columnar-section sheet vi, Pt. II of the Atlas. Good examples of this are found at the Wood's colliery, and at Dundas colliery. No. G tunnels, cited above, the sections of which are not more than 3 miles distant from each otlier. sReporton Stony Creek Estate, iii>. II and 50, pi. ] 17, lijr. 'J. ^Rogers, Geol. Pennsylvania, \'(il. II. I't. I, \i. i;t7, .Sinitli, Suiiuiiary Final Report, Vol. Ill, Pt. I, p. 2145. 20 GEOL. I'T 2 54 850 FLORAL ZONES OF THE POTTSVILLE FORMATION. XXV of the Anthracite Atlas. That portion of the Dauphin Basin west of Yellow Spriny's was somewhat thoroug-hly explored l)y the owners, the Dauphin and Susquehanna Coal Company, under the direc- tion of Dr. Taylor. It is to this fact that most of our knowledge of the coals in this reg-ion is due. The entire thickness of the PottsvilU^ formation, from the ))ed which 1 assiune to represent the Twin (Buck Mountain) coal to the top- most beds of the red shale is given by Taylor as 1,100 feet. Within that interval twelve coals were found. Two thin coals are reported within 170 feet of the red shale. The sixth coal from the red shale, the Reliance l)ed, not over 3 feet in thickness, was drifted for -lOl feet. Although a very "dry coal," it was not found tol)e at that time profit- al)ly workal)le. The distance, 410 feet, from this bed to the red shale suggests the position of Lykens coal No. 4. Twenty feet to the south of the Keliance bed another coal was opened, and still another l)ed was discovered 40 feet to the north. A coal at the supposed horizon of the Buck Mountain lied has l)een opened on both sides of the gap, it having l)een drifted for a considerable distance on the east side. No fossil plants were obtained at this gap. BIG FLATS. (Statiox 20, PL. CLXXX.) As Sharp and Stony mountains, which form the two walls of the Dauphin Basin, converge toward the west, the interior valley becomes less marked, until at a point nearly opposite Watertank Station, about 9 miles fi'om Dauphin, the Coal Measures completely till the interval from rim to rim and form a low, slightly convex knob, the Big Flats, about 1 mile in length, the north and south faces being composed of the Pottsville formation. The topography is shown on the Harrisburg sheet of the Topographic Atlas of the United States, as well as on mine sheets xxv and xxvi, Ft. Ill of the Anthracite Atlas. The shaftings on the Big Flats, dating from the early part of the century, are described by Tavlor, whose cross section was repro- duced by Rogei's.' Dr. Taylor reports the presence of three coals within an interval of 48 feet, from one of Avhich several hundred tons of coal were hauled to the Susquehanna River at Dauphin prior to 1840. The three principal shafts, l)ut a few yards apart, are platted on mine sheet xxvi. The shale from the shafts is now nearly disin- tegrated; yet fragments taken from the duni]) reveal the presence of Xruroptri'tx oriiid Ilollin.. N. ScJuKehzei'l IIollui.. Ptcopft-vh rUJom Brongn. ^ Arnt nhi rhi tph >/U um emar- i/iiHifiiiii Brongn., species chai'acteristic of the Lower Coal Measures. Althouiih we have no measurements showing the thickness of the ' Geol. Penusylvanitt, Vol. II, Pt. I, p. 198, fig. 184. WHITE.] SHOKT MOUNTAIN SHAFTINGS WEST OF BIG FLATS. 851 Pottsville formation west of Rattling- Gun Gap, my observations show that the upper group of conglomerate plates which have been traced the entire distance from Fishing Creek Gap nux}^ be followed, l)v the exercise of proper caution, when the trees are bare, for at least 2 miles west of Rattling- Run Gap. Westward from this point the amount of the error of the omission of the entire Lykens g-roup of coals on the State maps diminishes gradually. Yet in view of the lessening of the dips from 73- to IT- ' opposite the J^g- Flats shaftings, it is obvious that the supposed boundary of the lowest Lykens coal, which is rep- resented at a distance of not over 800 feet from the shafts, is too far north to include more than the upper 200 or 300 feet of the Pottsville formation, even if we suppose the coal mined in the shafts to be the Buck Mountain bed. than which, as is shown by the plants, it can hardly be lower. Were the strata nearlv vertical, it is proba])le that Lykens coals Nos. 2 and 3 would lie outside of the approximate boundary of the lowest Lykens bed as the latter is mapped in the mine sheet. That coals are present in both of the Lykens groups in this region is evident from the shaftings along Stony Mountain and Short Mountain, which will next be considered. SHORT MOUNTAIN SHAFTINGS WEST OF BIG FLATS. The disappearance of the Lower Coal Measures and the commence- ment of the spooning out of the upper beds of the Pottsville formation occur not far west of the Big Flats and the head of Watertank Run. It is certain that, unless the folding is much closer than the apparent dips indicate and the axis is not far to the north side of the crest, the Pottsville alone remains at the Fort Lookout shafting, less than li miles west of Big Flats. It should be remarked at this point that, with the exception of the Bavard shaftings on the north side of Short IMountain. no prospecting or exploitation of coal appears to have })een made in this region since the earlier half of the present century. In the following brief notes reference Avill be for the most part con- fined to those old developments from which fossil plants have 1)een obtained. The descriptive matter is derived from Taylor's report, the source of most of the information repul)lished in the State reports. A. The first of the developments west of the Big- Flats are the Fort Lookout shaftings (Station 27, PI. CLXXX), opened in 1S38. The location of this operation is shown in mine sheet xxvi, Pt. Ill of the Anthracite Atlas, and a cross section of the basin at this point is con- tained on cross-section sheet xxi, Pt. VI of the sanu^ pul)lication. The two shafts at this point are said to have reached a depth of 82^ feet. The coal was found too poor for mining. On searching the dump at the mouth of the shafts, a number of fossil plants were obtained, _^ . '. . » * See mine sheet No. xxvi. 852 FLORAL ZONES OF THE POTTSVILLE FORMATION. Avhich. as ^vill at oiu-o be clLscovenul in ulancino- at the list, arc dis- tiiKtlv of I'ottsville ao-o: Asterophyllites arkangamis. Annularia acicularis. Whittlef^eva elt'sians \ar. minor. ^Iari<)i)teris tennesseeana. Neuropterif! gigantea. Xeuropteris acutimontana. Neuropteris 8p. indet. Furthermore, nearly all of the species will be reeognized as having a distribution in the Upper Lykens division. It is certain that this flora can not be below the horizon of Lj'kens coal No. o, when Xciirop- tei'k gigantea^ Asterophyllites arhinmnus, and the peculiar form of Jfar/'opte/'is teiinesseeana strongly suggest a more intimate relation with the flora of Lykens coal No. 2. It is probable that the level of this flora is not lower than Lykens coal No. 2 or higher than Lykens coal No. 1. B. Of the economic results obtained by the Bayard shaftings (Station 28, PI. CLXXX) we have no other information than that given by mine sheet xxvi, which shows the location of four shafts, and by the fossils obtained from the rock dumps. From the lower of the shafts there were obtained an indeterminate species of Maviopteris^ Neurojyteris Pvcahmtas var. inaequalk, and a species of Sporocy-'^tts. From a higher shaft, not over 85 feet from the latter, the following-named species were obtained in shale mingled with waste coal: Mari last. From the thickness and order of the intervals (see tal)li\ p. 8(54). we may tentativeh' assume that the lowest bed represents Lykens coal 'Suininiiry Fiiiiil Kuport, Vnl. Ill, I't. I, \>. JMJ. WHITE.] LYKENS COALS IN STONY MOUNTAIN. 855 No. G; the thick bed, 40 or 50 feet higher, maj^ then be the Lykens coal No. 5, and the 2-foot 7-inch bed of good coal 100 feet above the last will perhaps correspond to the place of L3'kens coal No. 4, while the coal about 150 feet higher is possibl}^ near Lykens coal No. 2 or No. 3. As corroborating to a certain degree, or as slightly indicative of the correctness of, these hypothetical correlations, the small col- lection of plants apparent!}' derived from the roof of the third coal (numbering from the lowest), which we have assumed to be L3'kens coal No. 4, may be enumerated: jNIariopteris pottsvillea. I Asterophyilites parvulus. NeuiMpteris Pdcahoiitas. | Trigonocarpum ampullaeforme. The tirst of these species seems to be characteristic of the zone of that coal, while the third is more common in the same horizon. On the whole, it appears very probable that the three principal lower Lykens coals have been opened in the prospect shafts on the Dull and Hofi" lands. At a point nearly north of Rausch Gap^ two coals, which from sur- face appearances and the thickness of the intervals would seem to rep- resent the supposed Lykens coals Nos. 4 and 5 at the locality last considered, have been opened by trial slopes (Station 22, PI. CLXXX). The coal at the mouth of the upper of the two slopes is apparently of good qualit}' and in good condition. No information is at hand as to the thickness of the Ijeds in the slopes, which are now fallen shut. The lack of information is in itself indirectly indicative of no great thick- ness for the combustible. At the Kalmia colliery (Station 41, PI. CLXXX) Lykens coals Nos. 4, 5, and 0 were worked to some extent. Owing, however, both to the irregularity of the beds at this point in thiclvuess and condition and to the more advantageous conditions for mining about the Georges Head anticline, the greater part of the ''workings*' were aliandoned in favor of the latter area. The columnar section at this mine from the top of Lvkens coal No. 5 downward into the red shale is shown on PI. CLXXXiy in continuation with the section at the Lincoln collier}", from the connected gangways' of which a portion of the Kalmia ter- ritory is now directly mined. GENERAL CONDITIONS RELATING TO THE OCCURRENCE OF THE LYKENS COALS IN THE DAUPHIN BASIN. A review of the foregoing l)rief descriptive notes concerning the Pottsville formation in the Dauphin Basin shows that along the north side of the narrow trough, which is nowhere more than 2 miles in width, ^ several of the Lykens coals, one or more of which, usually in 1 Atlas Southern Anthracite Field, Pt. Ill, mine sheet xxiii. 2 Idem, rt. III. mine sheets xxi and xxii. *This refers to the distance across the basin from margin to luarKin; not to the length of the iiirve. 850 FLORAL ZONES OF THE POTTSVILLE FORMATION. the Lower Lvkens group, is nearly or quite of workable thickness, have lieen found in every district where a thorough search has lieen made. They are also found to extend along the base of the rising- axis in Short Mountain. Furthermore, it has been shown that, owing to the presence of an uno])served fault which cuts off the whole or nearly all of the Potts\-ille formation at I^orberry and Fishing Creek gaps, not only were the soft, inferior Productive Coal Measures coals exploited at tliese gaps mistaken for degenerate developments of the Lvkens coals, and consequently pronounced inferior or worthless, but on account of the trend of the former coals along the north side of Sliarp Mountain the. entire Fottsville group of coals has been sup- posed to lie to the north of the crest of the mountain. The outcrop, or su})posed approximate outcrop, of the lowest Lykens coal was there- fori' mapped by the late anthracite survey of Pennsylvania' along or near a horizon not lower in most places than the horizon of the Buck jNIountain coal, the conventional base of the Lower Coal Measures, from Fishing Creek Gap to a point about 2 mil(\s west of the Rattling Run (nip, an entire distance of over IT miles. To the same misinter- pretation at Lorberry and Fishing Creek gaps is also due the fact that no systematic search has ever ])een made for coals south of the crest of Sharp ^Mountain (where no coals were supposed to occur) between Fi-liing Creek Gap and Rattling Run Gap. It must not ])e understood from the above statement of facts that Lykens coals in good condition lie awaiting the search of the pros- pector along the south slope of Sharp ^Mountain. On the other hand, the vertical or very highly inclined attitude and the often crushed or slipped condition of the other coals along Sharp Mountain render it proljal)le that the Lykens coals will here also be found generally infe- rior in structure, and perhaps in composition, as compared with the present standard requisite for protitable mining. It is the purpose of this review of the stratigraphy of the region not merely to secure greater accuracy in the mine maps of the Dauphin Basin, or to add to our knowledge of the floras of the Pottsville formation in this region. l)ut to call attention to the facts: (1) That the soft or semi- bituminous coals on the north slope of Sharp Mountain between Fish- ing Creek and Rattling Run gaps, hitherto regarded as the Lykens coals, are really in the l*roductive Coal ]\[easuri>s; (2) that practically no search" has been made for coals in the Pottsville formation through- 'Crwlit is due the opinion expressed by Mr. A. DW. Smith in a footnote to the Summary Final Report of the State Survey (p. 2140), that the outcrops of the red shale and the lowest Lykens coal are drawn 800 to 1,000 feet too far north between Lorberry Gap and Rattling Run Gap, most of the coals in the Lorberry and Fishing Creek gaps being referable to the Lower Coal Measures, although he assumes the full thickness of the Pottsville formation to be present at the latter gaps. This foot- note, which I had Tiot seen until the writing of the i>resent i>apcr, is (juite at variance with all other portions of the text relating to tlie Dauphin Basin in Mr. Smith's report. -'Kxceptions of little importance are the discoveries of the Reliance coal at Rattling Run is of 500 to GOO feet. The transition beds and the lower beds of XII also exhibit decided variations in the materials composing them. At times heavy conglomerates predominate, with but few sandstones and shales, or again the whole series may be composed of coarse saTidstone.s and of shales, with the green and reddish tinge running high in tlie formation, making it difficult, 858 FLOKAL ZONES OF THE TOTTSVILLE FORMATION. transition, 400 or 500 feet in thickness, such us is exhi])ited in the sec- tions at the Pottsville Gap, Pis. CLXXXI, CLXXXII, and in the Lincoln region. Pis. CLXXXII, CLXXXIII. It is purposed in the following pages to present the results obtained by both methods. Begiiuiing with Mauch Chunk, at the eastern apex of the Southern field, and proceeding westward, the measurements of the Pottsville formation (XII), as given l)y Rogers,^ are: Mauch Chunk, about 950 feet; Nesquehoning, 792 feet; Tamaqua, about 803 feet; Pottsville, aljout 1,030 feet; Lorberry Gap, about 675 feet;^ Yellow Springs, about 660 feet;^ Kohlers Gap, 230 feet; Bear Gap, 460 feet. The measurements of all the intervals given by A. DW. Smith are of great value, since his statements are based on the enormous amount of instrumentally accurate data accunudated by the second geological survey of the anthracite regions, all of which were passed in review b}- him. As stated b}^ Mr. Smith, in the Final Summary Report, the thickness of the formation is as follows: Locust Gap, Tamaqua, 1,296 feet; Sharp Mountain Gap, Tamaqua, 1,130 feet; Broad Mountain, about 1,200 feet; Pottsville Gap, 1,350 feet; Swatara and Rausch gaps, 1,100 or 1,200 feet; Lorberry Gap, 1,500 or 1,600 feet;' vicinity of the Lin- coln mine, 1,475 feet; Kalmia region, 1,400 to 1,500 feet; Williams- town, about 1,400 feet. The preceding measurements begin with an ar])itrary Ijoundary, usually Avithin or below the transition series. The following measure- ments start from the topmost bed of red shale and extend to the supposed horizon of the Buck Mountain coal, except along Locust ^Mountain, in the Panther Creek Basin, where the measurements from both the A and B beds are given, it being nevertheless under- stood that the A coal at Tamaqua is referable to the Lower Coal Meas- ures.-^ The measurements opposite the names of localities marked l)y an asterisk (*) are compiled from the sections pul)lished by the State geological surve}-. even when a complete section is at hand, to decide where the line between the two formations should be drawn. It is not wife to always take the highest red shale bed as a Hiiiit, as Ix'ds of red shale, usually thin, but in appearance like the mass of No. XI, are not infrequently seen high up in the conglomerates of XII, and occasionally among the overlying Coal Measures; nor will it suffice to take the lowest conglomerate, as beds of conglomerate are often found well down in the red shales of XI. The fixing of a precise limit between the two formations becomes, in many instances, a mat- ter of individual i>reference and judgment." ' 'ieol. Pennsylvania, Vol. II, Pt. I, pp. 14C and 147. •-This, as we have already seen, consists in part, if not wholly, of the beds of tlio Productive Coal Mea.surcs. 3It is difficult to account for this measurement by Rogers at Yellow Springs, except on the supposi- ti(m that the dense, ferruginous surface deposits which occur in the lower end of the gap were mis- taken by him as indicating the presence of the Mauch Chunk red shale. : 230 Rausch (iap, Lebanon ( ounty Williamstown I Rattling Run Measured by- Rogers. Smith. 1,155 878 1,296 1,130 1,350 1,200 1,100 to 1,200 1,475 1,400± 1,100 White. 940 A (1,150B) 690 A (802 B) 750 A (952 B) 850=fc ] , 195 1,165 b210± 1,025± 1,205± 1,110± 1,160± 1,130-:: 1,219 l,165d: , 1,460? 1,100 1 Second Geo!. Survey Pennsylvuniii, Anthracite Region, Rept. 1, 18S3, p. 80. • Perhaps not over 9.50 feet. 860 FLORAL ZONES OF THE POTTSVILLE FORMATION. From the above tuIHe it appears that over 3i><.» feet of transition series has been inchided within the Pottsville formation in some of the measurements pul)lished 1)y the State survey. Among the deductions to be drawn from the table, perhaps the most important are: (1) "Whatever the arbitrary ])ase line employed in the measurements, the formation is found to ]»e thickest in the central portion of the field, i. e., the region including Pottsville and Lincoln. (2) The forma- tion appears to be as thick at 7 or 8 miles from the present southern ])order of the field as in Sharp Mountain. Thus on the Broad Moun- tain, near Altamont colliery No. 2, the diamond drill bore hole can hardly have begun nuich higher than the Twin coal, while the section at Kohlers Gap in Bear Mountain, which was carefulh' described and measured by Rogers, appears to be as thick as all those measured by myself in Sharp Mountain. It seems not improba])le that the great thickness of the formation in the Williamstown tunnel, as platted in columnar-section sheet vii, may be due to error in the identification of the liuck ^lountain bed, or in the computation of the the thickness of the beds. (3) The diminution of the thickness of the Pottsville Ix^tween the type section at Pottsville Gap and the Lansf ord railroad tunnel in Locust ]Mountain is well marked, as appears to be also the rapid increase which is noted in the region of Nesquehoning Gap. I am disposed to believe that in the Panther Creek Basin the B bed is per- haps nearer the level of the Twin coal, or supposed Buck jNlountain bed, than is coal A, which, although distinctly referable at Tamaqua to the Lower Coal Measures, seems to carry a rather less recent flora than that of the Twin coal. Neither is it certain that the A bed at the Nesquehoning Gap is identical with that similarly designated at Tamaqua. (4) Another diminution in the thickness of the section seems to occur along Sharp ^Mountain from Pottsville to Swatara Gap, where the interval from the supposed Twin bed to the top of the red shale is perhaps less than i»50 feet. (5) One of the most interesting facts brought to light in this comparison is the apparently but slight decrease of the formation in Sharp ^Mountain in passing westward along the Dauphin Basin, where at Rattling Run, near the Avestern end of the field, it still retains a thickness of 1,100 feet. This observa- tion is of greater weight because it is based on careful measurements apparently extending only from the iqipermost bed of r(Ml shales. The more marked variations in the thickness of the Pottsville are perhaps due to diff'erences in the horizons taken as (he upper or the lower limits, or to changes in the, thickness of tiic sc\crul t(M-ranes from i)oiiit to point, rather than to the existence of an uiiconforniily at the base of the formation. Even at Tamaqua, where the discrep- ancy between the thickness of the Pottsville, as measured in tht> two gaps, points, perhaps, toward discordance, the difference may )»(• due eitlici- to \iiriatioii. without uiicoiiformitv at the base, or to the a))solute 861 WHITE.] THICKNESS OF POTTSVILLE IN SOUTHERN FIELD. failure of the Twin coal to appear in the Locust Gap section/ Allow- ance for reduction by pressure and crushing should also ]>e nmde m some sections. -o^ff ,,r;iu in The rcktivdv slight diminution in tho thickness ot the PottsMlle m passing along 'the 'Oauphin Basin to Rattling Run as conclttsively hown°i„ th! table given above. '■™ the long radius of the fan-and by the size of the bowlders which are sometimes elun^ered far from the margin of the Keld. In illustration o the latter circumstance, the occurrence of bowlders , or b inches m diam- eter in Head Mountain, described by Rogers,' may be cited. As illustrating the thinning of the beds to the northwest, as well as indicating the radius of the thickened formation o the Southern Anthracite tield, it may be of interest to quote a number of measure- ments of the PottsviUe in other regions, in both the anthracite anc the iUuuinous basins. From a thi.kncss of about 1,200 feet in the type lection, ornearly^h^^ame^^^Ul^^ Anthracite Survey for 1883, p. 80 [202 feet on p. IOC], -I.e. White, Bull. U. S. Geol. Survey No. 05, p. 18o. 3Cieol. Pennsylvania, Vol. II, Pt. I, p. 22. 862 FLORAL ZONES OF THE POTTSVILLE FORMATION. margin of the Southern Anthracite tield. the Pottsville formation decreases to about 850 feet at Shamokin Gap, toward the west end of the Western Middle Anthracite tield, and to 830 feet, more or less, at the ]Mahanoy tunnel at the eastern end of the same field. Here the upper conolomerates often contain pel)bles of the size of an egg, while the lowest beds are interlarded with red shale, as in the Southern Anthra- cite field. A very rapid change is to be observed in the basins of the Eastern Middle field, where the contact with the red shale becomes distinct. Thus in the Silver Brook Basin, on the southern border of that field, the formation is but 500 feet thick, while in the Upper Lehigh, on the north, it is said to l)e not over 200 feet in thicknes.s. The meas- urements of the formation in the Northern Anthracite field varv, the average being about 225 feet. It is undoubtedly much less than this at points, such as the well-known fossil plant and insect locality at Campbells Ledge, near Pittston, where, if Dr. I. C. White is correct in the recognition of the equivalent of the Mauch Chunk formation, the Pottsville, assigned a thickness of but 54 feet by him, can hardly exceed 100 feet at most, as limited according to the standard employed in the bituminous basins. The diiriinition of the formation from 1,100 feet at Rattling Hun, in the Dauphin Basin, to lOO feet in the Broad Top field is perhaps less remarkable than the decrease in passing from the Southern field to Upper Lehigh, which is but 18 miles from Tamaqua and 14 miles from Nesquehoning. Both of the thinner sections may be considered as offshore stations, as compared with the thick sections farther to the southeast. It is, however, difficult to form an estimate of the relative remoteness of any of these points from the original coast of the inte- rior Carboniferous sea. Li the Bernice Basin, Sullivan County, the Pottsville does not appear to exceed 125 feet in thickness, and a similar measurement is reported where the formation touches the New York State line. Throughout most of the ])ituminous l)asins in southei-n and W(>stern Pennsylvania, including the northern margin of the coal field, near the Ohio line, the formation averages about 250 feet, more or less, in thickness. South- west of Broad Top. on the Potomac River, the section is somewhat thicker, and from that point the PottsviUe shows a generally, though not invarialily. increasing thickness until we reach the Kentucky- Virginia border, wiiere it })ro))ably exceeds 2.500 feet. VARIATION IN THE CONSTITUENT TERRANES OF THE FORMATION. It necnls ]>ut a (•onq)aris()n of the carefully measured, detailed colum- nar'sections of diamond-drill bore holes and of tunnels, published in Pts. IV and IV B of the Atlas of the Southern Anthracite Field, to demonstrate not only the varialMlity in the thickness and composition WHITE.] VARIATION IN PUTTSVILLE TERRANES. 863 of the Pottsville .stratii, l)ut uLso the astonishing laek of continuit}' among even eonspicnous and important strata. In fact, I know of no region in the Appalachian trovigh in which the local irregularities of the coal-bearing formations are more marked than in the Southern Anthracite iield. It is not diHicult to account for this irregularity on the hjq^othesis I accept in explanation of the conditions attending the deposition of the Schuylkill-Swatara and Virginia sections. The for- mation of beds of coal under such conditions seems to necessitate the assumption either that there existed, at various times on the surface of the Pottsville terrace or fan, coastal lagoons or protected basins, the sluggish water supply of which was laden for short periods with little else than vegetable matter, or, as appears more probable, that, as the result perhaps of occasional uplifts, large areas lying within bars or shoals were converted during short intervals of quiescent stability into Carboniferous swamps or lagoons in which considerable irregular deposits of plant matter accumulated before the current erosion of the l)arriers or the renewal of the general movement of submergence ter- minated the conditions favorable for coal formation and permitted the invasion of the coarsely deti"itus-laden waters. The interruption of the general subsidence by short periods of elevation and stability, while permitting at once the accumulation of vegetable matter in one region and the seaward extension of the submarine terrace in another during the periods of higher level, accounts also for the readiness with which the conglomeratic sediments, which usually almost directly, when not immediatel}', overlie every Lykens coal, were swept across the carbonaceous deposits on the recurrence of the general downward movement. The varialjilitv in the thickness of the coals, their irregular inter- vals and distribution, as well as the fact that the areas containing the lower Lykens coals are so restricted, compared with the area of the anthracite tields, appear to sustain this hypothesis as explaining both the deposition of the coals and the extent of the formation. As partially illustrating the variation of the several members of the Pottsville formation in the mining district of the Southern Anthracite field, while showing the prevailing intervals between the coals, the following incomplete table is presented, although it is extremely fragmentary and evidently insufficient to serve as the basis of any important generalizations. 864 FLORAL ZONES OF THE POTTSVILLE FORMATION. Tahlf nlioiiiiKj inft'rnih ht'tueen Oie principal Lykens coals in tJte Lincoln-Li/kcns mining region. [The intervals indicated are those between the horizons in whoso colnnins tlie numbers occur.] Locality. Distance from coal named below to ne.xt horizon under which record is placed. Lykens coal No. 1. Ly- kens coal No. 2 or 3. Lykens coal No. 4. Lykens coal No. 5. Lykens coal No. 6. Top of red shale. Total from Buck Mountain coal to red shale. Pottsville Gap... Broad ^Mountain (near Gordon plane) . Swatara Gap Ran sf li Gap (Sfhn yl kill Gonnty) . Colket Lincdin New Lincoln GoodSprintr Kainiia Buck Moun- tain. do 380? ' 170? 160? 480 640 60?' 425? 1.195 do 385 ifc 230 l,025zb 1, 205 ± d.. do do'.' .lo do 103 250 ± 250= 210 910 65 370 320 245 250 120 130 48 47 (+) 322? 980 110 87 140 (+) 70 75 90 75 66? 25 60 270? 100- 50?- 690 1.219 1,460? L 100 Kolderr^ (iaj) Williamstown Gratz . Buck Moun- tain. .do 372 288 Shinj tnnnel (+) Rattlinjr Run I Mill and Hoff s^ h a f t s , near Big Flats. Buck Moun- tain. <1<. 500± 410 200? 160? 100? 45± It is ot" interest, however, to note a few of tlie \ariations. sucii as that in the interval between the Buck Mountain coal and Lykens coal No, 1, which at Colket is 103 feet; at Good Sprino-. -t miles Mest, 210 feet, while at New Lincoln, farther south, l)ut in reality al)()ut 2^ miles from either (xood Spring- or Colket, it is 250 feet. Similarly, the inter\al between the Lykens coals 5 and ♦>, which is but -tT or 4S feet in tlie New Lincoln and Lincoln mines, measures 75 feet at Kalmia, with which direct underground connection is mad<'. wliilc the same thivkness is observed at Williamstown. In passino- from the subject of the variability of the terranes of the Pottsville formation, it should be ob.served tliat tlic succeeding Coal ^Measures also, especially in the Panther Creek r)asin and the regions west of Potts\ilie. show the continuation of conglomerate sedimenta- tion in enormous (jiuuititics. though the formation is generally softer WHITE] PLANTS OF THE POTTSVILLE IN SOUTHERN FIELD. 865 than tho Fottsvilli'. In certain instances cong-lomeratic .sandstones and conglomerates compose about one-third or more of the entire section. As might be expected, this feature, which is well illus- trated in the sections located in the Tremont region' and in the Panther Creek Basin, is not less striking than the astonishing varia- bility in the thickness of the intervals separating the coals of the Productive Coal Measures in the same regions. In this connection it is both interesting' and instructive to make a comparison of the columnar sections published in columnar-section sheets x of Pt. IV, vi of Pt. 11, and iii of Pt. I. of the Atlas of the Southern Anthracite Field. IS^OTES OX OR DESCRIPTIOXS OF SOME OF THE 3IORE CHARACTERISTIC SPECIES OF FOSSIE PLANTS OF THE POTTSVIEEE FORMATION IN THE SOUTHERN ANTHRA- CITE FIEED. It was my original purpose to have the description of the strati- graphy of the Pottsville formation in the Southern Anthracite tield accompanied by full descriptions and illustrations of the fossil plants, which, with the exception of Sp/'/vr/ji.s, rare crustacean fragments, or still rarer cockroach wings, appear to constitute the sole organic remains yet brought to light. When, however, it was found not only that the manuscript and plates were too voluminous for the present form of publication, but also that the subsequent preparation of a complete report covering- the fossil plants of the formation in other portions of the Appalachian province would include the repul)lica- tion of many of the descriptions of the fossils from the Southern Anthracite field, it was determined to confine this report to the description, limitation, and definition of the Pottsville formation as found in the type section and region, and such economic or general g-eologic results as had been reached in the course of the paleontologic and stratigraphic studies in the field, as well as such general or broad correlations as might be proper in a preliminary paleontologic publication. The following pages are devoted to descriptions of some of the more important stratigraphic species of the several zones of the Potts- ville formation or to notes, either relating- to species already known elsewhere or concerning forms closely allied to well-known types. Following is a list of the entire flora. 1 See polumnar-section sheet x. Atlas Southern Aiithnicitc Field, I't. IV B; and columnar-section sheets i and ii, respectively, of Pt. I of the Atlas. 20 GEOL. 11' 2 55 8(>6 FLORAL ZONES OV TUV. POTTSVILLP: FORMATION. LIST OF FOSSIL PLANTS FROM THE POTTSVILLE FORMATION IN THE SOUTHERN ANTHRACITE FIELD. Aiu-iiuite:^ pottsvillensis |). W . Aneimites t^p. Ereiii<)i)teris siibt'k'gaiis 1). \V. Ereniupteris sp. No. 1. Ereinopteris sp. No. 2. Ereiuopteris ditisecta Lx.'.' Ereinoptt'rit: lim-olniana ]). \V. Erein()i)UTis Cheathami Lx. Eremopteris (k'i'ipien.>< (Lx.). Ereiiiopteris Aldrichi D. W. Marioi)teris eremopternidcs I). W. Mariopteris i)i)ttsvill('a 1). W. Mariopteris Phillipsi D. \V. Mari()i)t«'ris Phillijisi var. intcriiKvlia D. W. Mariopteris pygiua-a J). W. Mariopteris nervosa (Bronsrii. I Zeill. var. liiu'olniana D. W. ^Iario])teriH tennesseeaiia I >. W. Marioi)teris tennesseeana var. liirsula D. W. Mariopteris cf. acuta (Broiitrn.) Zeill. Mariopteris sp. Pseudopecopteris (ihtiisildhalSti'rul).) Lx. var. luariopteroides D. W. Pseudopec!)i)teri.s cf. squamosa J.rX. Sphenojjteris iiiid)ratilis D. W. 8i>hen()pteris Lehinamii D. W. S])hen<>pteris Kaercheri I). W. Sphenopteris .siniulans I). W. Sphenopteris asplenioides StcrnI). Sphenopteris sp. Sphenopteris dadt-ana D. \V. Sphenopteris divaricata ((loepj). ) (iein. &(;utl). S])henopteris (lienaidtia) iiiicnu'ariia Lx. var. disseeta ]). W. Sphenopteris Ilarttii Dn. Splieiiopteris sul)pinnatitida l>. \V. Sphenopteris Monahani D. W. Sphenoi)teris (Diplothniema) patentis- siina (P^tt) Sfhinip. Sphenoj)teris (Diplothiiieiiia) fiircata P)ron<;n. Sphenopteris l\o\i Lx. Spheno])teris novaliin'olniana i >. W. Sf)lKno]tteris novaiincohiiaiia var. aule- 'eedens I). \V. Sphenopteris pahiiatiloba i>. W. S]iheiio])teris jiahiiatilolia var. s((iiarrosa 1). W. Sj)iien()pleris Lutheriana D. W. Sphenopteris niixtilis D. W. Sphenopteris pilosa Dn. Zeilleria cf. avoldensis Star. Aloiopteris (Corynepteris) geortriana (Lx.). Oligocarpia crenulata D. AV. Oligocarpia alabamensis Lx. Pecopteris serrnlata Hartt (non Heer). Pecopteris sp. Alethopteris Lacoei J). W. Alethopteris protaquilina 1). A\'. Alethopteris lonchitiea(Schlotli.) llmngn. Alethopteris lonchitica \ar. niultinervis D. W. AJethopteris alata 1>. W. .\lethopteris lincolniana 1>. W. Alethopteris niagnifolia 1>. \\'. Alethopteris grandifolia Newb. Alethopteris (hserepans Dn. Alethopteris coniposita D. W. Alethopteris Serlii (Brongn.) Goepp. Alethopteris coxtoniana D. W. .\lethopteris Evansii Lx. .\ lethopteris Evansii var. grand is I ). W. Alethopteris sp. Callipteridinni alleghaniense D. AV. Callipteridiuni suspectum D. W. Callipteridinni pottsvillense D. AV. Megalopteris phnnosa D. AA'. Megalopteris sp. Neriopteris lanceolata Newb. Nenropteris Pocahontas D. A\'. Neuropteris Pocahontas var. pentias D.AV. Nenropteris Pocahontas var. iiuequalis D. AV. Nenropteris Smitlisii Lx. Nenropteris Aldrichi (Lx.). Nenropteris Elrodi Lx. Nenropteris acntiniontana D. AV. Nenropteris tennesseeana Lx. .AISS. Nenropteris tennifolia (Schloth.) Brongn. var. hnniilis D. AV. Nenropteris sp. 1 Nenropteris aff. heterophyiia Brongn. I Nenropteris ovata Iloffm. Nenropteris hirsutina D. AV. Nenro]>teris Desorii Lx.? Nenropttr-ris fiinbriijta Lx. Neurojjteris gigantea Sternb. Nenro])teris hniata D. W. PLANTS OF THE POTTSVILLE IX SOUTHERN FIELD, 867 At-teroc-alamitef^ scrobiculatus (Schloth. ) Zeill. Calamites Roemeri Goepp. Calamites Haueti Stur. Calamites approximatus Schloth. Asterophyllites parvulus Dn. Asterophj^llites arkaiisanus I). W. Asterophyllites pennsylvanicus D. "W. Asterophyllites ef. rigidus (Stb.) Brongn. Annnlaria platiradiata Lx. MSS.? Annularia laxa Dn. Annnlaria acicularis Dn. Annularia cuspidata Lx. Annularia latifolia (Dn.) Kidst. Calaniostachys cf. lanceolata Lx. ? Calamostachys Knowltoniana D. W. Palaeostachya alabamensis D. W. Macrostachya sp. Volkmannia orassa Lx. Sphenophylluni tenerrinunu Ett. var. elongatum D. W. Sphenophylluni bifureatuni Lx. Sphenophylluni cuneifoliuni (Stl). ) Zeill. Sphenophylluni tenue D. W. Bowniannites ? .'^p. Lepidodendron alabamense D. W. Lepidodendron Veltheimii Sternb. Lepidodendron clypeatum Lx. Lepidophloios acutoniontanus D. W. Lepidophloios sp. Leiiidostrobus pennsylvanicus D. ^V. Lejiidostrobus cf. ornatus L. & H. Lepidophylluni quinniniontanuin D. W. Lepidophyllum campbellianum Lx. Lepidophylluni lanceolatum li. & H. var. virginianum D. W. Lepidophyllum linearifolium Lx.? Lejiidocystis fraxiniformis Lx. Triletes sp. Bothrodendron arbore.^cens (Lx.). Sigillaria ichtyolepis (Presl) Corda. Sigillaria kalmiana D. W. Sigillaria lincolniana D. W. Sigillaria cf. laevigata Brongn. Sigillaria sp. Sigillariostrobus ? incertus D. W. Stigmaria verrucosa (Mart.) S. A. Mill. Stigraariopsis Harveyi Lx. MSS. Cordaites Robbii Dn. Cordaites Phillipsi D. W. Cordaites angustifolius Dn. Cordaites grandifolius Lx. Artisia irregularis D. W. Cordaianthus spicatus Lx. Cardiocarpon bicuspidatum (Stb.) Newb. Cardiocarpon liicuspidatum var. ohioense D. W. Cardiocarpon Cuyahogjfi D. W. Cardiocarpon minus Newb. Cardiocarpon late-alatum Lx. Cardiocarpon disculum D. W. Cardiocarpon orbiculare Ett. Cardiocarpon cornutum Dn. Cardiocarpon elongatum Newb. Cardiocarpon elongatum var. interme- dium D. W. Cardiocarpon annulatum Newb. Cardiocarpon Phillipsi D. W. Cardiocarpon Wilcoxi D. W. Cardiocarpon Girtyi D. W. Cardiocarpon obliquum Dn. Trigonocarpum Noeggerathi (Steriil).) Brongn. Trigonocarpum ampuUaeforme Lx. Trigonocarpum Helenje D. W. Trigonocarpum Dawsonianum D. "\V. Trigonocarpum ornatum Newb. Rhabdocarpos (Pachytesta) speciosus D. W. Rhabdocarpos (Pachytesta) ^^'alc(lttianus D. W. Whittleseya Campbelli D. W. Whittleseya Lescuriana D. W. Whittleseya jnicrophylla Lx. Whittleseya elegans Newb. var. minor D. W. Carpolithes fragarioides Newb. Carpolithes orizfeformis Lx. MSS. Carpolithes sp. Carpolithes transsectus Lx. Sporangites sp. Fayolia sp. 868 FLORAL ZONES OF THE POTTSVILLE FORMATION. NOTES ON CERTAIN OF THE PREVIOUSLY KNOWN SPECIES, AND DESCRIPTIONS OF THE STRATIGRAPHICALLY MOST IMPORTANT FORMS. AXEIMITES roTTSVILLEN'SIS Sp. HOV. PI. CXC, Figs. 1, 2. Fronds lax, bi- or tri- ( () pinnate; pinna? slender, .slightl\' tlexuose or sul)geiiic'ulate, loose, slighth^ irregular, with very slender, sulcate, lineate, narrowly bordered {() rachis. Pinnules alternate, distant, open near the base, oblique above, poh'- morphous, usually asymmetrically ovate or rhomboidal-ovate, some- times obovate, obtuse, T to 18 mm. long, 3 to 11 mm. wide, the lower ones neuropteroid or even triangulo-semicircular, the terminal pin- nules cuneate-obovate, generally broad and truncate-rounded, the lowermost sometimes dissected to the base to form 3'oung pinnje of three pinnules, of which the middle one is similar to the ordinary ter- minal ones, the lateral being rhomboidal, all the pinnules being con- stricted to a verj^ narrow point of attachment, with straight proximal margins, and very hnely lineate lamina between the nerves. Nervation a little coarse, radiating flabellately from a single basal fascicle, forking three to live times while passing straight to the border, and counting about 25 to the centimeter along the distal margin. The most common form of pinnule met in the fragments of this polymorphous species is the rhomboidal type, such as is shown in PI. CXC. Fig. 2. which represents the normal lateral pinnule. In this illustration, which will })e supplemented by others in the larger work, th(> characteristic rhomboidal shape is very imperfectly shown. The sides, especially the superior proximal and the inferior distal margins, are in general nearly parallel. The distal angle is nearly always well markedexcept in the terminal pinnules, which are cuneate and roundly truncate. An example of the last is seen in PI. CXC, Fig. 1. Of the species heretofore published, that to which our species is most similar and most nearh' related is Aneimites adiantoides (L. and H.) Ett. The extremely close affinity of these two forms may l)e H'adilv noted )>v a comparison of the original tigure of S.phpferh adiantouh's^^ or that (Uvscribed })y Sauveur'~ as SpheiwjJterls o7>f(/siJoba, with the specinicns in hand. Tlic Pottsville })laMt seems to be distinguished from the fern from the elarrow colliery l>y its more rhomboidal and angular lateral pin- nules, the less dilated or capitate terminals, and, to judge from the ligure in the Fossil Flora, by the rather closer, more rigid nerva- tion. One of the specimens from llie Culm, figured bv Dr. Stur as 1 Lin are relatively short and compact. This feature as well as the form of the lobes, which in the inferior t)asal pinnules are sometimes palmately spread, as in Eremopteris missourlensis Lx., is one of the more prominent characters b}^ which the plant differs from Eremop- tei'is arteniisicefolia (Brongn), to which, as identified in our American collections, E. lineolniaua is closely related, or possibly ancestral. Although from the habit and mode of division of the tertiary pinnae the fern is apparently referable to the Sphenopteroid division of the genus Eremopteris^ the basal ramification, so far as it can be deter- mined from the specimens ))efore me, imparts a suspicion that the frond of this species, like that from Missouri, maj'' divide in the same manner as the fronds of Diplothmena. As stated in the discussion of the ferns from the Lower Coal Measures of Missouri, I believe l)oth species to have been derived from the Archseopteroid stock through the genus TrlphyUopteris. It seems far from improbable that Erem- opteris^ Rhacopteris^ Anehiiites^ Asplenltes^ Sphcnopjfer/'dium dissect am (Goepp.) Schimp., and S'plienop)tefris excelsa L. and H. are members of an early comprehensive group of Paleozoic ferns. This species has not yet l)een found above the top of the Pottsville series, although it appears to occupy a period near the close of that formation, and to be most closely related to an undescribed form in the lower pcn'tion of tht> Kanawha series in West Virginia. The type specimens are from the New Lincoln mine, where its association in the matrix with Xexiropteris Elrodi Lx. renders it nearly certain that it comes from Lykens coal No. 2 or No. 3, probably No. 2. AVith it is also found the E. Lehmanni. E. lincolniana is also present from the Lincoln mine, where it is associated with the same species as at New Lincoln, In the Pottsville Gap the species occurs 550 feet Ixdow the Twin coal; i. c. near the supposed horizon of Lykens coal No. 3. Eremopteris Cheathami Lx. This plant, described by Lesquereux from Rockwood and Tracy City, TeniH^ssee, is one of the most clearly marked and well differenti- ated fern species of the entire formation. Its nuxst prominent features WHITE.] NOTES ON CHARACTERISTIC SPECIES. 871 are the relatively short, remote, ultimate pinna?, the minutely rugose- striate limb, and the broadly cuneate, compact pinnules and lobes, cut on the obli({iie distal margins into short, irregular, blunt, claw-like, erect teeth. Unfortunately the presence of the latter, concealed for the most part })y their backward curvature in the matrix of the type specimens, is almost wholly ignored in the tig-ures, accompanied by details, published in the Coal Flora. ^ The pinnules of the species vary conspicuously in size, the larg-est seen, in terminal fragments, being nearh^ one-fourth larger than those tigured, while the smallest fragment yet observed is that illustrated in pi. civ., fig. 8. of the Coal Flora. The specimens from 550 feet below the Twin coal in the Pottsville Gap are specilically indistinguishable from the typical Tennessee form, though the northern representatives of the species seem more delicate and less coriaceous than the southern originals. In our Paleozoic plant collections Eremopteris Cheathaml has some- times been confounded with E. decljnens on the one hand and Tr/j>Jn//- loptei'h Le.scuriana (Meek) Schimp. on the other hand. The species described by Meek from the Pocono or Vespertine series, which, judged ])y its flora, is nearl}- contemporaneous with and certainly not later than the Calcif erous sandstones of Scotland, is easily distinguished b}' its clearly lanceolate pinnules or lobes, which are often slightly fasciculate in the impression, the Archseopteroid nervation, and the marg'ns not crenulate or sinuate. Besides its occurrence at the Potts- ville Gap this species is also found at the horizon of Lvkens coal No. 3 at the Lincoln mine. Eremopti:ris decipieks (Lx.) The form which 1 have described as Ereinopterlx decrpien><{Liy..) qow. stitutes, with its several variations near the top of the Pottsville series in northern Tennessee, in southern West Virginia, and in Arkansas, one of the most interesting types of our upper Pottsville flora, com- bining as it does, in its general aspect, some of the characters of the broad-lobed species of Ereniopterk with other details common in cer- tain forms of P.^evdopecoptcrw. In the general habit of the lower or pinuatifld pinnules of the frond it is distinctly a member of the group represented b}^ Eremopteris Chexdharm\jx. The flabellate-cuneate mode of division of the pinnatifld ovate-triangular pinnules or young pinnte, as well as the emarginate-sublobate upper borders of the lobes, l)ind the plant to the above-named group, although the nervation, which is also consonant with the latter, is seen to develop thePseudopecopteroid tvpe in the more broadl}' dilated, trifoliate forms. Among the hith- erto-published American types our species is probably most nearly related to the plants figured o\ identified as Pseudopecopteris macilenta Vol. III.pl. <-iv., (i^s. -l-i, J). 770. 872 FLORAL ZONES OF THE POTTS VILLE FORMATION. [Lx. ]. fi'om Olio fonii of which the (liflVrentiatioii is iiardly more than varietal in importaiico. Ki'i'iiioptrriH declp'teih'i differs from Krrinopfi'i'ix CJicKtluoiii^ typioally represented in abundant material from Tracy City, Tennessee, by the generally more distant and more distinctly cuneate lobes, which are always crenulate-denticulate along the distal margin, b}' the rather straighter nerves, and ))y the generally somewhat larger lobes of the latter. ?J. Chenthaml^ which seems also to be present in the anthracite region, occupies there, as is usually the case in other regions, a some- what lower stage than the Pseudopecopteroid group. Tlu> species occurs at both the Lincoln mines, at the North Brookside slope, n(>ar Good Spring, and at the prospect drift, near the mouth of the upper Eur(>ka tunnel, as well as at several horizons in the Upper Lykens (li\isioii at the Pottsville Gap. Makiopteris eremopteroides sp. nov. PI. GLXXXIX, Figs. 1. 2, 3, 3a. Frond (iua(lripartite(^). polypinnate, very large, rather dense; primary ])innie large, very long, of unknown form, with lineate rachis attaining a diameter of 1.5 cm. or more; secondary {'.) pinna? altei-nate. open, often at a right angle to the rachis, close, sometimes o\-erla))})ing nearly one-third their width, linear, or linear-lanceolate, tai)(Ming to an acute apex, with rather slender, ventrall}^ concave, dor- sally terete, very tinely lineate, slightly flexuose or flexuose-geniculate rachis; penultiiuate pinna^ alternate, open nearly if not quite at a right angle to the rachis, close, usually touching, or slightly overlap- ping, ])ut sometim(>s. especially in the upper part of the pinna, a little distant, lanceolate or linear-lanceolate, acute, hardly constricted at the )>ase. slightly Hexuose-subgeniculate, the lower inferior pinna not .«pe- <-ially heteromorphous: ultimate pinna> alternate, or sub-opposit(\ <)polate, somewhat acute, genei'ally slightly subfalcat(>. the apices inclined upward, the rachis re )und-sulcate, venti'a'.ly terete, dorsally mimitely lineate. and bordiM'ed by a nari'ow wing (lecurring from the limb of the ])inmiles. I'inniiics alternate, very ot)li<|ue or neai'ly erect in the younger ])innie. distinct to near the \\\)v\ of the largei' pinna', close, generally ovate or rhomboidal. rarely ol)ovate, obtuse or (>])tusely roundiMl. the upper ones connat<' foi" a little distance, the terminal ovate or ovate- triangular. obtu>e. obscui'ely sublobate. the lower ones attached by "\erv l>road. ol)li(|Ue. often pi'oduced bases, oidy the lowest lobed ])innnles l)econiiiig pinnatilid. the\' beini;' naiTowly constricted at WHITE] NOTES ON CHARACTERISTIC SPECIES. 873 the bases, all tlu> ])innul('s showintr at an early stage a marked ten- dency to division in two to live obtuse, rounded lobes, which, appear- ing- at lirst as one or two rounded teeth a little above the middle, are gradually cut one-half way to the rachis, sometimes, especially in the somewhat heteromorphous basal pinnules, appearing slighly obovate as the pinnule ])ecomes pinnatifid in its development into a pinna, though generally the ovate or ovate-rhomboidal form, with confluent or hardly constricted bases, is preserved to an advanced stage; lam- ina of the pinnules not thick, very slightly depressed over the pri- mary nerve in the pinnatitid pinnules, very faintly rugose, especially on the dorsal, minutely striated, surface, and rolled rather strongly backward at the margins so as frequently to make the pinnules or lobes appear more acute thati they really are. Nervation of moderate strength, distinct and very slightly depressed on the ventral surface, very close and in relief on the dorsal surface of the pinnule: primary nerve originating at a nai"row angle, nearl}^ opposite the proximal basal sinus of the pinnule, forking at an open angle near its point of origin, and curving strongly outward in the base of the pinnule, then forking pinnately and a little widely to sup- ply a secondary nerA^e for each lobe, the secondary nerves forking one to four times, at a moderate angle, in passing, a little distant, in a gentle, slight curve to the distal border. The relation of this graceful and beautiful fern to the genus 2£arloj)- ter'ix appears to be shown by the developmetit of the frond as well as by the general details of the pinnte. Nevertheless, the aspect of the pinnatifid portions of the frond, particularly when seen in small frag- ments, showing the spreading, lobed, relatively unconstricted.' extended pinnules, such as that shown in PL CLXXXIX, Fig. 8, as well as the nervation, is often so similar to the corresponding parts in some of the smaller, more compact forms of Erc'inopterh as to call in question its generic attitude to the latter. The examination of a large series of speci- mens shows the species in hand to be, however, one of easily recognized in(li^'iduality. The very large size of the plant is evidenced by portions of its rachis over 3 cm. in diameter, fragments of rachis, apparently representing one of the larger of the four divisions of the frond, being al)out 1.5 cm. in diameter when compressed. The rachises of th(» lateral pinme are more slender than in most species of this genus, and are slightly flexuose, in correspondence with the pinnation, even where the axes have attained considerable development. The salient features Avhich are to be observed at the first glance at small fragments of the fern ai-e a relatively close pinnation, with a tendency to curve upward in the smaller j)irma% the closeness or connatcMiess of the obtuse pimudes, and the marked tendency to loba- tion. which shows even in the small and half-developed pinnules, the lobes ap])earing as one or two or three inconspicuous shoulders, or 874 FLORAL ZONES OF THE POTTSVILLE FORMATION". broad, obtuse, or rounded teeth on the sides of the limb. When further developed this sublobation, which may be seen in specimens from nearly every part of the frond, becomes conspicuous, givino^ the pinnule in its pinnatitid stau"e a Sphenopteroid or Eremopteroid phase. Probably the only species of the genus in our Hora with which Mariopteris eremopteroides is liable to be confused is M. pottsvillea. But althou.o-h there is a resemblance in portions of the fronds of the two species, sometimes appearing close on a casual glance, it is randy difficult to distinguish the two forms, even in small fragments bearing simple piiuudes. from the upper part of the pemdtimate pinna'. The pinnules of J/, erciiujpteroidv-s are not so constricted at the base, not so triangular or dilated just above the point of attachment, and, as may almost invariably be noted, they are more or less distinctly lobate or sublobate. even in a younger stage, in which they are still attached ])y the whole base or even slightly connate. In general, the short pinnae of the latter species are more dilated toward the })ase, both the piniiiv and the pinnules being usually smaller, the latter being more frequently connate, as well as lobate and alate. Ver}^ often, too, the pinnules are set out from the rachis by a slight elongation of the basal portion or attachment so as to suggest a very short, broad pedicel, sometimes nearly equaling the pinnule in width. The nervation of M. pottsvillea is somewhat coarser and noticeal)ly more distant and ari-hed. The species is abundant at all mines in the horizon of the roof shales of Lykens coal No. 5. Possibly it is present also in the roof of Lykens coal No. (\. MaRIOITKKIS I'OTTSVILLEA Sp. UOV. PI. CXC, Figs. 3, 3a, 4, 4a, 5, 6. Fronds (juadri- or poly- {() pinnate, robust, not very dense; penulti- mate pinna,' alternate, open, the lower at a right angle to the rachis, the upper slightly obliciue. rathiM* distant, lanceolate, or linear-lanceo- late, slightly contracted at the base; rachis somewhat flexuose, coarsely lineate in the major divisions, more finely and irregularly lineate in the smaller divisions, while in the penultimate and ultimate pinna' they are slender, slighth' ilexuose-geniculate, ventrally sulcate, dorsally round, and broadened by narrow decurrent wings of the lamina; ultimate j/nuiie alternate oi- subopposite, open at a right angle or slightly oblic^ue, distant, usually one-half their Avidth or moi-e apart, lanceolate or linear-lanceolate, clearly constricted at tiie base, with a nai'row decui'iing border. Piniudes alternate or siihalternate. usually distant, ()l)li<}ue. broadly ovate, or ovate-triangular, asynunetrical, obtuse, or obtusely rounded, venti'ally arched, distinctly constricted at the broad base, which is marked in all the well-developed examples by an inferior rounded sinus, the uppermost pinnules becoming confluent, more olilique and WHITE.] NOTES ON CHARACTERISTIC SPECIES. 875 rounded, blending- into the ruther long' terminal, which often has its obscurely sublobate or sinuate margins rolled back so as to make it appear acute or even muricate; lamina of the pinnules not ver}^ thick, dull, becoming" decurrent in a very narrow wing along the rachis. Nervation rather strong, distinct; primary nerve originating low at an acute angle, arching outward, not rigid, forking repeatedly at a moderately wide angle; nervils a little distant, forking one to three times and curving more or less in passing, with diminishing- dis- tinctness, to the margin. The examination' of the collections from the geological sections of the Pottsville series, from Pottsville in Penns3dvania to the southern extremity of the Appalachian coal field in Alabama, shows the species described above, with its minor variations, to be one of the most ubiq- uitous as well as the earliest American representatives of the genus Marioptei'is. Under the name Pmudopecopteris murieata (Schloth.) Lx., it has long been known in the collections from the whetstone beds of Indiana, the Dade coal (Lookout sandstone of Hayes) in Georgia and various points in Alabama. Recent studies in the field show it to be specially prevalent in the middle division of the Pottsville series, to which I have given the name Horsepen group. It is more particu- larly characteristic of the upper part of this group. The ordinary ultimate divisions of Mariopterls potUvillea^ such as are illustrated in PI. CXC, Fig. -I, are clearly characterized (1) by the comparative remoteness of the pinna? and pinnules; (2) by the form of the pinnules, which are broadly ovate, obtuse or obtuseh' rounded at the apex, dilated above the base, and plainly constricted at the base, and arched ventrally ; (3) by the large size of the pinnides, which is greater than any of the earlier round-ovate, inflated-pinnuled forms yet found in what may for convenience be called the Mariopteris nmricata group; and, finally, by (4) the rather coarse, distinct, curved, rather close nervation, which approaches near to that of Mariopterk tennesseeaiia^ a form intermediate between J/, pottsvillea and M. murieata or M. nervosa. The present status of Filiciten muricatus Schlotheim,' or of the types of Brongniart's Pecopteris murieata^ seems slightly ambiguous, since the /"*. murieata has latterly been united by Zeiller^ and other European palco])otanists with ^Larwpterh nervom (Brongn.) ZeilL. a form quite distinct from the American material hitherto recognized as Schlotheim's species. It is true that the difference between the forms originally described under the two names is much less than ^\■e have been led to believe from the American interpretations of the illustrations and figures. ' Petrefactenk., p. 409; Flora d. Vorwelt, pi. xii, figs. 21, 23. "Hist. vi'g. fo.s.s.. Vol. I, p. :i52, pi. xcv, fig. 34; pi. xcvii, fig. 1. ^ Fl. Foss. bassia houill. Valenciennes, p. 173. 87<') FLOKAL ZONES OF THE POTTSVILLE EORMATION. Still. Nvhilo there i.s sc-airoh' room for doubt that J/. ;?t'/'r6>6'rt was either derived from J/! inuriatta or a common, slightly earlier stock, the analogies of the vertical distribution of the American species of Mari- optii'ls lead naturally to the expectation that the latter type will be found to occur consideral)ly lower in the stratigraphic series of Europe and to disappinir much earlier than the former, although through a portion of the Coal Measures they may have existed side by side. In the American sections the form designated in this report 21. tennesseea7icu which is possibly nearest to the plant figured as PecopteriH niurieata by Brongniart. predominates at the base of the Sewanee group and hardly survives in the normal type to mingle with the small, delicate, thin-nerved variety which appears, in the American Carboniferous, to l)e the earliest representative of 21. nervoi^a., occur- ring in the uppermost portion of the Pottsville series. The relations of 2Iar'iopterU miiricata and the type designated in our Amei-ican literature 21. nervom have been specially discussed in my remarks on the forms occurring in the McAlester, Indian Ter- ritory, coal rield.^ The characters enumerated above readih' distinguish 2Lai'iopferi8 potiHnlha from those European forms known as 2L. murkata and 21. nervosa. The form typically described in the American literature as I^eudopecopteru nerrosa (Brongn.) Lx. has larger, broad, triangular, acute, closer, unconstricted pinnules, with much stronger, more dis- tant, straighter nerves. The species described in manuscript by Dr. Newberry as Ve<-optlka sp. nov. PI. CXCII, Fig. 2-6. Frond small, compact; rachis rtdatively strong, lineate. deeply depressed, ventrally canaliculate; pemdtimate pinna^ilternate. nearly at a right angle to the rachis. close, touching oi' overlapping, lanceolate or linear-lanceolate, acute or acuminate; ultimate pinnte alternate, Nineteenth Ann. \W\>\. V. S. Gcol. Snrvey. Pt. HI, i>. 175. WHITE.] MOTES ON CHARACTERISTIC SPECIES. «^^ verv compart, v«-v open, usually touehin.; or .^lightly overlapping, teneeolate or oblong lanceolate, acute or sometimes obtusely acute, ^T- Imit;:?:^ sman. 1.25 to . nun. long. 1 to « nuu. wid.^ ^Heniat. usuallv .-ontiguous or slightly overlapping, crowded my Igl y inflated, generally ovate, the lowest reniform-ovate, s^tb dsbUb api.-ulate. dilated near the base, conspicuously const. U'ted at the 'ft ide of the rather broad attachment, those ot the m.ddle^ix. - ions dilated-ovate or ovate-triaugular. apiculate or "huse^^the ter- minal usuallv short and obtuse or apiculate, or, at the end ot the n Itiniate pinna., slightly sinuate-margined, acute or mucronalx., ai 1 ick. very mucS inflated or arched and smooth ven.rally. the Inlrrius curving -strongly backward, and decurnng in a narrow wing '"'Ner^tior l^her coarse, the nervils concealed on the ventral su. fac'e Init somewhat distinct on the concave dorsal surface of the pinn'ules. primarv nerve strong, originating at a narrow angle and si vplv marked in the largest pinnules by a vanishing furrow on the , d urfaceof the lamina; nervils originating at a - her nariw m those in the lower part of the pinnule arching near the primaiy rvc" and passing, straight or curved, relatively close together, the l^vcVones flrking^once. or rarely twice, the upper nervUs often simple. This the .smallest form of &<^ Man.^derh group known to me, is un ,, lu". ot onlv for the minuteness of its pinnules, but tor the degree : '.':ir inflation and for the crowded arriuigementot^he ,jii>nu and pimne. The fragments represented ,n PI. CXCII. Figs. - <•. aie of the average form and size, such examples being abundantly dis- persed on some of the shale slabs from the Lincoln min.-. ' Both 2[an.,,f.rU„..a and M. P^f'P^^^'^^f^f;';^^^ forms of Jf,„-;ol>t..n. that is almost exclusively conhned to the bewanee ■ppe, Lvkiis division of the Pottsville series. The smallest repre- nta ves are seldom found outside of the uppermost beds of he ■^ vlnee division. The fern from the Tremont region, which l.s hai dly m::Xau varietally diflerent from '^^^'^!^'"'f'^!:'^^^^ «t lemon-s coal mine, in the "coal bearing shale ot \\ ash ngton dm t Vrkln.sas. is evidently closely related to that described ,n ~;ipt bv Dr Newberry as /',«,*-« !,rff«M from the Sharon ;r^ , theastern Ohio. Both plants, each of which is very abun- . t , „ i, ,,!', loi-alitics. are very rarely found at the same locality, ■ • 1/ I, iu..- c.enerallv confined to lower beds in the Sewanec zone. Thf / , i rela ."s "( the plant from New Lincoln and that from thJsh.u Old are corrobonitively indicated in the Pottsville Ba^n IF bv he presence of the .Vrkansas form mentioned above in the ::t\l u'^N:: Lincoln. TI,.. latter is ;H.^tin."f cd f^m the no. ,„„| .1/. ;,,,,,„.,,„ by the rather larger, thinner. I.'ss ,nfl.it«l. 878 FLORAL ZONES OF THE I'OTTSVILLE P^ORMATION. strong-ly apiculuti' })iniiul('s. the lUTVution beiiiii' clcarrr on the ventral SUl'fiU-O. This singular littlo species is nearly always found in a])undance asso- ciated with Neuropteria Elrodi\j.^ AletJiopterls Lacoei^ and Sphenophyl- hiiii tcnernmuni Ett. var. elongotion in the roof shales of Lykens coal No. 2. It has been collected at the lower Eureka drift, the old Lincoln mine, the New Lincoln niin(\ and at the corresponding horizon in th(^ ty])e section at Pottsville. jVIarioptekis tennesseeana sp. nov. The fossils which I shall eventually describe in full as ^fnriopterh iciincHHeeana comprise the unpublished Tennessee material included by Lesquereux in PseudojMCopteris dimorpha. The comparison of the specimens from the horizon of the Sewanee coal at Rockwood, Tennessee, and from the zone of Lykens coals Nos. 2 and 3 in the Southern Anthracite field, with the specimens from the higher Coal Measures at Mount Hope, Rhode Island, and Oliphant, Pennsylvania, which constitute the originals of the species, shows the former to l)e undoubtedly specifically distinct. They are easily recog- nized by the open, a little distant, constricted pinna.% and especially by the obtuse or rounded piiuiules, generally ovate-triangular in form, distinctly oblique, never constricted at the base except in the largest, which are ])ecoming pinnatifid; and, though separated almost to the base in the lower portion of the pinnte, they are seen to be more and more broadly confluent in passing upward, blending in the obtusely subl()})ate, usually rather l)lunt terminal portion of the piimse. The Tennessee species is further distinguished by the not vei-y strong nerva- tion, which is usually indistinct beneath the rathei- thick epidermis. This species occurs in its normal form or as a variety in the Sewanee zone at the Lincoln mine and in the Pottsville Gap. SPHENOPTEKIS KaERCHERI S]). noV. Tnder this name I have descritu'd a fern which, in pinnation, size, and general form and ai"i"angement of the pinnules vcmv cl()s(^ly i-(\s(Mn- bles Ereinopter'lx iiil<-r(ij)/ii/ll(i i^i Lesijuereux, from IkiIs pi'esumal)iy in the Sewanee zone at the ll(>lena mines in Alabama. The saliiMit features of this species are the slender ])innn concealed by the interneural striationof the somewhat inllated limb. The plant is found at both the Pottsville (Jap and at the Ne\N Lincoln mine, where it is tissociated with Xna-optrris lumdl^ Sj)/,( iiojdii/l/inn trnrrr'niniiii and Kr< uio/ffrr/s Jiucol nhirid^ species indicative of the hori- zon of the roof of Lvkens cond No. '_!. WHiTE.J NOTES ON CHARACTERISTIC SPECIES. 879 Sphknopteris asplenioides Sternb. It is much to be ivgretted that a consistent observance of the law of priority in nomenchiture appears to necessitate the use of Sj)/ienoj->feris asj)lenioides Sternberg in place of the more familiar name SpTienopteris ILvn'uujliauxi Brongn., under which the former name is inscribed by most authors as a S3'non3aii. Although the species seems, in the Southern Anthracite iield, to lie very rare in the zone of Lykens coals Nos, 2 and 3, its more common occurrence being in the roof shales of Lykens coal No, 4, in the Lower Lykens division, it has generally a wide range in the thick sections of the Potts ville in the Southern Appalachian region. In the Clark for- matitm, below which it does not yet seem to have been found, the fern is represented by a form with small, compact, round-lobed pinnules and verv narrow pinnte, close to if not identical with >henp>teris aspleni- oides is probably referable to the genus Renaultia. As such it may be regarded as generically identical with Sp)henopteris inlerocarpa liX., which it resembles in its punctate rachis and the mode of the develop- ment of its pinnules. In the Southern Anthracite field this species is found chieflv in the horizon of th(> roof shales of Lykens coal No. 4, at East Brookside, and the Lincobi collieries. Examples of a very small form are present in the roof shales of Lykens coal No. 5 at Williamstown and Big Lick, while the normal form is present in the Pottsville Gap. Sphenopteris dadeana sp. nov. The specimens which will eventually be described as Sphenopteru dadeana comprise several of the types which were included by Les- quereux iiiHler the wAme Sphenopdei'ls Gravenhorstii \?iv. ft Brongn. They dillVi- From the examples figured under the above name' by the 1 Coal Flora, Vol. Ill, pi. ci, figs. 1, 1% V>, p. 763. 880 FLORAL ZONES OF THP: POTTSVILLE FORMATION. punc-tat(^ rac'his. l»y the hroader-lohod, shorter pinnules, the texture of which is more delicate or membranous, and by the relatively simple nervation, the nerves forking- more distantly at a narrower angle and curving- upward so as often to become nearl}' parallel in the lobe. The species is quite distinct from the Spiicnopterk fmg'dls Sternb., which is cited b}' Brongniart' as a synonym of Splieixopteris Graven- litn-sfH. This plant, which occurs at TlO feet below the Twin coal in the gap at Pottsville, appears to be characteristic of, though of rare occurrence in, the Mtrriopterin pottsmllea zone in the Southern Appa- luchiaii region. SrilKNOl'TKUlS I)1\'AKI('ATA ((jOCpp.) (xcin. & Gutl). This species probably bears the closest relation to Sph type locality. St. John. New Brunswick. SlMIENOl'TERIS I'ATENTISSIIMA (Ett. ) Scllini}). PI. CLXXXVTIl. Fig. 1. Primary pinna* pi-obal)ly arranged ]/nuiateiy along an axis; prin- cipal divisions bipinnate or tripiiuiatitid. ovate-triangular or trian- gular-acute, inetiuilateral. lax, with relatively slender, more or less llexuose. naiTowly alate rachial axis, wdiich is lineate, narrowly suieate Ncntraily, sutx-arinate dorsally; ultimate pinna* alternate, distant, open, often at a right angle, llexuose. linear or linear-lanceolate. a(-ute, ()!• somewhiit obtuse. Pinmdes alterntite. usually distant, \cry ()[)eii. often nearly at a I Hist. vc'g. foss., p. 191. '^'"'™J NOTES ON CHARACTERISTIC SPECIES. 881 right ano-le, ovate, round-ovate, or frequently more or less narro\vly triangular, elongate and acute, generally briefly subpetiolate, cut alter- nately to near the base or midrib into one to five close or distant, more or less divergent, cuneate or rhom])oidal divisions, which in turn are once or twice deeply or laciniately incised in narrow, simple, bifid or trifid divergent lobes, each simple linear lobe or tooth having its mar- gins parallel or but slightly converging upward to the very narrow, obtusely rounded apex; lamina not very thick, finely longitudinally lineate, apparently by rows of scaly epidermal cells parallel to the nerves, which are often partially obscured. Nervation usually visible and of moderate strength; primary nerve curving strongly outward from a very acute-angled, decurrent origin, forking low, the divisions forking repeatedly to furnish a single nervil for each lobule or tooth. Representatives of this interesting species are not rare in the lower portions of the very thick sections of the Pottsville ser-es in the Virginia, Tennessee, and the Alabama regions, as well as in the Southern Anthracite field of Pennsylvania. While, however, the specimens from some of the localities in the Southern Appalachian coal fields are typical of the form delineated by Ettingshausen, the form described above from the collection before me appears to differ slightly from the Old World types ^ by the generally slightly more flexuose pinnse, a little g-reater coherence of the lobes, and a rather less marked tendency of the latter to curve outward. In the second particular they are extremely close to the fragments illustrated by Stur.^ As may be noted in the fragments illustrated in PI. CLXXXVIII, Fig. 1, considerable difference exists in the form and elongation of the pinnules in different portions of the frond. The essential characters of the species are the lax habit, the distant, large, very open, and short pedicellate pinnules, the elongated and loose development, with deep, open sinuses, of the subdivisions, and the linear, very blunt or round-pointed lobules which are hardly con- tracted ])elow the middle. The somewhat irregular lineation seen in the lamina of the Pennsylvania specimens is clearly visible with a weak lens. Frequently the apices are partly l)uried in the matrix, or the margin is a little re volute, so as to give the lobules a sharp profile on the rock, ])ut when carefully worked out the tip is found to be rounded. The large pinnules seen in PI. CLXXXVIII, Fig. 1, are comparable to figs. 7 and S, pi. ix of the first part of the Culm-Flora. Spliniopterk furatta Brongn., a species whose pinnules resemble those of 8. patentissima^ is distinguished from the latter by the more rigid pinnae, the closer and more compact pinnae and pinnules, which ' Ettingshausen, Foss. Fl., Mahriseh-Sfhlesischcn Dachschiefers, p. 26, pi. vii, flg. 4, text-fig. 13, 2Culm-Flora, I ; Die Culm-Flora d. .Miihri.stfh-Sehlesi.sehen Dachschiefers, p. 36, pi. ix, figs. 1-9. ii(J (JEOL, PT 2 '){') 882. FLORAL ZONES OF THE POTTSVILLE FORMATION. are sessile, often less deepl^v dissected; and the relatively shorter, less diverovnt. and frequently slightly constricted lobules of the former species. In Sphenopteris Royi the lobules are more oblique, more broadly coherent, and acute. Finally Sphenopteris patentissima is in general more characteristic of the Culm or Carboniferous limestone of the Old World or of the lower Potts vi He in the New, while S. furcata^ its probable descendant, is later in its appearance, passing from the upper Pottsville into the Lower Coal Measures. ^plienojjterls j^tenti'^shna is common in the roof shales of L3'kens coal No. 5, and more especially of Lykens coal No. 4, of which it is largely characteristic, at the Brookside and Lincoln mines, as well as at the mines in the Lower L3'kens division, on Broad ^Mountain. It is also present in the zone of Lykens coal No. 4 at the Pottsville Gap. Sphenopteris (Diplothmema) furcata Brongn. The examination of the American material belonging to the group represented l)y Sjy/ie/wpfen'}^ furcata shows an interesting series of slight modifications. The earlier forms, characteristic of the upper- most beds of the Pottsville formation, are so closel}'^ related to Sphen- ojyh'vU Royi that the two are sometimes difficult to distinguish. The pinnules of the former are, however, generally more dilated, the lacin- ia3 more divergent, acute, and less coherent. The species appears to have diminished in size in the Lower Coal Measures, where it is per- haps inseparable from the type described by Lesquereux as Sphenop- teins trichomanoides Brongn. Frequently the reduced size and the obliquity of the pinnules and lobes appear to distinctly relate it to Sphe- nopteris dissecta and S. alata. The species is readily separated from Sphenopteris patentissima^ of the Lower Lykens division, by the very distant and deeply palmately lobed lower pinnules of the latter, the lobes being relatively long and hardly contracted near the base. In the Southern Anthracite field, Sphenoptei'is furcata occurs in the upper part of the Sewanee zone at the Pottsville Gap and at the New Lincoln mine. Sphenopteris Royi Lx. 'i'hc salient feature of this species, which was described by Les- ({ucn'ux,' from the roof of the Sewanee coal at Rockwood, Tennessee, is the oblicjuity of the rather distant, pinnatisect piniuiles, whose very narrow lobes are fixed on a somewhat more elongated axis than in S. fun-afa. or S. p}ate)itissiiii(i^y\h\\e at the same time they are very obJKpu', tapering from a slightly coherent base to an acuminate ])oint. Superficially, this species seems to be intermediate between S. alata on the one hand and S. furcata on the other hand. It is found with S. pahiiatiloha at the New Lincoln colli«>ry. A small and d()ul)tful fragment comes from the Upper Lykens division in the Pottsville Gap. 1 Cnal Flora, Vol. Ill, p. 708. pi. civ, fig.s. 7-10. WHITE.] NOTES ON" CHARACTERISTIC SPECIES. 883 Sl'HENOPTERlS PILOSA Dn. Tho specimens, including' the types, from the upper portion of the Pottsville in Washington County, Arkansas, described by Les(|uereux' as Sj)henopterls communis^ appear to present the identical specific characters seen in examples of Sphenopteris pilosa from the so-called middle Devonian beds at St. John. New Brunswick. In the Sewanee zone, which includes the Arkansas beds, of which the species is characteristic, we find it associated, as at St, John, with Pecopjteris serrulata. The specimens from the Southern Anthracite field are found near the supposed horizons of Lykens coals No. 2 or 3, at about 465 feet below the Twin coal, iti the gap at Pottsville. Aloiopteris georgiana (Lx.). The material described b}^ Lesquereux as Pecopteris georgiana" is generically identical with the Aloiopteris Sternhergii^ A. Witislovii, and A. e/'osa of the Productive Coal Measures. The species is notable for the great length and proportionate narrowness of the rather distant ultimate pinna?. The pinnules are nearly always at least tridentate, the primary nerve forking below the middle, and once or twice again in the upper part of the pinnule. The species which is, 1 believe, the antecedent representative of this group in our American Carbon- iferous is readily distinguished from A. Winslovii by its narrower ultimate pinnae, the small pinnules, and the coriaceous texture. The pinnules are proportionately a little farther distant, narrower, and distinctly curved, the teeth sharp and directed forward, the nerves curved and strong. Compared with A. Sternhergii^ the Georgia species is much more cristate, the teeth more acute, the nerves stronger, more open, and more upturned. The normal type is apparently character- istic of the Mariopteris pottsvillea zone in the Lookout formation in northwestern Georgia. The specimens in hand are from the roof of Lykens coal No. 4, at the Lincoln colliery. Oligocarpia alabamensis Lx. Specimens from the Lincoln colliery agree in all respects with typical material from Helena, Alabama, the original locality of the spe- cies. The plant appears to be, on the whole, characteristic of the basal portion of the Sewanee zone, though it may be found in the upper part of the Marioj)terls potUtrUlea zone. Pecopteris serrulata Hartt. The representatives of this species in the Pottsville formation appear to be in all respects in agreement with those from the type locality at 1 Coal Flora, Vol. Ill, p. 762, pi. xeiv, figs. 1 and la. 2Ibid., p. 759, pi. xcviil, figs. 6 and 6a. 884 FLORAL ZONES OF THE POTTSVILLE FORMATION. St. John. New Bnuiswick. With the exception of a few fragments supposed to have come from the roof of Lykens coai No. 4 at Brook- side, the species is not known below the Sewanee zone in the United States. Its geographic distril>ution in this zone is wide, the same form l)eing coUected in the shales over the Sharon coal in northwestern Pennsylvania, the Sewell coal in the Virginia region, the Sewanee coal in Tennessee, and in the coal-bearing shale in Washington County, Arkansas. In the upper part of the Sewanee zone this plant appears to merge into the form connnonly known as Pecopteria ijluim isa Artis. In fact, in the region of the Fayette formation the latter appears to have succeeded the New Hrutiswick fern. Alethopteris Lacoei sp. nov. PI. CXCIII, Figs. 1. ± Secondary pinnfe linear-lanceolate, very long, acute, slighth' nar- rowed at the base, often gently curved, with a distinctly lineate rachis; ultimate pinna^ of moderate size, usually alternate, but often sub- alternate, sub-opposite, or rarely opposite, open at nearly a right angle to the rachis. usually close, sometimes slightly overlapping, more rarelv a little distant, somewhat curved, linear-lanceolate, or linear, tapering to a slender, acute point; rachis finely lineate, rather deeply depressed, strongly concave and canaliculate ventrally. dorsally terete. Pinnules alternate or sub-alternate, the lowest open nearly to a right angle, close or a little distant, slender, linear- triangular, or ovate- triangular when small, hardly constricted at the base, tapering o-entlv through the lower third, the margins converging a little more rapidly in the upper two-thirds, acute or slightly obtusely pointed, cut to near the rachis with a rather broad, acute sinus, regu- larly and strongly crenulate-inflated in all portions of the frond; lamina rather thin, very strongly ventrally convex, alternately strongly inflated and transversly depressed at regular intervals of from 1 to 2 mm., according to the size of the pinnules, the margins being often slightly infolded dorsally so as to give the pinnules a tapering acuminate apex. Ner\'ation distinct, regular; midrib moderately strong, hardly decur- rent, straight, deeply depressed, tapering gradually, but distinct to the apex: nervils slender, relatively regular, rather close, hardly decur- rent, sim})le or forking once close to the base, and passing nearly straight and ])arallt'l to tlie liorder. Ahtltoptri'ix. Ldcoci is one of the smaller or more delicate species of tile geiuis, its forni and general proportions being closely comparable to .1. .IA///AV//,' or the smaller. iiaiTow-pinnuled phases of .1. Av/- c/iltica. It is, however, ver}'^ well marked by the alternating strong ' Brongniiirt, Hist. vt-g. foss., \\. l'T.h, |.l. Ixxxiii, fiK^i. '■'>. in, 1. WHITE] NOTES ON CHARACTERISTIC SPECIES. 885 inflations of the ventrally convex lamina, which gives the margin a slightl^v sinuate trend. This alternatel}- inflated development of the lamina, suggesting a series of rectangular cushions on either side of the midrib, or the expressions of distant sori through the substance of the fern, is a constant character plainly discernible in all parts of the frond. The aspect of the larger pinnules is very much like that represented by Brongniart^ in Pecopteiis margmata. But, while the resemblance of the undulate surface and sinuate margins is close, the species from Pennsylvania difi'ers by its smaller size, greater delicacy, narrower pinnules, longer and narrower terminals, as well as by the absence of the flat border of the Old World type. Quite independent of the infla- tion of the lamina, A. Lacoei is separable from the A. MaiitelU and A. Joncliltlca series by the form of the pinnules, which are hardly nar- rowed at the base, but which taper from the base upward, the apices generally appearing as acute. The nerves, which may fork close to the midrib, are nearly straight and regular in passing at a right angle to the margin, thus diflering from those of the A. SeMi or A. grandi- folia types. The latter, however, are too distinct in other respects to require further comparison. Although this species is common at the horizon of Lykens coal No. 2 in the Southern Anthracite field, it is generally rare in other regions of the Appalachian province. Yet when present it is usually represented by large numbers of individuals. In the anthracite region the species has been found at the New Lincoln mine, the lower Eureka tunnel, and at the supposed horizon of Lykens coal No. 2 in the gap at Pottsville. Alethopteris protaquilina sp. nov. The essential features of this species are the small proportions of the pinni\? and the close, very compact, oblong, or linear-lanceolate, obtuse pinnules, in which the limb is of about equal breadth at the base, the terminal being small, rather short, undulate and sub-lobate, . while the curved nervation is often concealed within the thick, strongly inflated lamina. The fern belongs to the straight-pinnuled group represented by A. aquilina, A. ambigua, A. Gibsoni^ and A. j>enn8yl- vanica. It is in most cases easily distinguished from the J.. Lacoei by the tapering, corrugated, more acute, and thicker pinnules of the lat- ter, in which the nervation is more distant. The species is for the most part confined to the roof shales of Lykens coal No. 4, at which horizon it occurs at the Brookside mines and in the Pottsville Gap. 1 Op. cit., p. 291, pi. Ixxxvii, figs. 2. 2a. 886 FLORAL ZONES OF THE POTTSVILLE FORMATION. Alethopteris loxciiitica (Schloth.) Sternl), The specimens from the roof of the Sharon coal in Ohio, described by Dr. Newberry under the above name, are probably in closer agree- ment with the original type than any of the other forms that have been identitied under the same name from the Allegheny series of this country. Most of the latter are probably referable to Alethopteris SerJil or A. aquiUna. A form with narrow pinnules, sometimes approaching AhtliopterisdecuTrens^ is found in the coals of the Kanawha series in southern West Virginia. The species is represented by the normal form, or by varieties, in the Upper Lykens division at the Pottsville Gap and at the Lincoln mines. Alethoi'Teris grandifolia Newb. This species, described by Newberry ^ from the roof of the Sharon coal, is, in general, characteristic of that horizon throughout the Appa- lachian province. Alethopteris Ilelence Lx. and (ydlipteridium Oireni Lx., both occurring" in the Sewanee zone, are so closely related to Newberry's species as to render their specific individuality very doubtful. The group is, in general, characterized by the lax, irregu- lar, and uneven open pinnules, the relatively thin midribs, and the distantly and irregularly ilexuose thin nervation. The flora of the Kemble drift constitutes the oldest plant association in which this species has yet been found. Alethopteris discrepans Dn. The specimens from the New Lincoln mine, which I refer to Sir "William Dawson's species, appear to agree in all respects with speci- mens from the fern ledges at St. John, New Brunswick. The occur- rence of this species, together with SpJienopteris Ilarttil^ S. pfhsa, and Pecopterk Herrvlata^ in the Upper Lykens division of the Potts\ ille formation points strongly to the close relationship between the flora of the latter and that of the supposed middle Devonian beds at St. John, a relationship so close as to convince me that no appreciable dirtcrciicc ill age exists between the plant beds at the two localities. Alktiioi'Teris coxtoniana sp. nov. The material which will be described under this name includes the types provisionally referred by Les([uereux" to Calliptcrhl'nnii Dotr- naku. The originals are from Campbell Ledge, in the Northern Anthra- cite field. The specimens from the Southern Anthracite field were found in the thin parting of the conglomerates at 245 feet below the > Report Geol. Survey Ohio, 1S7;5, Vol. I, Pt. II, p. :>M, pi. xlviii, figs. 1, la, iind 2. - Goal Flora, Vol. Ill, p. 747. WHITE] NOTES ON CHAEACTERISTIC SPECIES. 887 Twin coal in tlie Pottsville Gap. while doubtful fragments, representing a rather more elongated form of pinnule, with rather more distant nerves, come from one of the upper Lykens coals at the New Lincoln mine. Alethopteris E^vansii Lx. PI. CXCII, Figs. 7, 7a, 8, 8a. This species, which was described by Lesquereux ^ from the shales accompanying the Sew^anee coal in Tennessee, resemhlos. Zonchojjteris in the form of its pinnae and pinnules, while the nervation suggests one of the more oblique-nerved, straight-pinnuled species of AletJiop- teris, or Call ipter Id turn. The surface of the lamina is rugose and distinctly, though finely, punctate. The nerves are close, regular, rather oblique, forking once or twice. The normal form of this species appears to be generally confined to the region of the Sewell and Sewanee coals of the Southern Appalachian regions and to the approx- imate horizon of Lykens coal No. 3, in the Pottsville Gap. A later form, with very much larger, semi-membranous pinnules, occurs at higher horizons in the Sewanee zone, both in Arkansas and in the Sotithern Anthracite field. Typically, this species occurs 650 feet below the Twin coal in the Pottsville Gap and in the dump from the upper Lykens coals at the Lincoln mine. A variety grandis appears to have come from the roof of Lykens coal No. 1, at the latter localit}". Callipteridium pottsvillense sp. nov. In this species w^e have one of the composite Pottsville types, pre- senting at once characters of Megalopteris and Alethopteris. The pin- nules are elongate, acute, and thick, resembling very closely those of Megalopteris marginata Lx.^ It is also apparently related to Nerlop- teris lanceolcda of Newberry.' The fern is closely allied to C. tracy- a?ium Lx., from which it differs by the oblong, acute, or acuminate pinnules, w^hich are more unequal at the base, and the less distant and generally oblique nerves. The plant occurs in the Upper Lykens division in the Pottsville Gap. Megalopteris plumosa sp. nov. The species of the rare genus Mcg(dopti, including the Megalop- teris Daiosoni described from St. John, New Brunswick, appear to be characteristic of the Pottsville formation. Furthermore, the greater number of species thus far described are confined to the Sewanee zone in Arkansas, Tennessee, West Virginia, and Ohio, as well as in 1 Coal Flora, Vol. Ill, p. 834. 2Idem, Vol. I, p. 152, pi. xxiv, figs. 4, 4a. 3 Kept. Geol. Survey Ohio, 1873, Vol. I, Pt. 1 1, pp. 378-381, pi. -xl v, figs. 1,2, 3, 3a. 888 FLORAL ZONES OF THE POTTS VILLE FORMATION. Penns3'lvania. The most iiitorostino- species of this genus occurring- in the Pottsville formation of Pennsylvania, Megalopterin plumom^ closely resembles J/ii>^///j.yw// Hartt from the so-called middle Devonian of New Brunswick. It ditfers from the latter chiefly l)y the very oblique nervation. The specimens were obtained from a slope in the Upper Lykens division at Yellow Springs (jap. Xkl'roptkkis Pocahontas sp. no v. PI. CLXXXIX. Figs. 4, 4a; PI. CXCI, Figs. 5, 5a. Fronds large, tri- or quadri-(0 pinnate, with very broad, strongly lineate, slightly flexuose rachis, which may attain a diameter of 4 cm. or more; penultimate piniue generally alternate, open nearly at a right angle, becoming somewhat oblique above, close, often touch- ing or slightly overlapping, linear-lanceolate or linear, very slightly narrowed at the base, the margins nearlv parallel in the middle por- tions, tapering a little rapidly near the top to an acute apex, the ultimate pinnse being followed by a few large pinnules, rapidly succeeded by a narrow, l)asally sublobate, small, obtuse, inequilateral, ovate-triangular terminal, the rachis being rather strong, slighth' depressed ventrally, lineate, and a little flexuose near the apex; ultimate pinnae, alternate or subalternate, open at a right angle below, slightly oljlique above, usually slightly overlapping or touching, rarely a little distant, the smallest narrowly oblong, becoming linear, 5 to 30 mm. wide, 12 to 15 mm. in length, the lower small pinnte very obtuse, the more elongated being rather narrowly obtuse; rachis strong, depressed, lineate, slightly curved or flexuose. Pinnules small, slightlv polymorphous and irregular, alternate or subalternate, rarely subopposite, those in the lower portion of the largest pinna- or the basal pair in the small pinnse at a right angle to the rachis, the others more or less oblique, usually touching or even overlapping, more rarely a little distant, laterally unecjual, often ovate- round when very small, the lowest pair in the very small pinntB being often nearly reniform, the succeeding pinnules broadly ovate, nar- rowly ovate to ovate-ol)long, round at the apex, onl}' the lowest pair in the smaller pinrne, or the lower large pinnules in the large pinnje, or those a little below the apex of the penultimate piiuue, constricted to near the midrib, the others being less constricted, especially at the proximal angle, those near the top of the large pinnse or throughout thi' greiitcr })art of the small pinnse l)eing attachinl by more than one- half the width, often nearly the whole width, of the piniue, after the type of CalUpteridium^ the terminal being ovate or ovate-ol)long in the sniallei- pinna'. lat(M-ally unecjual, subl()l)at(^ by coiifluenc*' with the last piniude on one side, slightly undulate, usually obtuse or rounded at the apex, the terminals of the \'erv large piiunv biMug rather inoi'iu'longated NOTES ON CHARACTERISTIC SPECIES. 889 and less broadly rounded or obtnse; lamina thick, sliohtly coriaceous, a little concave along the middle, somewhat convex ventrally at the border. . . , Nervation rather coarse, distinct, regular, usually slightly in reliei, more or less Habellate in all except the lowest pinnules or the lower part of the largest ultimate pinnae; primary nerve but slightly dilTerenti- ated in the small pinnules, or of moderate strength, vanishing near the middle in those pinnules of intermediate size, or passing three- fourths the length of the largest pinnules, decurrent in the smallest, nearer the distal sinuses of the laterally unequal pinnules; nervilles very oblique, often nearly equally close in all parts of the lamina, a large portion springing directly from the rachis, especially in the proximal half of the pinnule in all but the very large or the lowest pair of pinnules, forking twice at a very narrow angle, one or more of the divisions forking again even in the very small pinnules, usually forking three times, sometimes four, in the largest pinnules, while passing, with slight or sometimes no curvature in the smaller pinnules, obliquely to the ..largin. The group of modifications or very nearly related forms which is typified^ by the fern just described is at once the most predominant, interesting, and complex in the fern flora of the entire Pottsville series. The genus is typically distributed in the lower division of the Pottsville serie^ of the Appalachian trough, and wherever fossil ferns are to be found in that division some form or other of the group is present and constitutes by far the most abundant, if not the exclusive, fern spe- cies of the flora. Essentially this type is characteristic of the lower division of the Pottsville, it being especially abundant in the vicinity of the Pocahontas coal in the greatly expanded section of that for- mation in southwestern Virginia. The distribution of the allied forms, as well as the typical form, will receive special attention in another place. The typical form, described a1)ove, the illustrated specimens of which were collected from the roof of the Pocahontas coal in the Flat Top or Pocahontas coal field of southwestern Virginia and southern West Virginia, is especially distinguished from the related forms of the same 'group by its broadly attached or Callipteridioid pinnules and the obliquity of its nervation, which is close, regular, coarse, and derived in part from the rachis, the midrib being very poorly defined in the small pinnules. Like the other forms from the basal portion of the Pottsville series, it is essentially a Neuroatn!)>terk. This syn- thetic character of the group of old Neuropterids in the earlier Potts- ville is particularly important as indicating the connuon origin of the genera Nenrupterl^, CalUpteridlum, and Miroiu^tira. Certain other Neuropterids of the type of N. hlform,^, whicli is a typical Neu- mlethopteris, similarly serve as connecting links between the genera 890 FLORAL ZONES OF THE POTTSVILLE FORMATION. Xenropteru and Alethojyteris. The subject of the origin and relations of these genera to Megalopteria and other Paleozoic types has Ijeen discussed by me in connection with the description of a somewhat composite form from the Des Moines series of Missouri/ The pinnules of the typical Xewy/ptcris Pocahontas are, in general, rather broadl}' ovate when small, and rounded at the top. The form of the terminal pinnules of the younger pinnte resembles that of the true X. SmitJisll as originally described and figured from Alabama, but the latter is considerably smaller and less sublobate, besides l)eing rather narrower. The species in hand is in reality readily distinguished from X. Smithsuhj its generally larger pinnules, which are broadly attached, more ovate, instead of oval or nearh?^ round, when small; by the far less developed median nerves, and especially by the much less curved nervilles, which are oblique, springing in part from the rachis, and which seldom meet the border at a right angle. Throughout the Appalachian trough the typical X. Smithsii has hitherto been found to occur in later beds than the typical X. Poca- hontas. The former is, in the central and southern Appalachian dis- tricts, fairly characteristic of the next higher divisions of the Pottsville series, the Clark and Quinnimont formations, or the Horsepen group. The Xeurojyterls ScKiehani Stur is with little difficulty distinguished by its narrower pinnules, which are constricted at the base, the midrib well developed, the nervation strongl}^ curved and meeting the margin at nearly a right angle. The form described by. Stur as X. Dluhoschi^ which Zeiller regards as inseparable from the preceding species, has much that is suggestive of the largest phase of the X. Elrodi of Lesquereux. Among the several modifications or variations of X. Pocahontas found in different regions of the Appalachian trough two fairly well- marked forms are present in the Lykens coal region of the Southern Anthracite field. One of these is so difi'erent from the ordinary t3'pe as perhaps to entitle it to more than a varietal distinction. But since its shape and mode of develophient are so similar to the normal type, since its earlier examples are somewhat intermediate, and because it is often difficult to discriminate between small fragments of the latter and material apparently derived from young fronds or apical portions of primary j)innte of the normal form, it seems most practicable to give it only formal or varietal rank. It may be termed and i-harac- terized as follows: Neikoiteris Pocahontas var. iN.F.QtAi.is n. var. PI. XCLXXXVIII, Fi- r>: I'l. CXC, Fig. 7; PI. CXCl. Fi-rs. 1-4. Ultimate pinnte and i)iiHuiI('s nuich larger than those of tiie normal form, often twice as large, somewhat lax, rareh' opposite, the pinnules 1 A new Tseninpteroid fern and it.s allie.«: Bull. Geol. Soc. America, Vol. IV, 189.3, \>\-i. 119-132, {>!. i. WHITE.] NOTES ON CHARACTERISTIC SPECIES. 891 a little more distant, oblique, slightly polymorphous, though generally ovate to oblong-ovate when small, becoming oblong and gradually constricted at the base, the largest attached at the midrib, narrowed a little toward the usually very unequal, obliquely rounded, asymetrical base, the attitude and form a little variable, tapering somewhat in the upper two-thirds, obtusely rounded or rounded at the apex, the termi- nal often elongate-ovate, often nearly acute, the margins, as in the largest pinnules, more or less distinctly sinuate; lamina thin, very slighth" convex ventrally at the border; nervation thin, but very distinct; midrib, when developed, becoming flexuose and vanishing in the upper part of the largest pinnules; secondar}^ nerves a little dis- tant, ver}' oblique, usually forking close to the point of origin, the divisions forking twice or three times in passing to the margin, which, even in the largest pinnules, they meet at varying degrees of obliquity. This variety presents certain phases which would appear to entitle it to full specific rank, though the presence, at one point or another in the Appalachian trough, of other forms showing every degree of transi- tion or genetic connection renders its specific separation impracticable. It afl'ords a fine illustration of unquestionable modification. Thus, certain of the largest of the pinnules are less obliquely narrowed at the base, more distinctly oblong, and even-margined. Such have in our collections sometimes been confused with Neuropteris Informix Lx., and accordingl}^ so recorded in the distribution of that really Alethopteroid and very rare form.' The phase whose fragmental (often detached pinnules) representatives have been the subject of this error is somewhat characteristic of the middle portion of the Pottsville series. Another variation, whii-h can be considered as onh" varietally or per- haps formally distinct from the one in hand, is seen in the plants from the Dade mines in northwestern Georgia, described and illustrated as secondary types of JVewopteris SinithsU. ^ While the Dade plants show an outline and habit hardly distinguishable from the variety in hand, they reveal a rather coarser nervation, which is more open near the midrib, often slightly flexuose, and usually a little more distant. A comparison of the types, or even of figs. 1 and 2, pi. xiii, of the Coal Flora, the originals of JV. SmitJisU^ with the illustration of the Dade specimen, pi. xcvi, fig. 3, of the same work shows at a glance the specified difi['erences in the fossils. The true Df. Smithsii has small pinnules open at a right angle, quite constricted at the base, well iThe latter, as seen in examples from Alabama, some of which were identified by Professor Les- quereux, have long, tapering pinnules, thiek. persistent midribs, strongly arched close nervation, the terminal and preceding pinnules being almost typically Alethopteroid. These features, slightly imperfectly shown in the Coal Flora (p. 121. pi. xiii, fig. 7) will later be more fully illustrated from typical material. *Le.squereux, Coal Flora, Vol. Ill, p. 734, pi. xcvi, figs. 3, 3a. No. 1156 Lacoe Collection, T'nited States National Museum. 892 FLUKAL ZONES OB^ THE POTTSVILLE FORMATION. clevolopod iiii(lril)s cvon in the siiuiU piiimiles. with v«M'v open nerv'ation orig-inatino- from the midrib. It is, moreover, a much smaller species. Many of the larger pinnules, with strong^' oblique, unequal bases, arc possibly suugestive of the XeMi'opteris anteceden.soi Stur, from the Ilainichen- Ebersdorf ])eds. The very strong resemblance of specimens from the Kalmia mine to Cardiopteris eriana Dawson^ is worthy of note, as is also the association of the latter with the Odontopteris squamosa Dn.,' which deserves a special comparison with the Neuropteris Pocahontas group of Pottsville forms. Xeiroitekis Pocahontas var. pextias n. var. PI. CLXXXVIII, Fig8. 2, .3, 3a, 4; PI. CLXXXIX, Figs. 5, .5a, 5b. The other Lvkens form of Xeuropteris Pocahontas^ to which refer- ence has ])een made above, is, so far as I have observed, nearly every- where slightly older than the one above mentioned, its habitat in the southern tield being essentially in the lowest of the Lykens beds, coals Nos. 5 and 0. It may be distinguished f rem the normal type and other forms as follows: Pinnules smaller than the normal type, very broadly attached, hardly so crowded, more distinctly triangular, laterallv unequal, oblique, the terminals as well as the largest pinnules more elongate, often sinuate- margined, narrower and more acute, the proxmial basal pinnule situ- ated in the angle of the pinna, the nerves regular, more slender, rather closer, often less oblique at the border, the lamina being thick and faintly irregularly striated between the nervilles, very many of which, in all 1)ut the large pinnules, spring directly from the rachis. Occasionally the younger pinnules of this variet}^ assume a distinctly and rather broadly triangular form, while the pinnae are much more sliMidei- and acute, the narrow terminal being not infrequently sin- uate-margined. The nervation of this form is fairly distinct, though thin, the nerves close, regular, and in the larger pinnules often nearly at a right angle to the border, although occasionally they turn upward slightly just before reaching the margin. The surface of the rather thick ])innules is often shiny, though when viewed under the lens it is seen to be ii-regularly striate as though impnvssed by minute scaly hairs nearly parallel to the nervation, as indicated in PI. CLXXXIX, Fig. ftl). The terminal pinnules of the larger pinna^ are slender and acute, those of the smaller lateral piniiic being proi)()rti()nately long. This, the more apiculate variety of Xciirojift'rl.s Pijca/ioidas^'xa not likely to l)e confused with any of tlie other forms or varieties of the species, on account of the form and attachment of the pinnuU^s and the nervation. .\^. A'////V//.v//. which at times it somewhat resembles, differs, among many characters, 1)}' its basally constricted, short pin- ' Foss. PI. Brian, Pt. II, 1882, p. 114, fig. 4. 2 Op. cit., p. 114, fig. 2. ,.„,„.] NOTES de slope,at westlide of Rausch Gap, Schuylkill County, and from a low Ixd the Pottsville Gap. Neubopteris Smxthsh IjX. The typical form of this species, first described ^>yP^°^'^^ xiii, tig^. 1, ^^, ""^^L^,,:,,!,,^! or true form is characterized l>y the --S:rpin^l:XLre«^.e..t^^ ,L margin, meeU^^^;^^^^^^^^^ West Virginia and whjh apPeai^ ^^f^^^J^^, ^..^u-acite Sf ^:fr^^:;;:;t;.^? "- -- -ume of .he co. ..ra under the above name jiffip,,Uv distin.-uished from *Kny- .fe«/,^^i. SmM.n IS - ho; j^;f; '>,^'^., : ,;,, ,s those indicated teru M,>«t<. ^->;--f'^^'^^f;:"".n,e pinnules of the variety above in spcjakinK ot the Uaae ler, 1 ^ . ; ■■ .^^^ ^.„„. pentias, while often .,f the same size -f , ^ ^ ^^ /j.;, ;,_ ,„d bdont«p- spicuousfor their broad attachment, ' '-"'S"^"'^ ' ^j „„,, ,,eut«. teroid nerves, while the '-■'''""^'? ^^'^^^ *X « ^.nec zone of the srtst;,t dt^.;;:;:- ;^ t...«., .«-. ^y it. 894 FL()RAL ZONES OF THE POTTSVILLE FORMATION. trianoular. more unequal, inflated pinnules, with persistent midribs, and more open, strong-ly backward-curved nervation. Neuroptei'ls Sniithsil has been collected at the Brookside and Lincoln mines; at Kalmia. the upper Eureka tunnel, the Broad Mountain mines, and the Lower Lykens division in the Pottsville Gap. Neuropteris Elrodi Lx. The plants from the Southern Anthracite field identified under this name agree with the original specimens described and figured by Lesquereux^ as having been derived from the Montevallo coal in Ala- bama. The character of the matrix is, however, yer}^ distinct from that of other material in the collection from that locality, it being in \ery close agreement with specimens from the AVhetstone beds of Indiana, in which this species is also said to have been found. 1 sus- pect that the t3^pes originated in the Whetstone beds. The originals represent a species quite different from that in the roof of the Se- wauee coal in Tennessee, later figured"' by Lesquereux under the same name. The latter is, perhaps, inseparable from Neuropteris Schlehani Stur. The differences between this species and Neuropteris Smithsii., the only species in the Pottsville flora with which Neuropteris Elrodi is liable to be confused, have been noted in the remarks on the former. N. K. 107; Atlas, p. 3, pi. .\iii, fig. 4. -Il)i(l.. Vol. HI. !>. 73.i, pi. xcvi, (ij;s. 1. 2. sibid., Atlu.s, p. 7, pi. x.xxviii, tigs. 1, 1\ l'', 2, 3; text, Vol. I (18S0), p. 171. white.] notes on characteristic species. 895 Neuropteris tennesseeana Lx. MSS. The fragments representing this species, from a bed about 550 feet below the Twin coal in the gap at Fottsville, appear to agree in all respects with the types described in manuscript by Professor Lesque- reux as Neuropteris tennesseeana. This species, which seems to have been derived from the original Neuropteris PocaJiontas stock, and which has many features in common with Neuropteris heterojyhylla., appears to be characteristic of the lower portion of the Sewanee zone, in both the Walden formation in Tennessee and the Upper Lykens division in the Southern Anthracite field. Neuropteris ovata Hoffm. The normal form of this species, which elsewhere is not known at so low a horizon as the uppermost beds of the Pottsville formation, appears to be present in the Southern Anthracite field in one of the shale partings in the topmost group of conglomerates, at 245 feet below the Twin coal. The variety antiqua^ which is characterized by slender, apiculate pinnules, very broadly attached to the rachis, and b}^ the unusually oblique nervation, is present in material from the rock dumps at the Lincoln collieries, although I am not certain to which of the Upper Lj^kens coals the fragments should be referred. Neuropteris gigantea Sternb. Under this name I provisionally refer the material described bj" Professor Lesquereux ^ as Neuro_pteris suhfaJcata. The original form, as described and figured by Sternberg, if present in our Carbonifer- ous basins, appears, so far as yet known, to occur only in the topmost beds of the Pottsville formation. Most of these specimens from the roof shales of Lykens coal No. 2, or from the higher horizons of the Sewanee zone, are generally more elongated than the Old World type. The examples from the Kemble drift appear to represent a new variety, clavata^ the most salient or distinctive features of which are the generally greatl v elongated, though somewhat polymorphous, pin- nules, which are frequently broader in the upper third than in an}' other portion of their length; the extremely slender and poorl}^ defined median nerve; and the veiy oblique, close, hair-like, parallel nervation, a portion of which springs directly from the rachis. Neuropteris lunata sp. nov. PI. CXCIII, Figs. 3-T. Frond and mode of development of the pinnse probably similar to those of Neuropteris gigantea., the rachis attaining a width of 12 mm. or more, very distantly but coarsely punctate, the penultimate rachis being 'Coal Flora, Atlas, p. :?, p). xiii, figs. 5, 6; text, Vol. I (1880), p. 102. 896 KLOKAL ZOXES OF THE POTTS VILLE FORMATION". provided. l)t't\vet'n tlir ultiiiuite pinna', with polyuiorphous broad, .short pinnules. Pinnules of the ultimate pinine linear-, or sliixhtly triangular-linear- subfaleate, usually short, four to six times as long as broad, with narrow cordate or slightly squarrose base, tapering upward toward the obtuse or obtusely acute apex, and usually turned upward with a uniform curve throughout the whole length, though often nearh' straight; tiie rachial pinnules triangular, triangular-ovate, cordate, or even cordo- reniforni, and very small; lamina a little thick, depressed in a narrow, rather shallow furrow along the median line, slightly convex ventrally, especiall}^ at the margin. Nervation sharply distinct, ventrally depressed, dorsally in relief; median nerve not very strong, but distinct and traceable to very near the apex; nervilles originating at a moderately open angle, a little distant, forking at a slightly open angle near the base, and similarly once or twice again, according to the size of the pinnule, in curving toward the margin, which the}^ meet obliquely, except in the lower part of the pinnule, counting about 30 to 38 per centimeter. Although the pinnules of this species occur in great abundance on the surface of the shales, no large segments of pinnae have been found. Nevertheless, the presence of the small and somewhat polymorphous pinnules, corresponding to those from the rachis in the preceding species, as well as the phases and similar characters of the large pin- nules, strongly indicates a development of the pinna? in the same general manner as in Neuropteris gigantea^ to which it is undoubtedly closely related. The most noticeable features of the pinnules are their slenderness, the crescentic curvature, the squarrose-cordate base, the distinct mid- rib, and the slightly distant nerves, which fork a little widely, although the divisions may at once assume a nearly parallel direction. The piniudes of N'. hmata are more slender than those of y. gigan- tea, and proportionately less acute, the curvature, when present, being generally more uniformly distributed, slightly crescentic, through the whole length instead of being expressed as an upward turn n 'ar the apex. Often, however, they are but slightly curved or nearly' straight. The midrib is much stronger and more persistent even than in X. ZeUlarl Pot.,' it being clearly traceable, though thin, very nearly to the apex. The angle of division of the nerves, as noted above, appears greater than in the last species, and is certainly much greater, while the nerves are less oblique, than in N. glganfea var. clavata. Some of the small roundish rachial piiuiules seem, when detached, to be not separable from similar small basal pinnuUvs in some other species of Nenropterla. But the average piimules may be i.Tnhrb. fl. k. Pr. Oi-ol, T,!iiKlcsiinst n. Bcrgakad., Vol. XU, p. 22, pis. li-iv, text fig. 1-4. WHITE.] NOTES ON CHAEACTEKISTIC SPECIES. 897 eatsily differentiated by their form and nervation from anj^ other species with "which I am familiar. Occasionally a small narrow pin- nule, such as that in Fig. 7, suggests, particularly in its nervation, pin- nules of JT. rarinet'vis. Even these may be separated b}^ recalling the dilated and distinctly cordate bases and the slightly concave lateral margins of the latter. This species is abundant in the roof shales over L3'kens coal No. 1 at the Lincoln mines. ASTEROPHYLLITES PARVULUS Dn, The specimens from the Southern Anthracite field which I refer to this species appear to be in complete agreement both with the plant originally described under the above name from the supposed middle Devonian beds at St. John,^ New Brunswick, and with the mate- rial from Rushville, Ohio, described by Andrews^ as AsteroplnjUites? m hi utus. Subsequent stratigraphic and paleontologic study of the deposit of dark leather}' shales at Rushville shows them to be probably not far from the horizon of the Sharon coal in Ohio. Many of the examples of the AsterojyhyUltes from the Southern field are even smaller than those figured by Sir William Dawson and Professor Andrews. This species, perhaps the smallest of its genus, is closely related to Astero- phyUites arhansamis and A. grandis. Although to a certain extent characteristic of the Sewanee zone throughout the central Appalachain region, the Aderojyliyllites ])armil%is is not only common in the same zone of the anthracite region, but it is often abundant in the roof shales of Lykens coal No. 4, in which horizon it is found at the Brookside and Lincoln mines. ASTEROPHYLLITES ARKANSANUS nom. nOV. The species to which I give the name AsteropJiyllites arhansanus is that known in the American literature as Asterophyllites gracilis^ a name which, unfortunately, must be a1)andoned, the A. gracilis (Sternb.) Brongn. having priority. As originally described^ by Les- quereux, from the coal-bearing shales of Arkansas, the species is especially characterized by the thick axes of the branchlets, the angu- larity and dilation of the internodes, and the strongly reflexed, out- ward-curved, slightly angular, and rapidh" tapering, small, acute leaves. The plant is found associated with the Upper L^'kens coals at the Lincoln mines and at the supposed approximate horizon of Lykens coal No. 3 in the gap at Pottsville. 1 Dawson, Acadian Geology, 1868, p. 540, fig. 188 A»-=; Fossil Plants of the Devonian and Upper Silur- ian Formations of Canada, 1871, p. 27. 2 Kept. Geol. Survey Ohio, 1875, Vol. II, Pt. II, p. 424, pi. li, figs. 4, 4-'. 3Rept. Geol. Surv. Arkansas, Vol. II, ISGO, p. 310, pi. ii, figs. 4 and 4«. 20 GEOL, PT 2 57 898 FLOKAL ZONES OF THE POTTSVILLE FORMATION. ANXUT.ARIA ACICUT.AKIS (Du.) Reil. This species, which is, perhaps, most closel}' related to Anmdaria Todtata Brongn., and which was described b}' Dawson from the fern ledges (Lancaster formation) at St. John, New Brunswick, as A^'tcro- j>lnjl.Jttes acicidar'ts^ is not rare in the Sewanee zone of the Pottsville formation, its most frequent occurrence being in the upper part of that zone. Its salient features arc the slender axis and the very slender leaves, which are distinctly narrowed, in tapering form, both toward the base and toward the apex. It has been very well illustrated in the publications of Sir AYilliam Dawson.^ The ]:)lant occurs in the Sewanee zone at the Lincobi mine and in the Pottsville (iap. AXXULAKIA CUSPID ATA Lx. The typical form of this species, as illustrated bv Lesquereux," from the dark shales at Kushville, Ohio, has a rather wide geographic dis- tribution throughout the Appalachian region, in the uppermost part of the Sewanee zone, or in the zone of Lykens coal No. 1, it being one of the characteristic forms of that portion of the sections. In the Southern Anthracite field it occurs in the roof of Lykens coal No. 1 at the Lincoln mine, and in the plant beds 380 feet below the Twin coal in tli(> Pottsville Gap. AXNULARIA LATIFOLIA (Dn.) Kidst. A comparison of the material in hand with. specimens frcmi the middle Devonian at St. John '' unquestionably shows the specific iden- tity of the American material with that from New Brunswick. It is, perhaps, specifically indistinguishable from the leaf verticils from Campbell Ledge described l)y Lesquereux* as Calamites raniifer Stur. In the Pottsville formation this species, which appears to stand in an ancestral relation to Ammla7'ia .sfeUafa, occurs near the horizon of the Lvkens coal No. 1 — i. e., near the base of the Fayette formation in the Virginia region and in bed L, 380 feet below the Twin coal, in the Pottsville Gap. Sphenophyllum tenekrimim Ett. \ iir. euoxgatim n. \ar. PI. CXCIII, Figs. 8, 1». This variety ditlers from the species as described by Ettingshausen '^ chiefl}' b}' the considerably larger size of the verticils and the more > Quart. Jour. Gcol. Soc. Lond., Vol. XVIII, 1862, p. 310, pi. xiii, fig. 16a 16b; Acadian Geol., 1868, p. 5.55, fig. 194 H' ; Fossil Plants of the Devonian and Upper Silurian Formations of Canada, 1871, p. 28, PI. V, figs. .5-la-c, 57. ' -Coal Flora, Vol. Ill, p. 72.5, pi. xeii, fig. 7, 7a. *Da\vson, Quart. Jour. Ccol. Soc. Loud., Vol. XVIII, p. 311. pi. xiii, fig. 17a-c: Acadian Cii'ol., ISCiS, p. 53S, fig. 1S7 A and I)'; Geol. Hist. Plants, IKSS, pp. 78, 26.5, fig. 28a, D., D'. ■•Coal Flora, Vol. Ill, p. 703, pi. xci, figs. 4 and 4a. & Ilclmhackcr, Beitr. Kcnntn. Siidrandes Oberschl. Pol. Pteinkohlenf.. p. 2S, pi. iii. figs. 5-16; Stur Culm.-Flora, Pt.II, p. lOH, i.l.vii, text figs.21a-e, p. 110; 22,p.in- 2:i A-F, p.lH: LM,i).n.5. WHITE.] NOTES ON CHARACTERISTIC SPECIES. 899 frequent dichotomy. The proportions of the variety are very nearly those of the form described 1 )y Star ^ from the Schatzlar series. The plant in hand differs from Spheoiophylliim hifurcatum b}' its more slender stems, by its more delicate, narrower, more rigid, and deeplj^ dissected leaves, and by the linear, obtuse form of the lobes. S. htfur- catwn is more nearl}' connected by its ^•ariable leaf, Avhich is often much less divided, to the S. cuneifoUuin group than is the above- named variety, all of whose leaves are of the more linear, entirely dissected, delicate tj^pe. In the Southern Anthracite field this variety appears to be one of the plants characteristic of the horizon of the roof shales of Lykens coal No. 2, where it is almost invariably associated with Neuropteris Elrodi^ 3fariopteris pygmcea^ and a small species of Eremopteris. It is especially common in this horizon at the Lincoln mines. Sphenophyllum bifurcatum Lx. This species, as originally described by Lesquereux, from beds now known to lie within the Sewanee zone of the Pottsville formation in Arkansas, appears, as has already been suggested, to be intermediate between SphenoplnjUmn tenerrimimi and S. cuneifolhnn. The less- divided leaves of the Arkansas plant might easily be mistaken in some cases for the latter species. It differs, however, in the position of the leaves and their aspect in the verticils, in the wider form of the leaf, and in the more distant and generally more rigid teeth. The distinc- tions between Sj)Jieno2)hyllu7n hifmxatmnSindi S. tenerrimxiin have already been indicated. The former is not infrequent in the upper part of the Sewanee zone; more rarely it is found in the lower part of the same zone. In the Southern Anthracite field it is found over the coal mined at Kohlers Gap; at the horizon of the roof of Lvkens coal No. 2, at the North Brookside slope; and in the Pottsville Gap at the approximate level of Lykens coal No. 3. A form of doubtful specific identity occurs at a level supposed to be not far abo^•e that of the roof of L^dcens coal No. 4 at the Kemble drift on Broad Mountain. Sphenophyllum cuneifolium (Sternb.) Zeill. As has already been noted in the discussion of the floras in the sev- eral horizons of the Pottsville formation, two forms which appear to be referable to this species occur in this formation. The first, which is represented b}' the type figured ]^y Lesquereux in fig. 9, pi. xciii, Vol. Ill, of the Coal Flora, is especial!}^ characterized by its slender, narrow, lax form, thin, rather membranous texture, equal teeth, and slender, fine nervation, the forking of which occurs mostly in the lower part of the leaf. The second form may be briefly described as more closely resembling the smaller, narrower, and often dissected leaves ' Die Caliimarien rt. Schatzlarcr Seliichten, 1HS7. p.202, pi. xv, figs, la, lb, 4. 900 FLOKAL ZONES OF THE POTTSVILLE FORMATION. of Sjphenojjliylhim emarginatum. It exhibits the more rigid type of leaf, with the coarse nervation of tlie latter, to which the Pottsville form appears to stand in an antecedent relation. The leaves of the first t3'pe, which is more typical of the horizon of the Sewell-Sewanee coal, or Lvkens coals Xos. 2 and 3, in the Southern Anthracite field, are relatively rarely dissected, while those of the second or more rigid type, which occurs in the Upper Intermediate division in the anthracite region, are nearh' always unevenly dissected. Sphenophyllum tenue sp. nov. PI. CXCl, Figs. 6, T. Stems slender, distinctly though not very prominently ribbed, cari- nate, 1 to 3 mm. in diameter, branching a little freely, the branches, often in verticils of three, springing from within the bases of verticils of very deeply dissected leaves; nodes distinct, 1 to 5 cm. distant, usually having the leaves still attached; leaves in verticils of six, oblique at the base, very rapidly spread In nearh^ equidistant radia- tion, broadly cuneate, 1 to 2 cm. in length, usually about 8 mm. in width, very thin or membranous, lax, the apices round-truncate, more rounded near the angles, sometimes with a faint sinus at the cen- ter, crenulo-denticulate in 12 to 2-J: short, broad, round-obtuse teeth, or rarely more or less dissected in broad, \ev\ lax, obtuse, usually bi or tri dentate laminae, the lateral margins distinctly, sometimes rather strongly, concave and converging downward in a slender, very narrow, relatively long, slightly thickened base; primary nerve single and rather strong for some distance in the lower part of the leaf, forking four or five times at a slightly narrow angle in passing upward, slen- der and delicate, to furnish one nerville for each tooth or denticulate crenulation. The salient features of this species are the large, membranous, very broadly cuneate, crenulate-denticulate. slightly rounded leaves, with distinctly il: not conspicuously concave lateral margins and long slen- der bases, in verticils of six, on ver}^ slender stems. The aspect of the somewhat rounded, rarely sinused apex, the transparent lamina, and the narrow, slender bases, below such l)road apices, make the plant easily recognizable among the other species of its genus. The nerva- tion is thin and delicate, derived In' numerous bifurcations from a rather thick, single, primary bundle that passes for some di.-tance throuji'h the narrow base of the leaf before dividing. Leaves " the dissected type, such as that shown in PI. CXCI, Fig. 7, are more rare. Although they bear a closer resemblance to the laciniate forms of other large-leafed species, th6y are easily distinguished by the irregular, very broad, obtuse, and lax type of the lacinea\ which show a markedly broad spread at the top, and b}- the concave lateral profile and the WHITE] NOTES ON CHAKACTERISTIC SPECIES. 901 slender, almost stalk-like base, as in the normal form. The leaves at the base of the branch verticil are much more elongated and deeph^ and narrowly cut than in the normal t^^pe. Sp/itnajj/if/fhim ten lie ii\:>pein's to be related to the S. rnajiisoi Bronn, though the differences noted above seem to readily distinguish it from that as well as other species. It is also closely allied to another very large- leafed species, not yet described, from the Vespertine or Pocono series (basal Carboniferous) of southwestern Virginia, between which and the former it appears to be intermediate. The species here described is one of the most beautiful as well as wideh' distributed plants of the Pottsville series, ranging in identical forms along the Appalachian trough from Pennsylvania southward to Warrior, Alabama. In general it is quite characteristic of the middle or Horsepen division (Clark formation) of the Pottsville series, to which its horizon in the roof of Lykens coal No. 4 also belongs. Although several fragments from near coal No. 2 or No. 3 at the New Lincoln mine appear so nearly in agreement with the tA^ pes as to be not readily distinguishable from the typical form, I have not met it elsewhere in the Upper L3"kens division or the Sewanee zone. It is not certain that the specimen is from the Upper Lykens division. In the Southern Anthracite field the species occurs in the roof shales of Lykens coal No. 4, at the Brookside and Lincoln collieries, the Kemble drift, the North Brookside prospect shaft, and in the gap at Pottsville. Lepidodendron alabamense sp. nov. The species which will later be described in full and illustrated under the above name includes the specimens from the upper part of the Clark and from the Quinnimont formations in the Virginia region, and from the vicinity of the Warrior coal in the Alabama section, included b}^ Lesquereux^ in Lejyidodendron Sternbei'giL It belongs to a Lepidodendroid type presenting several phases or modifications in the different zones of the formation. Typically, however, as seen in the region of Lykens coal No. 4, or in the Quinnimont formation, in which it is generally found in association with Marlopterls 2)ottsvillea^ NeurojJteris Smii/i.'iii, Spheiiopliylluni tenue^ LepidopliyUimi qiiinni- montanum, and Trigonocajpitni Ilelence^ it is especialh' characterized by the robust, thick branchlets, densely clothed with thick, rigid, rather short leaves, which, though verv oblique at the l)ases, rapidly curve outward, upward, and then inward, while tapering gradually, so that their upper portions assume a distinctly incurved or somewhat uncinate form. Stems and branches of this type are present in the roof shales of Lykens coal No. 4 at the Brookside mines and at Kalmia, and in the Pottsville Gap at a horizon about 775 feet below the Twin coal. 'Coiil Floni, Vol. II, p. 36C. 902 FLORAL ZONES OB^ THE POTTS VILLE FORMATION. Lepidodendron clypeatum Lx. To this species, which has been well illustrated b}' Prof essor Lesque- reux/ I refer most of the specimens from the Sewanee zone in Arkansas and Alabama hibeled in our collections as Lepidodendron VeUJieimlamim Sternl), Some of its phases are onh' with difficulty distinguished from the type descri})ed h\ Sternberg as Lepidodendron R}ind<'tinunt^~ which, in turn, appears to be closely related to and often confused with Lepidodendron ohovatuni of Sternljerg. The forms I refer to Lepidodendr(m clypeatum have a wide distribution in the Pottsville formation in the Southern Anthracite field, their range being from the horizon of the roof shales of Lykens coal No. 4, or possibly No. 5, to the level of Lvkens coals Nos. 2 and 8, from which the specimens are specially abundant as well as typical. Lepidophyllum quixximoxtaxum sp. nov. This species of Lejndojyhyllurn constitutes one of the characteristic t3"pes of the Mariopteris jyotUvillea zone throughout the Appalachian region, although it rarel}" occurs in the horizons of the roof shales of either the Pocahontas coal or the Lykens coal No. 5. The collections from the Southern Anthracite field contain one or two specimens which ma}' possibly have come from the shales of Lykens coals Nos. 2 and 3. The species is especially characterized by the linear-lanceolate form of the bracts, which are 5.5 to 8 cm. long and 10 to 13 mm. wide at the widest point — some distance above the middle. The sporangi- ophoros are cuneate and proportionately long, their length being usually over one-fourth that of the blade, which is dilated and auricu- late at the base, slightly contracted just above, and which tapers with slightly convex borders from its wndest point, above the middle, to an unusually narrow acuminate apex. The midrib is broad and strong throughout, while the lamina is obscurely lincate longitudinally, the lines slightly diverging toward the margin. The most important dif- ferentiative features are the relative length of the base, the dilation in the upper part of the blade, and the acute apex. The bracts are much larger and generallv longer than those of Lepidophylluni canip- iM'lJianxnn^ the sporangiophores of which are short, small, and gener- ally rather broadly cuneate. So far as the distril)ution of these two species has yet been observed in the Pottsville formation, it ai)pears that L. eain2)heJlianKin is characteristic of the Sewanee zone and the l^pper Intermediate division, while Z. quinnimo7itanuin is, in general, almost exclusivelv confined to the Lower Lykens division. The latter species is present in the roof shales of the Lykens coal No. 5 at the 'Geol. Pennsylvania, Vol. II, Pt. II, p. 875, pi. xv, fig. 5, pi. xvi, fig. 7. Coal Flora, Atlas, p. VI, pi. Ixiv, figs. 16 IGa (not figs. 17, 18) . 2SeeStur, rulm-Flora. Pt. II. p.2S3, pi. xxiii,flg. 1: pi. xxiv, fig. 1-3. ^Lesquereiix, Coal Flora. Vol. Ill, p. 7s0, pi. cvii. fit;<. G, 7. AVHiTE.] NOTES ON CHAKACTERISTIC SPECIES. V)03 Lincoln colliery, though its frequent occurrence at other localities is at or near the horizon of Lykens coal No. 4. BOTHRODENDKON ARBORESCENS (Lx). The examination of the stem fragments in the types described by Lesquereux as LycopocUtes arlorescens' reveals leaf scars showing the typical characters of the genus Bothrodendron. The collections in hand from the lower portion of the Kanawha series in West Virginia show that this genus is not rare at that stage in the Coal Measures in the United States. Bothrodendron arhorescens, the originals of which are from the Sewanee zone in Washington County, Arkansas, while the Pennsylvania representatives occur in the plant beds 380 feet below the Twin coal in the Pottsville Gap, appears to constitute the oldest representative of the genus yet discovered on this continent. CORDAITES ROBBII Dn. The identity of the leaves from the Pottsville formation with the species described by Dawson' from the fern ledges at St. John, New Brunswick, seems to be fully assured by a comparison of material from the type localitv. The species appears to be especially common in the Upper Lykens division of the formation, although it has a wider vertical rano-e. It is quite possible that the form which will eventually be described as Cordaites Phinipsl is not more than varietally distinct from the St. John type. In the Southern Anthracite field the species occurs, as at St. John, in association with Cardiocarpon cornutum. It is found at the supposed horizon of Lykens coals 2 or 3, about 650 feet below the Twin coal, in the Pottsville Gap, in the shales accom- panying the upper Lykens coals at the Lincoln mines, in the Lower Lykens division at the Brookside mines, and in the Lincoln mine. It is'also found accompanying the upper Lykens coal in Yellow Springs Gap. Cordaites akgustifolius Dn. The material which I refer to this species appears to be in agree- ment with the species figured by Sir William Dawson from the tern ledges at St. John,^ rather than with the material earlier described from the Devonian at Gaspe. If the Gaspe fossils are specifically different from those at St. John, as appears to be the case from an inspection of the figures, the name should be retained for the Gaspe "coal Flora, Vol. III,^, pl. evi, fig. l! Lacoe collection, United States National Muse-nm, Xos. 'TquS' Jouf Ll Soc. Lond., Vol. XVIII. p. .10, pL xiv, fi.. 31a-e; Acadian Geol 186S, pp. 534 and ,544, flg. 190; Fossil Plants of the Devonian and Upper Silurian Formations of Canada, 18,1, p. 43, '^Qi^r? Jout'Sr^oe. Lond., Vol. XVIII, p. 318; Fossil Plants of the Devonian an,l .pper SilnrianFonnationsofCanada, lS71,p. 44, pl. xiv., fig. ]63a-e. ,h.- „ -..i li^ 428 " H'anadian Naturalist. Vol. VI, pp. 170 and 17G, fig. He; Logan, Geol. < unada, Lso.., p. -U, l.g. 428. 904 FLORAL ZONES OF THE POTTSVILLE FORMATION. t3'pes. From an oxauiiiiation of examples from St. John, I am disposed to regard the latter as possibly young leav'^s of Cordaites Hol'hii. to Avhieh they are undoubted!}' at least very closely related. In the anthracite Held, as at St. John, both species occur in the same beds. Cordaites grandifolius Lx. The most conspicuous features of this interesting- species are the great breadth and broadly cuneate form of its large leaves. The base is narrow, the lateral margins always more or less strongly concave, and the distal margin, or top, rounded or round-truncate, and cut, in the older examples, into short, broad, unequal, round-truncate lobes. In fact, the general form, texture, nervation, and the mode of lobation at the apex are suggestive of the Ginkgoales. This species appears to be present in the material from the west side of Kohlers Gap in Bear Mountain, and from the rock dump from one of the upper Lykens coals at the New Lincoln mine. The originals described by Lesquereux^ are from Campbell Ledge, near Pittston, Pennsylvania. Whittleseya elegans Newb. var. minor n. var. The interesting tj'pe of vegetation described by Dr. Newberry^ from the roof of the Sharon coal at Tallmadge, Ohio, as Whittleseya elegans^ has been discovered at a number of other points in the Sewanee zone, throughout the Appalachian province. At various localities and hori- zons a number of additional species or varieties have also come to light. The leaves of the variety minor are much smaller than those of the normal form, and are proportionately broader, they being usualh' a little broader than long. The nerve fascicles are also a little more crowded, numbering about fifteen to the centimeter at the top. The form, texture, and nervation of this species, which can hardh^ be else than a gvmnos])erm. are such as to app(nir to justify its reference to the Ginkgoales. The variety was o])tained from the parting between Lykens coals Nos. 2 and 3 at the Lincoln mine. Whittleseya micrgphylla Lx. Tiic salient features of this species, described by Lesquereux,^ are the small size, the distinctly cuneat(^ foi-m. and the fasciculately fibrous texture of the leaves, which range iwnw 5 to 15 mm. in length and average about <; mm. in width. The fasciculate nerve ])undles, which, as in 11' rlegxii-s^ are in i)art derived from the thickened lateral margin in the base of the leaf, constitute rather poorly detiiicd longitudinal ribs, each of which enters a usuall\- ol)scure, rounded tootii in the 1 Proc. Am. Philos. Soc, Vol. XVII, 1878, p. 318, pi. xlviil, figs. 1, 2, 2a. Coal Flora, Atlas, p. 16, pi. Ixxvii, figs. 1, 2, 2a; text, Vol. I, p. .530. SAnnals Sci., Cleveland, Vol. I, lAi8, p. iifi, fig.«. i, 2a-b. Lesqucreux, Coal Flora, Atlas, p. J, pl.iv, figs, 1,1a. Renault, Cours. hot. fo.ss., Vol. IV, p. C9, pi. v, figs. 9 and 10. 8 Coal Flora, Vol. Ill, p.S43. WHITE.] NOTES ON CHARACTERISTIC SPECIES. 905 distal margin. Tbie latter appears .slightly creuulate. The distinctly cuneate form and small size of this species, which is intermediate to the broad W. elegans on the one hand and the lineate IH Cainj.>h^lli on the other, readily separate the leaf from those of other species in the genus. From a stratigraphic standpoint this type appears to ])e one of the most important, since it seems almost exclusively contined to a small vertical range above the level of the Sewanee coal. The representatives of this species in the anthracite tield are from the roof of Lykens coals 2 or 3 at the New Lincoln colliery. AVhittle.seya Campbelli sp. nov. PI. CXC, Figs. 9, 10, 11. Leaves ver}^ small, linear or slightly oblong-linear, 12 to 22 mm. long, 2.2.5 to 5 mm. wide, generalh' 12 to 15 mm. long and 2.5 to 2.75 mm. wide, often very slighth' cuneate, straight or slightly arched later- ally, acuminate at the rapidly contracted base, the lateral borders nearly parallel from a point less than one-fourth of the way from the base upward to verv near the truncate, acutely though very obscurely denticulate apex, where they normally converge somewhat, the outer teeth being inclined inward and usually crowded against or overlapping the interior teeth; texture densely fibrous, thick, more or less dis- tinctly rounded-ribbed b}" 3 to 5 longitudinal, parallel, finely lineate ridges produced by the ver}' thick contiguous fascicles of nerves, each of the dense, broad fascicles entering a tooth ; petiole filamentoid, lax, very faintly striate, blending with the slighth" thickened margins of the acuminate base of the leaf; nervation distinct, dividing at or near the base in 3 to 5 or 6 close, greatly thickened fascicles, giving the leaf a parallel-ribbed and striated surface, the nerves of each fascicle being slightly connivent in one of the apical teeth of the leaf. The species described above is one of the most widely distributed plants in the Pottsville series, in which it ranges horizontally from northern Ohio and northeastern Pennsylvania to the overlap of the Cretaceous in Alabama, and vertically from near the base of the Potts- ville to the base of the main upper plexus of conglomerates. So far as is yet known, this species was the earliest of the representatives of its genus to appear, it being found, in the deepest section of the Appalachian Basin, nearly 1,000 feet Ijelow the levels of WhJttleseya elegans or W. inicrophylla,, and several hundred feet lower than any of the other closely related, more or less linear forms. In general it presents a great uniformity in its features, there being no marked variation in either form or size among the abundant examples which are to be found at nearly every locality. Some modi- fications are, however, to be seen in the course of its vertical range, as well as in its local development. Thus the oldest observed forms were 906 FLORAL ZONES OF THE POTTSVILLE FORMATION. prevailinj4"ly small, not over 13 mm. long, unci narrow, though the species soon assumed its normal proportions in the Clark and Quinni- mont divisions of the Pottsville in the central Appalachian region, and near the horizons of the Lvkens coal No 4 and the lower Lykens coals of the Southern Anthracite field. Occasionally larger or longer leaves occur locally in the Lo\ver Lykens division; but in the Sewanee-Sewell zone, or I'pper Lykens division, a larger form is somewhat characteris- tic, while in the upper part of the series, at certain points in the Southern States, and in the roof of Lykens coal No. 2, a very much elongated slender form is found to prevail, almost if not quite exclu- sively, locally. It is also possible that the large, arcuate leaves which occur near Birmingham, Alal)ama, and at the Kemble drift and Koh- lers Gap, are hardly specifically separable from the species in hand, notwithstanding their local abundance and exclusiveness. The characters of the bases and petioles of these leaves are shown in PI. CXC, Fig. 9. Typical examples are illustrated from the Lykens coal No. 5 in PI. CXC, Figs. 10 and 11. The specimen seen in Fig. 10 is below the average in size. As will be observed on an inspection of the drawings, the mode of origin of the nerve fascicles, the slightly ribbed character of the leaf, its dense fibrous texture, the teeth, in which the nerves of each fascicle appear to be somewhat con- nivent, and the nature of the petiole, all indicate its common generic relation to Whittleseya inicrophylla and W. elegans. Owing to the thickness and inward inclination of the teeth, the latter are often obscured by imbrication or superposition. Not infrequenth% how- ever, they are spread out erect, when the clearness of the dentation, the vascular system, and the slightly cuneate outline show their con- generic relation to the species considered in the preceding sections, some of the cuneate forms thus approaching the IT" mlcroj^hyUa. R«Mnark should be made of the presence, about 470 feet below the Buck ^Mountain coal, of a straight, normally cuneate-linear form attaining a length of 35 nnn. and a width of 6 mm. As with the other species of the gemis, none of the leaves of 117///- tJeseya CanijjhtUl have yet been seen with absolute certainty to be attached to any branch or axis, though they are sometimes found in large num))ers matted together, their bases o])li([uely converging toward a conuuon axis. 1 suspeet the fruits of the plant to be refer- able to RliahdocarpOH. Wli'iffhxt'tjti C(iii,j)l>tlJl^ is at once distinguished by its lini^ar fonu and small siz(> from aU the species of the genus yet descril)ed. The species occurs at nearly all localities of the Pottsville formation in horizons below the Fayette sandstone of Virginia, the Conoquen- 1 The spcc-it's is named in honor of my esteemed colleague, Mr. M. R. CiimpV)ell, geologist in charge of theareal cartography of the Centnil Apf>alachian coal fields, to whom I nni greatly iiidt'bte. 373, pi. xliii, fig. h. •'Lesfiuereux, Coal Flora, Vol. II, isso. p. ."itkS, pi. Ixxxv. figs. 40. 47. «NewlH-rry. Ilept. Geol. .Survey Ohio. I.s7:i Vol. 1. I'l. 11. ).. :'.73, pi. xliii. li.'. I. WHITE.] NOTES ON CHAKACTEEISTIC SPECIES. 909 Cardiocarpon obliquum Dn. The identification in the Southern Anthracite lield of this species, hitherto known only from the so-called middle Devonian at St. John, New Brunswick,^ rests upon the entire agreement of the American examples with specimens from the type locality' of the species. This fruit, which appears to me to be unquestionably distinct from C. acutum L. and H.. is especially common at the drift in the upper Lj'kens coal at Kohlers Gap, in Bear Mountain. It is also found in the roof of Lykens coal No. 2 at the North Brookside slope, and in the rock dump from the Upper Lykens coals at the New Lincoln mine. Cardiocarpon cuyahog^ nom. nov. This name is here proposed for the fruits which were described by Newberry^ as Cardiocarjxni orl)iculan\ the latter name having been employed in the preceding year bj' Ettingshausen^ for a similar fruit. The species occurs in the roof of Lykens coal No. 2 at North Brook- side, at the lower Eureka tunnel, and at the New Lincoln mine. It has also been found in the rock dump from the upper L3"kens coals at the Lincoln mine. Trigonocarpum ampull^forme Lx. PI. CXCI, Fig. 8. Examples of this species from the anthracite regions appear to agree in all respects with the types described b}^ Lesquereux* from the Sewanee zone in Tennessee and Arkansas. Throughout the greater portion of the Appalachian region this type of fruit is more common in and slightly characteristic of the Sewanee zone. I have not yet seen the species in the Lower Coal Measures. Trigonocarpum Helen^e sp. nov. The species of fruit which will eventually be described under this name includes a portion of the types described by Lesquereux' from the Pottsville formation in Alabama as Rhahdocarpos clavatusf Gein. Trlgonocarjyuni HelencE is distinguished from Trigonocarjmm amj)id- la^'on/ie var. spectabile by the generally narrower nuclei, the very much narrower and less prominent ribs, and the proportionately thicker envelopes with their relatively much broader, shorter, micro- 1 Dawson, Quart. Jour. Geol. Soc. London, Vol. XVIII, 1862. p. 324, pi. xiii, fig. 25; Fossil Plants of the Devonian and Upper Silurian Formations of Canada, 1S71, p. 61, pi. xix, figs. 225-226. -Annals Sci., Cleveland, Vol. I, 1853, p. 153; Kept. Geol. Surv. Ohio, Vol. I, Ft. II, 1873, p. 374, pi. xliii, fig. 10. ^Steinkohlenfl. v. Stradonitz, 18.52, p. 16, pi. vi, fig. 4. K'oal Flora, Vol. Ill, p. 823, pi. cix, figs. 18-21. Nos. 16536-165.38, Lacoe Coll., U. S. National Museum, 6 Coal Flora, Atlas, p. 18, pi. Ixxxv, fig. 20 (not fig. 14); text. Vol. II, 1880, p. 581 (part). 910 FLOKAL ZONES OF THE POTTSVILLE FOKMATIOX. pyUir necks. T. inmJlum Brongn. is much shorter and more pointed, besides having less distinct ribs. The type from St. John, New Brunswick, which I have designated T. Dmosojiianum differs from T. Ilelt-nw by its still more slender form and narrower acuminate valves. This species is, in the central and southern Appalachian regions, apparently confined to the Clark and Quinnimont formations: and more particularly to the basal portion of the latter. In the Southern Anthracite region the fruit has, however, a much greater range, examples having been collected from the roof shales of Lykens coals Nos. 4 or 5 at the Lincoln, Brookside, and Kalmia mines, and from the upper Eureka tunnel. It also occurs in the rock dump from the upper Lykens coals at the New Lincoln mine, as well as at the supposed horizon of Lykens coals Nos, 2 and 3 in the Pottsville Gap, Trigonocarpum Dawsoniaxum sp. nov. Accompanying the specimens of a very narrow and rather small Trlyonocarjmriu there occur in the same matrix numerous detached valves which agree so completelv with the fragments figured by Daw- son from the '" Fern Ledges" at St. John as ''fruits or bracts of uncer- tain nature,'' that 1 have ventured to include a portion of the latter material as well, in the same species. The figures given in the " Devo- nian flora "^ will serve to illustrate the Pottsville material which I name in honor of the late distinguished paleobotanist of America. The dif- ferences between T. Dav'sonianum^ which will later be more fully described and illustrated, and T. Tfelenm have already been indicated. The species is found in the roof of Lykens coal No, -i at the Lincoln mine and at the Kemble drift. It also occurs in the rock dumps at P^ast Brookside and at the New Lincoln mine. Examples probably belonging to the same type occur in the plant beds 550 feet below the Twin coal in the gap at Pottsville. Carpolitiies transsectus Lx. Detached semicircular bracts, or possibly sporangiophores, identical in form with those (h'scribed by Lescpiereux ' from the ''coal-bearing shale" of Wasliington County, Arkansas, occur at a munber of locsili- ties in the shales from the coals of the Upper Lykens division. Tlie structure of the organ from which these small semidiscoid fossils are derived is still uncertain. From their mode of occurrence and their association. I am. however, disposed to regard them as possiblv belong- ing to a strohih' simihir to or identical with that described 1)v Lesque- reux'* as Li-j^ldocysfix (piadrauijuJdrix. > Fossil Plants of the Devonian and Upper Silurian Formations of Canada, 1871, pp. 64, 92, i)l. xix, figs. 230a, 231, 231u-b (not fig. 230). -•Cf)al Flora. Vol. Ill, p. H26, pi. cxi, figs. 27, 27a-l). 3 Op. tit.. Vol. II, p. 455, pi. l.xix, fig. 5. WHITE.] AGE OF THE POTTS VILLE FORMATION. 911 Fayolta ^p. Althouoh the collections in the National ]\Iuseum already contain specimens from the Allegheny series (XIII) at Mazon Creek, Illinois, referable to this genus, it is represented, so far as I know, in the col- lections from the Pottsville series l)y only two obscure specimens. From an examination of the material from Mazon Creek, and of the types from the Chemung of northwestern Pennsylvania, described by Dr. Newberry as Sjjirasi'f's,^ I am convinced that the latter genus is essen- tialh' identical with the FayoJia ~ of the Old World. AGE OF THE POTTSVIELE FOE3IATION. In the absence of the full descriptive paleontologic evidence, I should prefer to refrain from a definite statement of conclusions as to the age or the equivalents of the Pottsville formation. Since, how- ever, the questions of age and correlation directly affect the classi- fication and nomenclature of the formations now being mapped in the Appalachian province, it is proper to offer a few brief generalizations which ma}' be considered as preliminary and, so far as they relate to European coal fields, as tentative or suggestive. The persistency of the formation, or some portion of it, in some phase or other throughout the American Carboniferous basins, its generally well-marked lithologic characters, the different conditions governing its deposition, its thickness, which may exceed 2.. 500 feet in the Virginia-Tennessee region, and its mostly ver}' distinct vege- table contents, as compared with the basal portion of the Lower Coal Measures, or the Allegheny series, in Pennsylvania and Ohio, appear to me to merit for these terranes distinct recognition as a forma- tion or series, coordinate not only with the Allegheny series, Cone- maugh series, etc., but with Lower, Middle, and Upper Coal Measures, as those terms are used in this country. It is to be regretted that while vmder the name "Pottsville series" the formation is ranked by most geologists with Allegheny series,^ etc., man;\' authors treat it as a part of the Lower Coal Measures, although it was originall}" distinguished by Rogers as coordinate with the latter. Its occasional inclusion bv geologists in the more comprehensive, but equivocal, "Coal Measures" is perhaps not wholly satisfactory, even when that term is used in the broader sense of "Upper Carboniferous." As has already been remarked, no conclusive proof that the oldest beds of the Pottsville ma\' be contemporaneous with the last beds of the red shale, or other marine Lower Carboniferous sediments, has yet come to light. Never- theless, if the explanation of the conditions of the deposition of the iXewborry, Annals New York Acad. Soi., Vol. Ill, 18*5, p. 217. 2ReniUilt iuifl Zeiller, Comptt-s Kendus Acad. Sci., Vol. XCVIII, p. 1393; Fl. Foss. ba.ssiu hoiiill. Commentry, 1888, Pt. I, p. 15; Pt. II. l>-90, p. 3(i9. 3 Bull. U. S. Geol. Survey No. tib, p. 1-9. 912 FLORAL ZONES OF THE POTTSVILLE FORMATTOX, formation here jiceepted is oorroct. it is possible that thoro is a slio-ht overlap of Pottsville time on that of the Mauch Chunk formation, in which case its designation as Coal Measures would he lithologic and economic only, rather than strictly accurate from the chronologic stand- point. The flora of the Vespertine (Pocono X), which has received atten- tion from Lesquereux', Meek' and Fontaine,^ like that of the cor- responding Horton series of Nova Scotia, studied bv Sir William Daw- son, consists of an almost exclusively Triphyllopterid or Aneimites flora, with several laciniate-lobed Sphenoptrrldx, and great numbers of Lepidodendron of the corrufiatum, type. The flora of the Mauch Chunk formation is as yet l)ut little known; but such material as has come to hand from the upper portion of the formation shows a marked affinity with the Pottsville flora. The Chester limestone of Illinois is said to have furnished some fossils which are closely related to those of the basal Pottsville beds. I may add that the Anet notes from the topmost ])ed (bed A, PL CLXXXII) of the red shale at the Westwood and Pottsville gs^.ps appears to be more closeh' bound to the Lower Carboniferous types than to the ordinary plant life of the Pottsville formation, and should, therefore, perhaps be excluded, together with the accompanj^ing Sphenopteris umhratilis^ from the flora under con- sideration. The plants of the Lower Lykens division, as a whole, appear to stand in the closest relation to the flora of the Ostrau-Waldenburg beds described by Stur* and generally regarded as Lower Carboniferous ("Culm"), though many geologists and paleontologists are strongly disposed to refer the terranes to the Millstone grit. The intimacy of the relationship, and the probalfle contemporaneity of our flora with the I'pper Culm flora will ])e more fully indicated when the Pottsville flora is treated more at length, in the monographic report. The flora of the Upper Lykens division seems to be directly related to that of the Millstone grit (rf Canada and portions of the Carbon- iferous basins of the Old AVorld, though the data for comparison are hardly satisfactorily complete. Th(> upper horizons of this division have also much in conunt>u with {\w flora of the Lower Coal Measures of Great Britain. The latter, it luay ])e noted, are, for the most part, paleobotanically oldt»r than tiie foiMiiation known by the same name in the Northern States of this country. The Upper Intermediate division would seem, from the identities and distribution of its plant sp(HMes, to be as late as the Lower Coal Measures of Great Britain, or the lower zones of the Westphalian in continental Europe. Mk-ol. Pennsylvania. 1h.tS, Vol. II, Pt. II. -Bull. Philos. Soc. Wa.>;liington, 1K79, Appendix. "Am. Jour. Sci., 3(1 surits. Vol. XIII, 1S77, pp. 3.\ ll.=>. < Abh. A. k.-k. geol. Reieh.sanst., Vol. VII, 1.S77. Pt. II. WHITE.] SUMMARY OF CONCLUSIONS. 913 One of the most surprising, as well as interesting, facts observed in the stuch- of the Pottsville floras is the large element that is common to the latter and to the flora described b}' Sir William Dawson from the supposed middle Devonian beds at St. John, New Brunswick. In fact, taking into view the entire flora of the Pottsville formation in the Appalachian province, the identities in the composition of the floras are so great, Avith respect to both genera and species, as to leave little room for doubt that we have in the "Fern Ledges"' at St. John beds of nearly the same age as the Pottsville formation in Pennsyl- vania. In fact, as has been remarked in the preceding notes, the characteristic forms of the St. John flora, such as Megalopteris^ NeurojAerls retorquata^ AlethojJteris discrej^o/ns^ Alethojjteris ingens^ Sphenopteris inlosa^ Sphenopteris Harttii^ Sphenopteris marginata^ Pecopderis serrulata, Annnlarla JatifoJla^ and xVnnularia acicidnrh^ as well as the numerous gymnospermous fruits, are so far identical with, or obviously most intimatel}' related to, the upper Pottsville types as to render it highly pro])able that a flora contemporaneous with that of the Sewanee zone is present in the section along the St. John Harl)or. On the whole, as may already have been inferred, while recognizing in the Pottsville formation a group of terranes equal in rank to Lower Coal Measures, Allegheny series, etc., I not only do not favor a classification which relegates the entire formation hard and fast to the Upper Carboniferous, but I even anticipate a possible necessity for its permanent division into two groups, the lower of which may even- tuallv perhaps be referred to the Lower Carboniferous. From the paleobotanic standpoint the Pottsville formation is the beginning of the Mesocarboniferous. SUMMARY OF COXCI^ITSIO^STS. 1. The Pottsville formation in the Southern Anthracite field is com- posed chiefly of massive conglomerates and conglomeratic sandstones of varying composition, the lower terranes being somewhat hetero- geneous and irregular, the upper generalh' more uniform and persist- ent, with better assorting of materials. The coals (Lykens), locally of great economic importance, exhibit the general variability of the formation, though they sometimes appear to extend over relatively large areas. 2. In the Schuylkill region the passage from the Mauch Chunk (XI) to the Pottsville (XII) is by a transition of heterogeneous conglomerates, intercalated in red and green shales, the proportion of sandstones increasing to the top of the red shales, which are later represented by red and green argillaceous materials washed into the soft but more distinctly arenaceous conglomerates and bowlder beds of the lower portion of the Pottsville. 20 GEOL, PT 2 58 914 FLORAL ZONES OF THE POTTSVILLE FORMATION. 3. Tho irrogularity and the lack of selection in the materials inter- larded with the upper beds of red and green shales (Mauch Chunk) appear to l)e natural results of the submergence and somewhat imper- fect working-over of an intermittently subsiding coastal plain under the action of strong and varying detritus-laden currents. 4r. No evidence of a marked or general iniconformity between the Pottsville and Mauch Chunk is noticeable in this regicm, though at various points within several hundred feet of strata beds of small bowlders or coarse conglomerates are imposed, in a knife-edge contact, on the distinctly uneven surfaces of olive-green mud beds. 5. The conditions are such that it is impossible to determine upon a persistent stratigraphic basal line of division which can be traced or recognized throughout the basin. Different geologists have taken different horizons. In this report the topmost bed of tvpical red shale or sandstone in the section is arlntrarily taken as the upper limit of the ]Mauch Chunk. This is the usage of geologists in the bituminous fields, w^here, it should be noted, an unconformity probably exists in most areas. 6. The upper limit of the Pottsville formation has, for reasons of necessity or practicability, been placed by the various geologists and surveys of Pennsylvania at the base of the lowest '* considerable coal,'' which usually occurs not far above the main plexus of massive conglom- erates at the top of the Pottsville formation. Such a horizon, though usuallv traceable for a distance of several nliles, is not always definite where, as happens in portions of the field, the distinctly conglomeratic character of the terranes continues into the Coal Measures and a num- ber of thin coals are interspersed. Yet this mode of delimitation, employed in conjunction with the kno^edge of the stratigraphic rela- tions of the low coals to the main upper group of conglomerates, has probably rarely led to any consideralile vertical error throughout the central portion of the field, including the vicinity of the type section; and, from the standpoint of field practicability, it is probably the most satisfactory method of definition at present available. 7. The flora in the roof of the Buck Mountain coal, or its supposed equivalents, at the very base of the Lower Coal Measures at Pottsville is a typical Coal Measures flora, very distinct from the floras typical of the Pottsville formation, although a few of its speci(>s appear in the upper 250 feet of beds in the latter, which contain a mixed flora. It is even slightly later than that of the l)asal l)eds of the "Lower Coal Measures"^ in the Northern Anthracite field or of the Allegheny 1 The term Lower Coal Measures is used in the anthracite fields in the original sense as proposed by Rogers for the series next above the Scral (Pottsville) eonglomerate. It is similarly applied in the northern bituminous basins, where it is, in part at least, synonymous with Allegheny series, Desmoines series, etc. It is, however, shown by the fossil plant-s to be as a whole somewhat later than tlie Lt>wer Coal Pleasures of the Old World. See Bulletin of the Geological Society of America, Vol. XI, pp. 1 1.^178. WHITE.] SUMMARY OF CONCLUSIOISrS. 915 series in the northern bituminous basins. Furthermore, the plants from the thinner coals in the tj^pe region, in some cases about 100 feet or more lower than the ''Buck Mountain" (Twin) bed, and close to or partl}^ within the top of the upper dense complex of conglomer- ates, with which the formation culminates, are also clearly paleonto- logically referable to the Lower Coal Measures, the}' being- comparable to the lowest floras a})ove the Homewood sandstone in the bituminous basins. In other words, the conventional stratigraphic boundary between the Pottsville formation and the Lower Coal Measures is, in the Southern Anthracite field, slightly higher than the paleontologic boundary. The paleontologic boundary appears to lie close within the outslvirts of the upper plexus of conglomerates which form the most conspicuous feature of the formation from the Pottsville Gap westward. 8. In the central districts of the field the formation probably attains its maxinnun thickness of a little more than 1,200 feet. Westward it thins very gradually on the whole, the thickness at Rattling Run Gap being 1,100 feet. Eastward it appears to I'apidly decrease to about 830 feet at Tamaqua, and perhaps less than 800 feet at the Lansford railroad tunnel, though it seems to expand somewhat to the east- ward of this point near the apex of the field. Little, if anj", diminu- tion is observed within this field in passing from south to north. 9. The fossil plants of the Pottsville formation in the type region exhibit a rapid development and series of changes or modifications, which, if treated with great S3^stematic refinement, are of high strat- igraphic value. With the exception of the species from the topmost beds of the formation, the ferns are, in general, readily distinguished specifically from those at the base of the Lower Coal Measures, or Allegheny series, as recognized in the northern United States, while the floras of the lower portions of the section are found, in passing- downward, to b(^ar still less resemblance to those of the Lower Coal Mcasurt\s. Two principal divisions of the formation, to which com- paratively few fern species are conmion, are recognized. These divisions, which coincide with the natural grouping of the Lykens coals, are here termed the Lower Lykens division and the Upper L3kens division. A portion, including about 200 feet of the type section between these two paleontologic divisions, contains a mixed flora, and has been temporarily designated the Lower Intermediate division. A portion of the type section, at about 200 feet below the Buck Mountain coal, contains floras largely characteristic of the Potts- ville, l)ut in association with a number of the earliest Coal Measures species. This is temporarily termed the Upper Intermediate division. The Lower Lj'kens division includes two floral zones, (1) that of the lowest beds, up to and including the roof shales of Lykens coal No. 5, and (2) that of tiie vicinity of Lykens coal No. 1. The first contains 916 FLORAL ZONES OB^ THE POTTSVILLE FORMATION. relatively few ferns, and is specially characterized by the invariable abundance of a species of Neuropteris^ as well as b}' the absence of the forms typical of the other zone. The Upper Lykens division reveals a principal zone of the Sewanee-Sewell coal flora, typically pres- ent in the vicinity of Lykens coals Nos. 3 and 2, but extending in mod- ified form up to Lykens coal No. 1. The flora of the latter horizon is characterized by modified survivors from the older horizons of the Sewanee zone, accompanied b}^ elements apparently peculiar to this portion of the section. 10. The flora of the lower zone of the Lower L^'kens dinsion is found in the vicinity of the Pocahontas coal in the very thick section of the Pottsville formation in the Virg-inia-Tennessee region. It is unknown in the thinner sections along the northern and western borders of the Appalachian trough, Beds contemporaneous with the upper zone of the same division are present in the upper portion of the Clark and in the Quinnimont formations of Virginia and West Virginia and in the Lookout formation of the Tennessee- Alabama r(>gion. The Lower Intermediate division of the formation in the Southern Anthracite field is shown b}^ the fossils to occup}-^ nearly the position of the Raleigh sandstone in the Virginia region and of the Sewanee conglomerate, the top of the Lookout, in Tennessee. The flora (""Sewanee") in the lower portion of the Upper Lykens division is essentially identical with that in the vicinit}'^ of the Sewanee coal in the lower portion of the Walden sandstone in the Ala]>ama-Ten- nessee region, the Sewell and the Dismal formations and a portion of the Norton formation in the Virginia region, and the Sharon coal of northern Penns^dvania and Ohio. The lowest phytiferous horizons of the formation yet studied in the bituminous basins of Pennsylvania and northern Ohio appear to be distinctly referable to this zone. It is dou))tful whether ])eds older than the upper portion of the Quinni- mont formation are present in these regions. The upper and less dis- tinct zone of the Upper Lykens division appears to be represented in the greatly expanded later modification of the Sewell formation of Virginia and southern West Virginia, near the base of the Fayette formation. The time of the upper 200 or 300 feet of ponderous conglomeratic plates at the top of the formation, constituting the Upper Intermediate division in the Southern Anthracite field, is appar- ently represented b\' over 800 feet of sediments in the southern Vir- ginia region, only the lower portion of which, including, probal)ly, a part of the Fa^'ette formation, has the lithologic characters of the Pottsville. The horizon of the well-known plant bed at (.auipliidl Ledge, which is within a few feet of the supposed Mauch Uhunk. in the Northern Anthracite field, is probably not lower than this division of the type section. MHiTE.] SUMMARY OF CONCLUSIONS. 917 11. Further paleobotanio study of the Pottsville formation appears to fully confirm the earlier conclusion, based on the examination of the plants, that the thinner sections of the formation along- the north- ern and western borders of the Appalachian trough do not contain beds as old as those in the lower portions of the thick sections along the eastern border, e. g., in the Schuylkill and Great Flat Top regions. The positions of the respective Jtloras in the sections plainh^ indicate a transo-ression of the sea toward the north and west during- Pottsville time. 1'2. Both lithologically and paleontologically the Pottsville forma- tion constitutes a division of the Carboniferous coordinate with the "Lower Coal Measures," "Allegheny series," etc. As such it forms the lower member of what may, in a broad sense, be termed the Mesocarboniferous in the Appalachian province. 13. The lowest beds in the thickest sections, which appear to be continuous by transition with the deposition of the Mauch Chunk red shales, are perhaps to be regarded as coarse, coast-detrital redeposi- tions, contemporaneous with the uppermost beds of red shale or other marine Lower Carboniferous sediments in other regions. The flora of the Lower Lykens division appears to be contemporaneous, in part at least, with that of the Ostrau-Waldenburg (Culm) beds of the Old AYorld. The flora of the Sewanee zone of the Upper Lykens division is perhaps contained in the Millstone grit of Canada and portions of the Old World coal fields, while it is probable that the Upper Inter- mediate division is contemporaneous with a part of the Lower Coal Measures (Westphalian) of Europe.' li. The flora of the Pottsville formation is so far identical, in both its generic and its specific composition, with that from the supposed middle Devonian beds at St. John, New Brunswick, as to leave no room for a great difference in the age of the latter. In fact, the plants from the "fern ledges" include a flora essentially equivalent to that of the Sewanee zone, which appears to be represented by a portion of the section at St. John. 15. Owing to the hitherto unrecognized presence of an overthrust in Sharp Mountain in the vicinity of Lorberry Gap,^ and the conse- quent misidentification of the less valuable coals in Lorberry and Fishing Creek gaps with the Lykens coals, the boundary of the low- est Lykens coal has been represented from Fishing Creek Gap west- ward, on the State mine maps, as close to or north of the crest of 'The base of the LgwerCoal Measures or Allegheny series m this country appears paleontologically to be nearer the stage of the Middle Coal Measures of Great Britain, or the upper zone of the West- phalian in continental Europe. -The discovery of the fault at Lorberry and Fishing Creek gaps was tlic direct result of the testi- mony of the fossil plants, which was later completely corroborated by tlie ordinary stratigraphic method. 918 FLORAL ZONES OF THE POTTSVILLE FORMATION. Sharp :Mountain, i. e., near the place of the Buck :Mountain l)ed, whereas the outcrop of nearly the entire Pottsville formation, in its full thickness, including- both of the groups of Lykens coals, extends from near Fishing- Creek Gap nearly to the Big Flats, a distance of over 17 miles along the south face of Sharp Mountain, quite outside of the ••approximate boundary of the lowest Lykens coal" and partly within the territory represented on the anthracite maps of the recent State geological survey as Mauch Chunk red shales. = ■< "^'-^ -..sannsvaK nv(i.i >i3AV(ri.. •KoiivKaoa ariiAsixoa- X ^ ^ fm im ii^-ii'v^i^i KOUVKUoa ■S3 1 J St i ca-3 4-3 = i t £; c - o s >>= O 4) 3^ - I . ;:, nN jri — j:"* .«> ^ S V; S ^ f ^^ i x" 3 Ji ,z ? 3 i3 Si £ c a st-r •r u* c ^ pa .i 3 1 11 iiliEi^ UVl'lTl'il' ' I i l||,ili!ililili|i![ ^fv.m.J HJ.IVK c £ <3 -B U. S. GEOLOGICAL SURVEY TWENTIETH ANNUAL REPORT PART II PL. CLXXXIII ^.wmia^ !:tX.:::»n' i,*--!; = o'^o'o^ , .boO-o ° -ot^o^'- s^s-.^Qp^fiSSSs 3(y conglomerate. 13' sandstone. shale. 2' 0" oongrlonierate. COAL BED. 15' sandstone. 12' fine conglomerate. 2' conglomerate and shale ■r conglomerate. 3' 6" shale. 5' sandstone. 16' 6" conglomerate, r sandstone. fe'a^'°°o"#?o^?^' 32- ronglomerate. o„0 cpScPcO V oto^^^ ■ ■ • V ■ ■■.'■: ■■■■■ • ■•. : ■ ■ .• .1 4' sandstone. 4' sandstone and conglomerate. g&'Sfo°ilf^?^^%°^e' 8" conglomerate. l?°.v° ■.v."^:r;:-.v...v.v.v.al 6 sandstone. VfJP°c^&'Kf*i''^o°'d 13' conglomerate. S?Sq.9c.Ss3i, ^„ san.lst.iiic. ■ ;. ■.■••. J2' sandstone. £o«°^3' conglomerate. LYKENS YALLEY NO. 1 BED. 3' s.u.dstone. I ■ •••■iW"»i'«-.-. :»;<»7:r*'?9i r?? 413' brought forward. 11' sandstone. ■ . . p. 1 . . .-"■dS 105' conglomerate and sandstone. 2'J' sandstone ami conglomerate. 3fi' sandstone and conglomerate 2[y ('»" sandstone and conglomerate. 38' 0" sandstone and cojiglomerate. LYKK.NS VAI.T.EY NO. 2 BKO. 9' shale. I.YKEXS VALLEY NO. 3 BED. 4' shale. 5 sandstone. 12' conglomerate. 3 sandstone. -=1 R" shale. : 22' sandstone. 9- fi" shale. 4' sandstone. 'grgapp^^fiD^jjpopo.oc 4' 'J" conglomerate. 3p.°cS.ct-c,.-. ^.mwm^.,. Ocf* = 16' sandstone and congl'% 0° 2c 32' f»" congliimerate. r fi" shale. |L\^^NS VAI.I.KY No. .5 BKI). I'j S ' sandst(me. ■ •.too '.(sfii.c .* to ^ ' * 1,20X' 6" brought forward. 26' 8" conglomerate. ' 2" red shale. 7' 5" sandstone. 35' conglomerate. 1' 10" red sha'e. 4' 3" conglomerate. 6' 6" red shale. 2' 9" sandstone. 16' 6" red shale. 10' 3" conglomerate. 9' sandstone. 29' 2" red shale. I'y 7" conglomerate. 8' 4" shale. 13' 4" sandstone. 8' 6" conglomerate. 8' 6" sandstone. Si8SSSK=S=3s'8' 8" red shale. &>"/.«. .=>..---.-.o... 17' 9" sandstone. 13' 11" conglomerate. 1' 6" red shale. r>' 0" conglomerate. 1' 1' sandstone. 9' 3" conglomerate. :;' C" red shale. Total. 1.47S' 11". 18' conglomerate LYKENS VAI.I.KY Xci. 6 BKI). a' S" sandstone. COAI-. 18' 8" shale. BASE OK I'OTTSVIM.K KOKMATION. 15' 7 ' KKI> SIIAI.H. ir .sandstone. 1,208' 6". SECTION OF THE POTTSVILLE FORMATION IN THE LINCOLN MINING DISTRICT. (CONTINUED FROM PL. CLXXXIII.) This plate represents the lower portion of the Pcttsville formation and the upper portion (transition beds) of the Mauch Chunk formation, Kalmia colliery water-level tunnel. Compiled from the State anthracite survey. Scale : 1 inch = 50 feet. U. S. GEOLOGICAL SURVEY TWENTIETH ANNUAL REPORT PART II PL. CLXXXV Klii.l. UAlsmGAr. SCMIYI.KILI.C^IITSTV. Pennsylvania. Section from the Twin eoal to the top of the Mauch Chiinlv formation, showiiit; tlie Pottsville formation. 1 inch =200 feet. "P" Oo~, o 6 ^0'.0:i;.i-°0 Coal. Twill. Tor ov IMiTTS 1,-'0I)'. i.^;-!5'. ilMlMirikr.Mlshalr. Vu. Section of the Pottsville formation. from the '-A" coal to the top of the Mauch C'hunlv reil sliale. 1 incli = ■<5 O'-'O-' a oO 200 feet. » Co O •? o 5 O i < • IMKlllilipstn IrrefTularly bedded i cimjflomerates. MallrhChiiiikn-clshalc FlIV. :!. l.nUMEKRY (;aP I\ SlIAUr JtdlN- TAIN. Section of exposed beds from the '-Iv r- nace " coal downward. Datum lim- is "South " coal. 1 inch = 200 feet. l.;i.'i'.i'. Coal, Kurnace. l,.S0(l'. Fig. 3— Continued. Coal. o,=,ci»coOoP6Ci .700' Coal. Unibehaue 1,100". 1.000'. Coal I drifted I Coal , drifted .. Coal (di-iftedi 600'. Coal. Zimmerman. •.')00'. Coal. "IVacock.' .100'. Coal! ?jlPc)«>i5>b°i.D' smm .200'. Blassive confflomerate Coal. '-Soutli.' 300'. 3.32'. ' Total, i.iirr. SECTIONS OF LOCUST MOUNTAIN AT TAMAQUA AND OF SHARP MOUNTAIN AT RAUSCH GAP. SCHUYLKILL COUNTY, AND AT LORBERRY GAP. U. S. GEOLOGICAL SURVEY TWENTIETH ANNUAL REPORT PART II PL. CLXXXVI FIG. 1. Fishing Creek G.\p, Two Miles North of Ellwood. Pexnsyi.va.ni a. Section including the coals opened within 500 feet of the red shale on the east side of the jiup. Datum line is lower coal mined. 1 inch = 200 feet. ■ ^.opiOOOOf c 240'. Coal I mined 1. (The horizon of this coal is probahly 400 ± feet above that of the Twin bed. I 100'. Ked shale! Fig. 2. Black Spring Gap (MorNT RAGLE). Lebanon (Bounty. Pknnsyi.- VANIA. Section from the -Black Spring" coal to the Mauch Chunk red shale, includ- ing tile Pottsville formation and a portion of the succeeding Coal Meas- ures. Datum line is tlicapproxin)ate horizon of the Twin coal. 1 iiich = '.'OU feet. '■:Q% ^Po(X9t ■•6°P ?.^o.W)'. Coal. -Heister." "400'. •'. (Page 880.) Fig. 1. Tliis specimen shows the lax hahit of the species and the deeply dissected distant i)innnk'S. Lykens coal No. 5 (?), Lincoln colliery. - Loiter Ijikcnx dirhinu — Lower zoiw. Xkikoi'tkris Pocahontas vuf. i-entias I). W. (I'iige892.) Fig. 2. Ajiical fragment, showing triangular form of n})i)er junnules. From Lykens coal No. 6 at East Brookside colliery. Fig. 8. From Lykens coals No. 5 or 6 at Kalmia colliery. Fig. '^(1. Pinnule from the original of fig. 3, showing tlie line lu-rvation, which is Callipteridioid at the hase. Twice the natural size. Fig. 4. A fragment with lobate ijinzuiles. From Lykens coal No. 5 at tlie Lincoln mine. Jjoircr L^i/h'iis dirlsioii — Loiix'r zoiir. Nechoi'tkuis Pocahontas var. iN.KtjUAr.is I). W. (Page 890.) Fig. o. Fragment sliowiiig ndoiitoptcroid pinnules of tlu' upper piiuijc r>cd 1>. 710 feet helow the Twin coal in the gap at Pottsville. Fig. ")(/. Enlarged detail showing the nervation of two of the ])innuk's illustrated in fig. 5. Twice the natural size. Jjiiri r Liiki'iix (lirixioii. 920 U. S. GEOLOGICAL SURVEY TWENTIETH ANNUAL REPORT PART II PL. CLXXXVIII SPHENOPTERIS AND NEUROPTERIS. Lower Lykens division. PLATE CLXXXIX. !)21 PLATE CLXXXIX. MaKIoPTKKIS KKK.MOl'TKHOJDES 1 >. W. (Page 872.) Fi shown in fig. 1. Fig. 3'( Pinnule of the same specimen enlarged to show the nervation. Twice the natural size. Lover LiikrtiK (lirisloii — Loxrr zone. Nel'koitickis Pocajioxtas I). W. (Page 888.) Fig. 4. Specimen showing the characteristic form and development of the pinnules of this species. From the roof of the Pocahontas coal near Crozens, in the Great Flat Top coal field, West Virginia. Fig. 4'^ Pinnule of the same specimen enlarged to show the nervation. Twice the natural size. Aije of tin' Loirrr Lykrvs (Ik'isioii. Neihoi'tkris Pocahontas var. pextias D. W. (PageSiVi.i F'ig. o. rppcr pimiic and succeeding sublobate ]iiniiulcs. Fnini the roof nf Lykens coals 5 or (i at the Brookside colliery. Fig. 'vi. Sublobate pinnule, enlarged to show close, relatively i>aralk'I nervation. Three times the natural size. Fit:. -i/'. Fragment from the lamina of the same jnnnule. I'jilarLred to illustrate the interneural striation. Lower Li/ktns (lirision — Loirer zone. 922 PLATE CXC. 923 PLATE CXC. AXEIMITES POrrSVILLENSIS D. W. (Page 868.) Figs. 1 atid 2. The frasiiieiit? show the small cuneate ujiper pinnules and the larger Adiantitoid form. From the roof of Lykens coal Xo. 4, at the Lincoln colliery. Lower Lykens division — Mariopteris pottsvillea zone. Mariopteris pottsvillea D. W. (Page 874.) Figs. 3, 4, 5, and 6. Figs. 3 and 4 represent the typical developments of the pinnpe and pinnules of this well-marked species; fig. 5 shows the apex of a compound pinna, while fig. 6 represents a fragment in which the pinnules are much more than usually close. Specimens from the roof of Lykens coal No. 4, at the Lincoln colliery. Fig. ?ta. Enlarged detail of one of the pinn?e shown in fig. 3, illustrating the lobes and nervation. Twice the natural size. Fig. 4(/. Similar detail fiom the original of fig. 4. The fossil is somewhat distorted by pressure in the shales. Twi(;e the natural size. Lower Lykens division — Mariopteris pottsvillea zone. Neitrofferis Pocahoxtas var. ix.equalis D. W. (Page 890.) Figs. 7 and 8. These specimens show the elongated pinnules with Callipteridioid bases, such as occur in the middle and upper portions of the pinnje. ■ The specimen shown in fig. 7 is from bed D, 710 feet below the Twin coal in the gap at Pottsville; the original of fig. 8 is from the Kalmia colliery. Lower Lykens division. WniTTLESEVA Camprelli D. W. (Page 905.) Figs. 9, 10, and 11. These specimens show the characteristic aspect of the fasicular rilis terminating in blunt, often obscure teeth, and the very slender jjctioles of the leaves of this species. The originals of figs. 9 and 11 are from beds H and D, respectively, in the section of the gap at Pottsville; the specimen shown in fig. 10 comes from the roof of Lykens coal No. 5 or No. (5 at the Lincoln colliery. Lower and Upper Lykens divisions. 924 U. S. GEOLOGICAL SURVEY TWENTIETH ANNUAL REPORT PART II PL. CXC ANEIMITES, MARIOPTERIS. NEUROPTERIS, AND WHITTLESEYA. Figs. 1-8. Lower Lykens division. PLATE CXCI, 925 PLATE CXCI. NErijoiTEKis PocAnoNTAs var. ix^qialis D. W^ ( Page 890. ) FigH. 1-4. Fip?. 1 and 2 nhow the characteristic development of the pinnules in the small pinna\ Fig. 2a shows a pinnule from the original of fig. 2, enlarged to twice the natural size, to illustrate the nervation. Fig. 3 includes pinnules with the elongated form, constricted at the base, immediately above the pinnatifid pinnules. The specimen shown in fig. 4 is drawn twice the natural size, to show the characteristic out- line of the base of the pinnule and the nervation. The original of fig. 1 is from the drift on the east side of Rausch Gap; that of fig. 2 is from the Kalmia colliery; that of fig. 3 was obtained from the roof of Lykens coal No. 4 at the Lincoln mine; the original of fig. 4 comes from bed D, 710 feet below the Twin coal, in the gap at I'ottsville. Loinr Li/kciif: (llrixioii. NElItOPTKRIS PoCAUOXTAS D. W. (Page 888.) Fig. 5. This specimen shows the smaller upper lateral pinnje, suggestive in form and size of the Xeuropteris SniUlisil Lx. The lateral pinnules are, however, clearly seen to be broadly attached to the rachis, the nervation being Odontopteroid, the midrib scarcely developed. From the roof of the Pocahontas coal at Gilliam, West Virginia, Pocahontas quadrangle. Fig. 5a. Enlarged detail of pinna in the original of fig. 5, showing the form of the piniuiU's and the nervation. Twice the natural size. A'/i' (if till' Lnin I- [jikois (Uvixinn. SrillCNOI'lIVLLl'M TEXl'E I >. W. (Page 901.) Figs. 6 and 7. The figures show the slender, narrowed liases f)f the pinnules and the round-truncate, crenulate, distal margins. A fragment of one of the deeply dissected leaves from a verticel at the point of ramifi- cation is shown in fig. (5. The original of fig. 6 is from the Clark formation at Smith's Store, Virginia, Pocahontas (juadrangle; that of fig. 7 is from the Pottsville at Warrior, Alabama (No. 8501 of the Lacoe collection. United States National Museum). Loifir Lifkcns dlvimon — M(irU)]}li'rix pnttfo'lllea zone. TUKioNCUAKl'l-.M A.MP11J..EFORME Lx. Var. SIMXTABILE D. W. (Page 909.) Fig. S. This specimen shows well the thin, rather broad wings, the elongated micro- pylar neck, and the faint subordinate t"st;r. The spt'cimeii is from the Linciiln collit'ry. I'jijtf'r Lifkcnx ilivtsioii. 926 U. S. GEOLOGICAL SURVEY TWENTIETH ANNUAL REPORT PART II PL. CXCI NEUROPTERIS, SPH ENOPH YLLU M , AND TR IGONOCARPU M. Figs. 1-7. Lower Lykens division. PLATE (^XCII. PLATE CXCII. ErEMiMTERIS LIN'COLNIAXA I). W. (Page 869.) Fig. 1. Fragment imperfectly fallowing the devel<)])inent i>l the pinn?e and pinnules in the lower lateral pinnpe. From the roof of Lyken.* coal N(j. 2 (?) at the New Lincoln mine. Fig. la. Enlarged detail from the name specimen, showing the nervation. Twice the natural size. Upper Lykens division. IMaKU)PTERIS PYGM.EA D. W. (Page 876.) Figs. 2, ;], 4, r>. and 6. These specimens show the ordinary form and aspect of the pinnfe and pinnules of this species. The originals of fig.'^. 2, 4, 5, and 6 are from the horizon of Lykens coal Xo. 2 at the Xew Lincoln colliery; the original of fig. 3 is from nearly the same horizon, at about 500 feet below the Twin coal, in the Pottsville Gap. Figs. 4a and (ia. Enlarged details of pinnules of the originals of tigs. 4 and 6, showing the sul:)divisiou of the pinnules and the nervation. Twice the natural size. L'pper Lykens divisio)). Al-ETHOPTEKIS EvAXSII Lx. ' Figs. 7 and S. Fig. 7 represents the teriuinal ixirtion <>f the compound pinna, while in fig. S fragments of lateral ultimate pimue are seen. Fig. 7 is from the Lincoln colliery; fig. 8 was found in bed H, 550 feet beloAV the Twin coal, in the gap at Pottsville. Fig. 7a. Enlarged detail of pinnule shown in fig. 7. Twice the natural size. Fig. 8a. Detail of pinnule from the original of fig. 8. showing the nervation an 3 '"**9*^t ,\ ^' EREMOPTERIS, MARIOPTERIS, AND ALETHO PTER IS, Upper Lykens division. PLATE CXCIII. 20 GEOL, FT •> 59 9-^9 PLATE CXCIII. Alktiioi'tkkis I>\(()i;i 1). W. (F'age .S.H1. ) Fi^. 1 and 2. Tvpii-^l fragiufiit.s »»f the piniiic of thin spec-it's. From tlio. roof of Lykens coal No. 2 at the lower Eureka tunnel. Fijr. la. Enlarged pinnule from the fragment shown on the left of fig. 1, illustrating the nervation. Twice the natural size. I ji/ti r Liih'iix (lirisioii — St'innirc zniic. NeIROITEKI.S I-INATA I >. \V. ( Page 895. ) Figs. :!-7. Specimens showing the variation in form and size of the pinnules of this species. The originals of figs. 3, 4, and 5 are from the roof of L}'kens coal No. 1 at the Lincoln colliery; the type of fig. 6 is from the rock dump at the same colliery; that of fig. 7 is from i)roV)a1)ly the same horizon at the New liincohi colliery. Figs. ."ia, (ia. 7a. Enlarged details from the originals of tigs. 5, 6, and 7, showing the thin, outward-curved nervation. Twice the natural size. I'p/xr Li/kni.f dirisiou. SpHENOIMIVLMM TENEKKIMIM Ett. Var. KU)NG.\TrM D. W. ( Pago S9. 2. Fig. Sa. Enlarged detail, showing the division and n^-rvation of a young katlet of this species. Twice th? natural size. Ijijii r Li/kriix (lirixioii. (_'.\KniO(AUI'OX COKXITIM Du. I Piigi--9as.) Fiii. 1". Ordinary example; from lii . - ^-^ p. y 1 r ,: ^'T' » 1 Jr^*^^ \ 1 . m 4 k'^