yE929 .B432 Berry , Edward Wilber The physical conditions ind: icated by the flora of the Calvert formation / si DEPARTMENT OF THE INTERIOR Franklin K. Lane, Secretary UNITED STATES GEOLOGICAL SURVEY George Otis Smith, Director Professional Paper 98— F THE PHYSICAL CONDITIONS INDICATED BY THE FLORA OF THE CALVERT FORMATION BT EDWARD WILBER BERRY Published Ma; 27, 1916 Shorter contributions to general geology, 1916 (Pages 61-73) WASHINGTON GOVERNMENT PKINTING OFFICE 1916 DEPARTMENT OF THE INTERIOR Fraxklix K. Lane, Secretary United states geological Survey George Otis Smith, Director Professional Paper 98 — F THE PHYSICAL COXDITIOXS INDICATED BY THE FLORA OF THE CALVERT F0R:\LVTI0X BY EDWARD WILBER BERRY Published May 27. 1916 Shorter contributions to general geology, 1916 (Pages 61-73) WASHINGTON GOVERXMEXT I'KIXTIXG OFFICE 1916 CONTENTS. Page. Introduction 61 Character and extent of the Calvert formation 61 Fauna of the Calvert formation 62 Flora of the Calvert formation 62 Composition and enwonment 62 Probable age 66 Conclusions 66 New species of plants 66 ILLUSTRATIONS. Plates XI. XII. Fossil plants from the Calvert formation 72-73 II THE PHYSICAL CONDITIUXS INDICATED BY THE FLORA OF THE CALVERT FORMATION . By Edwaiu) Wh.heij ]5erey. INTRODUCTION. The object of till' present paper is to give a suminarv of the sniull flora preserved in the Miocene diatomaceoiis beds vi the Calvert for- mation in the District of Columbia and Virginia, and more especiidly to discuss its l)earing on the physical conchtions of the Calvert epoch. Sub- sequent to the middle Eocene the next abun- dant marine fauna preserved along tlie midtUe Atlantic coast is that of the Calvert formation of the Chesapeake group. ^Vlthough Miocene faunas so low in the stratigrapliic column are known south of Virginia only in the vicinity of Porters Landing, Savannah River, Ga., closely related but younger Miocene faunas extend southward at least as far as Florida, where the containing formation rests unconformably on bods in which occur the warmer-water faunas of the Apalachicola group. The change from the Apalachicola faunas to those of the Miocene constitutes one of the most striking faunal changes of tlie later Tertiary- in southeastern North America, and its emphasis by Diill and others has led to what I believe to be a misconception of the rctil climatic condi- tions of Calvert time. Likewise from the fact that the conspicuous deposits of diatomaceous ooze in existing marine waters are in the polar oceans it has been unscientifically assumed that the diatomaceous beds so cliaracteristic of the Calvert formation must have been formed by species which had a more or less comparable environment. CHARACTER AND EXTENT OF THE CALVERT FORMATION. The Calvert foi-niation, named l)y Shattuck ' in 1902 from the Cidvert Clill's, in Calvert County, Md., consists of 200 to 400 feet of ' Shattuck, G. B., The Miocene [onnation of Maryland: Science, npw ser., vol. 13, p. 906, 1902. 2.s924°— IG diatomaceoxis cartli, sandy clays, and marls. It is tyj)ic:dly developed in Maryland, wli(>re the basal member (Fairhaven) comprises about 65 feet of diatomaceous earth, locally argillaceous and showing occasional infhixes of sands. The Calvert has been recognized in southern New Jersey and Delaware. It is also well repre- sented in Virginia but disappeai-s in the south- ern part of that State by tlie transgression of the St. Marys formation, which complctel}- buries it in Nortli Carolina, overlajjping on the crystalline rocks of the Piedmont Plateau in some areas. FossU land jdants have been found in the Calvert at Richmond, Va., in the District of Columbia along the Penning road near the District line, and at Good Hope IIQl. The formation crops out on the Good Hope road at a point near the top of the hUl, where a few feet of light-colored daj^ (Calvert) rests unconform- ably on the fossdiferous beds of the Magothy formation (Upper Cretaceous) . The Calvert is overlain by gravels of late Tertiary or early Pleistocene age. This locality is one of the most northwesterly points at which the Calvert can be recognized with certainty, and it must also have been near the hxndward margin of the Calvert sea. No invertebrates are associated with the plants at tliis outci-op. At the Richmond locality the Calvert con- sists of very clayej^ diatomaceous earth 40 to 50 feet in thickness, which rests unconformably upon remnants of the Eo(u>ne or ujxm the underlying Lower Cretaceous or crj-staDine rocks and is overlain by Pleistocene deposits. That this locality also was lu^ar the shore lino of tlie Calvert sea, as it is near the landward limit of the existing Calvert deposits, is iny the more claj-ey nature of the materials compared with similar diatomaceous deposits elsewhere in the ('alvei-t formation but 62 SHORTER CONTRIBUTIONS TO GENERAL GEOLOGY, 1916. by tho containpfl pLuit fossils, as well as by considerable coiiimimited lignite, tliat in places forms layei-s 5 to 12 millimeters thick. The Calvert contains marine invertt^brates in the immediate vicinity of this outcrop, and asso- ciateil with the plants are Nassa peraltoides Martin and Disci niscd htguhtis (Conrad), the latter a species whicli occurs in shallow water free from sand. FAUNA OF THE CALVERT FORMATION. The fauna of the Calvert ft»rmation is exten- sive, particularly in the marls that overlie the diatomaceous beds. In addition to the re- mains of marine vertebrates, which are in gen- eral wide-ranging and of little significance in ecology, there is a considerable molluscan fauna. From the Calvert of Maryland 2.35 species of Mollusca have been described. About 100 species have been recorded from the Vir- ffinia extension of these beds, but this mmiber will be considerably increased when the studies of Dr. J. A. Gardner are published. Twenty- tliree of the Calvert species have been identified in the Florida Miocene (Jacksonville and Choc- tawhatchee formations), where 40 per cent of the fauna is common to the Chesapeake group, resembling more specificallj' the faunas of the Miocene formations younger than the Calvert. From an analyses of these Maryland faunas Dall,' a few j-ears ago, concluded that they in- dicated shallow marine waters and temperatures somewhat warmer than those of the present time in the latitude of Maryland. Of the species that persist in the existing fauna, the bulk are found south of Cape Hatteras. I see no reason for doubting Ball's conclusions, which obviously preclude the ' ' cold ' ' or even the ' ' cool tempei-- ate" conditions that have commonly been considered as characteristic of the Chesapeake group. The United States hydrographic charts give the mean temperatiu-e of the surface water of the Atlantic in the latitude of Maryland from December to May as 55°, and from June to No- vember as 71.5°. The figures for Massachu- setts Bay are 45° and 63.5°, respectively. The ^ Dall, W. H., Maryland Geol. Survey, Miocene, pp. cxxix-clv, 1904. figiircs for the vicinity of Beaufort, X. C, which is immediately south of Cape Hatteras and which would appear to afford conditions cor- responding to Dall's qualitative estimate, are 65+ ° for Decemlicr to May and 76° for June to November. If Dall's estimate is reasonably correct it would indicate that the mean annual temperature of the water in which the Calvert molluscan fauna lived was close to 70°. The mean for the year at the siu'face off the Mary- land coast is 63°, and at Beaufort it is 70°, and although the liottom temperatures are slightly l(^ss than the surface tcmpiTaturcs they are more unifoi'm and for shallow seas like the Cal- vert would not be appreciably lower. FLORA OF THE CALVERT FORMATION. COMPOSITION AND ENVIRONMENT. The most abundant plant remains in the Cal- vert are the siliceous tests of diatoms. These have been much studied by diatomists, who have identified a very large number of species. I am indebted to Dr. Albert Mann for the state- ment that the Calvert diatom flora indicates a comparatively shallow, strictly marine habi- tat with relatively wann or subtropical tem- perature. The remains of land plants preserved in the strictly marine beds of the Calvert arc few both in number of species and in number of individ- uals. As a result of the present study I am able to enumerate 26 species — 16 from Virginia and 17 from the District of Columbia, with 7 forms common to the two areas. The flora in- cludes 1 pteridophyte (Salvinia), 2 conifers (Taxodium and Pinus), and 23 dicotyledons. Leguminous forms and oaks predominate, al- though the cypress (Taxodium) is by far the most common form at Richmond, where cone scales and seeds as well as an abundance of the deciduous twigs are preserved. Although this flora is so small it fortunately comprises forms whose characters indicate very clearly the physical conditions along the shore of the Calvert sea. The accompanying table indicates the most similar fossil and existing species : PHYSICAL COXDITIOXS IXDIOATED BY THE FLORA OF THE CALVERT FOKMATIOX. G3 o 0 03 w s C3 •E o w i , o a ■ 3 ■c .9 ■a Q e c 1^ 'j^ Ti « c 3 33 5: 3 o u O J2 3 c3 "5) „ T3 ►? s « « •S 5 TO 3 - Ci c ^ c3 >^ 1^ o £ H > 1 r- -7- ^ 7i fer- -= -SS ■?££ 3 C C C3 X ci . -•<; ^* oi'C •S 2 Tr = «•£ ^'^ 5^ « c3 s s a 3. >^ X "^ w B 3 3 -1. .2 c -2 2 9"° S 9 3 •I S.S CZ3 &H o . t-l £ E 13 S 3 3 O'O' 3 o 5 C3 Pi o X Ph '^^ 7, >>— 3 1^ c9 X X X S Ph O 5 +-" 3 =3 w a £ W :w d a; P3 £ O - --- "S-C o cZ.~ c ~ P C3 ii _^ H H rms referred b Quercus myrti nger. >-> ;3 a g ^c^ x O t. S ® o3 ■5 rt yi ,^ o C5 Ed •d 0) < O ■£=3 O ^ m rf 5 fc- E E X X X X X « 3 X* — d d .^ d Si a d 5 « H "o "o 3 3 d d ■3 -2 o c o o «-3 2 X 1^ « ^- 0) 0 t, j: q; ~ 2 c 3 0 ST i^ "^ 3 ^ S-^ 3 " c; '"-0 -4-9 d -3 3 c ^P. 1 J '3 ^ ^. •aiid ,^S d f.r u -T""" d l«l "S a if =1 fi^< 0 0 cu J OPh XX 3 3 •^ '§ c 0 t 0 ^ > 0 a 0 X X XX X d 3 3 T3 d 3 d pa.ii 3- « 5 t: d ?> a •5-3 d o 0) » w ao- m .0 d a a X X X X X be c 3 d M fl >? !ji w ■2 r tj 6r '^ c fc: a) pa 3 O a e pa K. d o o o H 08 >• -5 o S ■30 § a d iJ .3 fi § S .S3 ■& -e; .^ -s j: pl. Ph Ph 2 5 ■ d o ^ 2 II o 64 SHORTER CONTRIBUTIONS TO GENERAL GEOLOGY, 1916. I C :-< 7- rf X V Q; cj n ^ -- c3 _o >1 "p - -1 -2 1 aJ 1 §" c3 C3 ■? _ o »< -ij S m > -a *^ > iL> a s d '^ 2S ^ .1 . o o 6C w c cs 5 o 1 •a J r° is /^-^— < ■ ■ i : , ' ^ 03 ;-. • ', S ! ^ H : O H v^^ : • .-* ;ii; 1 a ,3 2 a : s 3 o : 3'S : o a 1 ■c ■g * >1 1 1 .3 -2 a . '^ Hi 0) — ' >i 'S tn ■fe si- : li i aSSrd : ^ ffl e^ s :j u : g ^ III 0 • -§ : d cS 3 O =^ . o mw £ -n A K ■ N -*,^o o^ O *4-i 'C o o^ a o ^ o ■ O M c rt o 0.2 >> S g S •a : i*-i o o-S ■set; '=£ '"S -§ • O O OJ Co Q^ o : K HHH b c OJ a o3 g m ti a> 'go . s &.s$ W ^ "S^w? a ^ a 3 TO -' ^_ (D 3 ^ 6 a (S r^' --J . a a cw ■a §2 .2 .3 s; '-e a a J is (B (D JT; == © — >. KM Z a ^3 1 fC -d g XX X X tf a J,'^ a J en X : X X x>< > (- o. : x' . o >i >> :5 s 0) § ^^ •'S >, 1 > 33 !sa a C5 QJ ^S 1 s > o o .2 Ma B ^^ I- c >l o c a t 2, a .3 '3 ■si a" i^ ^ S r^ > a 5 3-3 S Ei I 1 1 I ,„ « S o j: cs .3 a P 3 S «j > 1 a PC - ^ 5 ;^ Q *-> -1 5 c \ J 3 ■a > 3 C8 60 3 •a 3 ■S 3 a PHYSICAL CONDITIONS INDICATED BY THE FLORA OF THE CALVKltT IdRMATlOX. 65 Among the recent species with which the Calvert forms arc compared three range as far north as soutliern Canada and one other ranges into New Enghvnd. On the other hand, all the forms, inidnding the fonr jnst mentioned, range southward to Florida or Texas and five of these extend into the Ameri- can Tropics. It is obvious that these wide- ranging forms are of slight value hi an en- deavor to estimate the climatic conditions along the coast in Calvert time. More sig- nificance attach(>s to the northern limits of the other existing forms with which those of the Calvert arc compared. Three of these have their northern limits in southern Ari- zona and New Mexico, two in Georgia, two in North Carolina, four in Virginia, one in Maryland, and two in Delaware. Some of the comparisons with recent forms are closer or more significant than others. Among thes(> may be cited those with Taxodiurn distichum, Quercus chapmani, Planera aqiiafica, Berclumia scandens, and Pieris niiida. Of these, tlie bald cypress (Taxodiurn distichum) reaches its north- ern limit in soutliern Delaware; Quercus chaji- rnani is confined to the coast of Georgia and the Gulf of Mexico ; Planera reaches northward to the Coastal Plain of North Carolina; and Berchcmia and Pieris extend to the Coastal Plain of Virginia. It is obvious that if these comparisons arc legitimate the Calvert flora would find its most favoral)le conditions for existence along the present south Atlantic and Gulf coasts. Another method of approach is furnished by comparing the Calvert flora with the contem- poraneous floras of Europe. Of the 26 Cal- vert species, 7 are identical with and 10 are extremely (dose to forms of the Tortonian of Europe. The Tortonian has a very extensive flora, and as a result of the elaborate analysis by Ileer it was considered to indicate a mean annual temperature of 6.5°. Regarding the specific environment of the difTerent Calvert species, I consider the follow- ing as denizens of river or estuary swamps: Fraxinus richmondensis. Finns sp. Salvinia I'ormosa. Taxodiurn duhium. Nyssa gracili.s. Tlie following may be considered as dwellei on river bars or behind coastal sand dunes: Salix raeana. Ulmus basirordata. Cassia torafonuis. Keris scrobiculata. Platanus areroides. Bercheinia priscaformis. Carpinus grandis. Those which probably flourished in low l)ot- toms, on the lower flood plains of streams, are Carpimi.'^ grandis. Planer.i uiigeri. Platanus aceroides. Berchemia priscaformis. Most of these could readily live also in river or estuarj- swamps. The species which may legitimately be con- sidered to have been inhabitants of the strand or of coastal dunes are Podogonium ^•irginianum. I^eguiuinosite.s falvertensis. Ficus richmondensis. Pieris scrobiculata. Quercus calvertonensis. Quercus Ichnianni. Quercus chapniauilnlia. Phyllites < ercocarpifolia. Csesalpinia ovali folia. Cassia toraformis. Dalbergia calvertensis. Rhus milleri. Ilex calvertensis. Vaccinium cf. V. textum. Celiistrus bruckmanni. Pinus sp. AU but Salvinia and Cassia were probably arborescent, and Cassia may have been. All the dicotyledons except Fraxinus, Carpinus, Platanus, Ficus, and Quercus calvertonensis show marked reduction of the leaf lamina-, clearly indicating a sandy habitat or a swamp habitat, the latter being physiologically dry and in its effects much like the former. The leaves of the oaks, Berchemia, Podogo- nium, PhyUitcs cercocarpifolia, Dalbergia, Rhus, Ilex, Vaccinium, and Celastrus are coriaceous or subcoriaceous, and those of the Pieris are tomentose, both features tending to prevent transpiration and indicating a physiologically dry habitat as well as exposure to abundant sunshine. In comparing the two localities, that at Richmond clcarlj- indicates a low coast which was lined with cypress swamps and in which the very inconsiderable run-off carried only the finest muds. That in the District of Co- lumbia, in its small-leafed oaks. Ilex, Vacci- nium, Pieris, and abundance of Leguminosa^, as clearly indicates a region of dunes compara- ble with the present Santa Rosa Peninsula, be- tween Pensacola Bay and the Gulf of Mexico. Physical conditions may produce decided differences in the flora of the continental mar- gin and the marine life of the sea which washes it, as for example in the English Pliocene, where the only avenue of immigration for terrestrial plants and fresh-water mullusks was from the south, while the onl}' avenue of immigration for the marine fauna lay in the cool currents that entered the North Sea from the north. However, the Calvert fossils do not represent a northward extension of tlie south Atlantic and 66 SHOBTEE COXTEIBUTIOXS TO GENERAL GEOLOfiY, 1016. Gulf flora along the Coastal Plain and a soutli- wai'd migrat ion of a marine shaUow-water fauna in the opposite direction along the coast, but both floral and faunal evidence are in accord. In this connection it is of interest to caU attention to the fact that five of the Calvert plants occur m the late Eocene or Oligoccne of western Greenland, in latitude 70° N. TliLs Arctic flora I believe to. be contemporaneous with those more southern floras of the late Eocene (Jackson) or early Oligocene (Catahoula and Vicksburg) which are most tropical in their facies. It seems also to be established that with the gradual lowering of temperatures during later Tertiary time this circumpolar flora spread southward over North America and Eurasia. PROBABLE AGE. Seven of the Calvert plants, or 26.9 per cent, are common to the Tortonian of Europe, and ten othei-s, or 38 per cent, are represented in the Tortonian by very similar forms. In view of tlie fact that these floras spread into both regions from a common and equally accessible source, as I haye just stated, the evidence that the Calvert flora indicates a Tortonian age is as conclusive as intercontinental correlations can ever be. Compared with other American floras of Miocene age, that of the Calvert has little in common with the described Miocene floras from Colorado, Idaho, Oregon, or Cahfornia, which are all lake or river vaUey floras of moist up- land forest types. CONCLUSIONS. The Calvert flora was a coastal flora of strik- ingly warm-temperate affinities, comparable with the existing coastal floras of Soutli Caro- lina and Georgia along the south Atlantic coast or with those along the coast of the Gulf of Mexico from western Florida to eastern Texas. The general climatic features it indicates are in accord with those which may be legitimately deduced from the evidence of the marine faunas. The climate of the Chesapeake Miocene epoch, cooler undoubtedly than that of the Apalachicola or preceding epochs, was neither cold nor cool temperate. The age indicated by the Calvert flora is middle Miocene, or in terms of European geology Tortonian. NEW SPECIES OF PL.4NTS. Class GYMNOSPERMiE. Order CONIFERALES. Family PINACE.ffi. Genus PINUS Linne. Pinus sp. Plate XII, figure 1. A somewhat macerated seed of Pinus of the Pinus txda ty])e occius at Good Hope Hill. The material is insufficient for characterization. Pinus txda Linne ranges from Delaware southward along the coast to Texas and extends up the Mississippi Valley to Temiessee and Arkansas. Class ANGIOSPERMjE. Order FAGALES. Family FAGACE.S:. Genus QUERCTTS Linne. Quercus chapmanifolia Berry, n. sp. Plate XI. figures 1.2. Leaves small, oV>long-ol)ovate in general out- Ime, with a rounded apex and a wide or nar- rowly cuneate, ultimately slightly rounded base. Length from 27 to 4.5 millimeters: maxi- mum widtli, at or above the middle, from 14 to 30 millimeters. Margms entire, slightly undulate. The lateral margins ascend to or above the midtUe, where they curve to form irregular and unequaUy developed rounded lobes subtending usually shallow and open rounded sinuses. A second short and broadl}' rounded lobe may be developed, usually ou only one side, in the apical region. Petiole short and stout enlarged proximatl, 2 to 3 miUi- meters in length. Midrib stout, usual!}- curved or flexuous, prominent on the lower surface of the leaf. Secondaries of varying caliber de- pendent on the extent to which lobes are de- veloped; in the prominently lobed specimen shown in figure 2 a subopposite pair are stout and prominent ou the lower surface of the leaf, where tlie lobation is feeble, as in figure 4. The secondaries are thin. Tliere are as many cras- pedodrome secondaries as there are lobes — two or three to a leaf; the rest of the secondaries are camptodrome; all are somewhat irregularly spaced and rather straight in their courses, PHYSICAL COXDITIOXS IXDICATED BY THE FLORA OF THE CALVERT FORMATIOX. their angles of divergence depending on the rel- ative width of the leaves. Texture coriaceous. This significant new species is rather clearly marked off among previously described fossil forms. Among existing forn^s its size and general outhne as well as its limits of variation are almost exactly those of Quercus chapmani Sargent. This may he readily seen without extended discussiim by the mk prints of the leaves of the latter species introduced for com- parison and shown in figures .3 to S of Plate XI. This similaritj" is, I believe, an indication of relationship, and in its light the habitat of the existing species becomes of special uiterest. Quercus cJiapmani is a small tree, or more com- monly a shrub, uihabitiog sandy barrens near the coast, of rare occurrence froni South Caro- lina to Florida along the Atlantic coast, reach- ing its maximum of abundance and develop- ment along the GuK coast of western Florida from the shores of Tampa Bay to Pensacola. The leaves figured come from the Santa Rosa Peninsula, where it is typicaU^- developed and where the environment is compara])le with that of Miocene time along the shores of the shallow Calvert sea. Occurrence: Good Hope road, Anacostia Heights, D. C. Quercus lehmanni HoUick. Plate XI. figures 9-n. Quercus Uhmanni. Hollick, Miocene, p. 483, figs, la, lb, Man-land Geol. .Sur\-ey. 1904. This species was briefly described by Hol- lick in 1904. Recent material collected from the tjT)e locahty enables me to give the follow- ing somewhat fuller characterization: Leaves small and narrow, oblong in general outhne. Length from 21 to 38 millimeters: maximum width, across the median marginal lobes, from 14 to 17 millimeters. Apex conically pointed. Base broadly rounded, generally somewhat inequilateral. Margins with one to three irregularly spaced reduced lobes on each side: these lobes are more or less developed, short, and conical and are little more than coarse teeth; they subtend usually open sinuses, although in some specimens the lobes are directed upward instead of outward and the sinuses are correspondingly narrow. Petiole short and stout. Midrib stout, straight, or curved, prominent on the lower surface of the leaf. Secondaries tliin, two t« four pairs: thej' diverge from tlie midrib at angle- of about 45' and ascend in straight or some- what fle.xuous courses: a craspedodrome sec- ondary runs to the tip of each lobe, and a basal one on each side arches along the lower lateral margin. Tertiaries obsolete. Texture coria- ceous. The lobes are irregularly spacetl: if there is but one on one side there are usually two on the other, and if there are two on one side there are tisually three on the other: they may all be above the middle of the leaf, as in figure 9. but often there is a small obhque one lower down, as in figures 10 and 11. Midway between the apex and the base, where the leaf is consequently widest, they are subopposite, as are the correspondmg secondaries. These characteristic leaves are not uncom- mon, but like most of the pLint material in the Calvert formation they are tisually much broken. Their size, form, and texture are indicative of barren soil, bright sunshine, and sparse rainfall. Compared with existing spe- cies they are found to resemble the toothed leaves sometimes developed on Quercmt inr- ffiniana Miller, a coastal species of southeastern North America, which is also abundant in the Pleistocene and which in the form of Quercus prei-irglniana Berry is exceedhigly abundant in the Phocene of the Gulf coast. Most of the leaves of Quercus virginiana are oblong, elhptical, or oboviite with merely undiUate margins, and the significance of the occasional toothed leaves is tmknown. A modem species with which Quercus lehmanni may be more legitimately compared is Quercus emoryi Torrey. The latter has the small narrow leaves with rounded bases and irregular teeth. TheA^ differ somewhat from the fossil leaves m their proportions, having generally a broader base, but many leaves can be selected that exactly match the fossil. Quercus emoryi is a stout tree of the uplands of western Texas and southern Xcw Mexico and Arizona, and although it is not necessarily directly afiihated with this iliocene form of the Atlantic coast, it is not improbable that the two forms have a common ancestor which once flotu-ished in an intermediate area. Occurrence: Good Hope road, Anacostia i Heights, D. C. 68 SnOETHE CONTRIBUTIONS TO GENERAL CEOI.OGY, 1916. Order URTICALES. Family ULMACEjE. Genus ULMUS Liim6. Ulmus basicordata Hollick. Plate XII, figure 2. Ulmns basicordata. Uollick, Miocene, p. 484, fig. If, >rar>-Uinf Ulmus minuta, from the Tortonian of Baden, on the Swiss border, de- scribed by Heer.' Occurrence: Richmond, Va., and Good Hope Hill, D. C. Order ROSALES. Family ROSACEJE. Genus PHYLLITES of authors. Phyllites cercocarpifolia Berry, n. sp. Plate XII, figiire.s 3, 4. Leaves small, obovate, with a broadly rounded tip and a cuneate base. Length about 6 millimeters: maximum width, slightly above the middle, about 3.5 millimeters. Margins en- tire, possibly shghtly revolute. Textui-e dis- tinctly coriaceous. Petiole short and stout, 1 Ettingshausen, Constantin, Fossiie Flora von Wien, p. li, pi. 2, flgs. 5-lS, 1S51. 2 Goeppert, H. R., Tertiiirc Flora von Schossnitz in Schlesien, p. 31, pi. 14, figs. 12-14, 1855. 5 Heer, Oswald, Flora tertiaria llelvetiae, vol 2, p. 59, pi. 79, figs. 9-13, 1856; vol. 3, p. 181. pi. 151, fig. 30, 1859. less than 1 milhmeter in length. Midrib rela- tively stout and promment. Secondaries thm but prominent, numerous, equally spaced and subparallel; about seven or eight subopposite pairs diverge from the midrib at acute angles and pursue a nearly straight ascending course to the margins, where they are abruptly camp- todrome. Tertiaries obsolete. Tills tmy leaf suggests a relationship with the genus Cercocarpus, of the Rosaceae, and the resemblance is so great that I feel justified in referring the fossil to that family. Cerco- ctxrpus is a small genus of about five species of trees and shrubs confined to the dry mterior and mountainous regions of the United States and Mexico. Tlie fossil greatly resembles the entire leaves of Cercocarpus brevifiorus Grtiy, a small tree of the pine and oak forests of the dry elevated ridges of southern Arizona, New Mexico, western Texas, and northern Mexico. Although I do not feel justified in referring the fossil to Cercocarpus without more conclusive evidence I beheve that the resemblance noted indicates such a relationship. Occurrence: Benning road, D. C, near District line. Family CffiSALPINIACEJE. Genus CASSIA Linne. Cassia toraformis Berry, n. sp. Plate XII, figures 6, 7. Leaflets small, sessile, inequilateral. Apex broad, ineciuilateral, slightly emarginate. Base V)road, .slightly pointed, inequilateral. ^largins entire. Leaf substance thin. Length about 15 millimeters; maximum width, slightly below the middle, 7.5 millimeters. Midrib of medium size, nearly straight. Secondaries thin, about seven pairs, diverging from the midrib at acute angles, those in one-half of the lamina more acute, sweeping upward in ascending sub- parallel curves, parallel with the lower lateral margins, camptodrome. Tertiaries thin, of the Cassia type. This species appears to be a new species of Cassia, although it iwAy represent an allied genus of the Ca?salpiniacese. Among the hundreds of existing species there are a number very similar to the fossil, which I have named from its resemblance to the existing Cassia tora Linne, an annual of sandy soils ranging PHYSICAL CONDITIONS INDICATED BY THE FLORA OF THE CALVERT FORMATION. 69 from Vir<;ini:i southward to I'lorida, Cuba, and tro])ic!d America. The fossil s[)eeies are numerous and range from tlie I'pper Cretaceous to the present. ^imong ])reviously described fossil forms Cassia ioj-aforniis is identical with some of the heaves from th(. Tortonian of Ooningen, Baden, which Heer ' refei-s to Cassia llgnUum Unger. It i.s also very similar to CxsaJpin'ta esdicri Heer,- from Ooningen. Occurrence: Good Hope Ilill, I). C. Order SAPINDALES. Family ILICACEjE. Genus ILEX T.inne. Lex calvertensis Berry, n. sp. Plate XII, figure 8. Leaves small, ovate in general outline, with a bluntly pointed apex and base, the latter slightly inequilateral. Length about 16. .5 mil- limeters; maximum width, midway between the ape.x and base, about 7 millimeters. Mar- gins entire for most of their length, apically with slight and remote denticulations. Texture subcoriaceous. Petiole short and stout, about 1.7.5 millimeters in length. Midrib stout and prominent proxiniad, attenuated distad, curved. Secondaries thin; five subopposite pairs diverge from the midrib ^^t wide angles and are conspicuously camptodromo in the marginal region. Tertiarj' venation obsolete. Ilex is a large genus with over 100 described fossil species and over 200 existing species. It ranges from the Upper Cretaceous to the pres- ent, and existing forms are found in all tropical and temperate regions except western North America, Australia, New Zealand, and New Guinea. The largest number occur in Brazil and Guiana. Hex calvertensis may be compared with the existing Hex romltoria Aiton, a small tree of the coastal region of the South Atlantic and Gulf States. It also resembles Flex I'omitoriafolia Berry, a coastal form of the lower Eocene (Wilcox group) of the Mississippi embayment, with which it may be affiliated. Occurrence : Good Hope Hill, D. C. ■ ITeer, Oswald, Flora tertiaria Uelvetiae, vol. 3, p. 121, pi. I.'i8, figs. 23-28, 1839. s Idem, p. Ul, pi. 155, fig. 21. Order KHAMNALES. Family KHAMNACEa:. Genus BERCHEMIA Necker. Berchcmia priscaformis Berry, n. sp. Plato XII. fi-ures 11, 12. Leaves small, lanceolate, with equally acumi- nate apex and base. Length about KJ milli- meters; maximum widfli l)etween apvx and the base, a])out 5.5 millimeters. Margins entire, sligiitly uiuhilate corresponding with the camp- todrome eiulings of the secondaries. Texture subcoriaceous. Petiole relatively long and slender, about 2 millimeters in length. Midrib relatively stout, somewhat curved, prominent on the lower surface of the leaf. .Secomhiries thin but prominent, numerous; seven or eight pairs diverge from midrib at regular intervals and acute angles, curving slightly and sul)- parallcl, camptodrorae close to the margins. Tertiarios mostly obsolete, a few percurrent ones seen. This characteristic little species is scarcely distinguishable from the smaller and more lanceolate leaves of Bcrchemia scandens (Hill) Trelease, a high-climbing shrub frequcntuig low woods from Virginia to Florida and from Missouri to Texas. The genus contains about 10 existuig species, the others being natives of Asia and tropical Africa. About half a dozen fossil species have been described. These include a rare form in the Aquitanian of France, another in the Burdigalian of Boliemia. species in the Pliocene of France anil llollaml, and tlie w'lAeiipvQ&A Bercliemiamultinenis (Alex. Braun) Heer. B. multinems has a recorded range from the Tongrian to the Astian throughout central and southern Europe. It has been iden- tified in the lower Eocene (Raton, Denver, and Fort Union formations) of the Rocky Moun- tain province and in the Miocene of Oregon. Tlie American material referred to B. multi- nervis is larger and wider than the Calvert species and lias a more pronounced tertiary venation. Some of the smaller leaves of B. multinerms from European localiti(>s are much like B. pnscaformis, as for example a leaf fig- ured by Heer ^ from the Tortonian of Oeningen, Baden, which is but slightly larger or wider. > Idem, pi. 123, fig. 17. 70 SHORTEE COXTEIBUTIOXS TO GEXERAL GEOLOGY, 1910. The Calvert specios is named from its resem- })lance to Bcrclumia prisca Saporta/ from the AquitaniaTi of Peyriae, Franee, a very simih^r but slightly mor(> oblong leaf. Occm-reneo: Ciotul Hope Hill, T). C. Order ERICALES. Family VACCINIACE^. Genus VACCINroM Liniie. Vaccinium cf. V. textuni Heer. Plate XII. ligm-es 14, 15. Several specimens of a small oval sub- coriaceous leaf are indistinguishable from Vaccinium textum Heer,^ from the Tortonian of Oeningcn, Baden. Heer's description is as follows: V. folius subroriaeeis, ovaUbus, integerriinis, apice ol)tusiusculis, basi rotiindatis: nervis secundariis campto- (Iruiuis, areis reticulatis. The Calvert form agrees with its European contemporary in size, general outline, and texture, as well as in secondary and tertiary venation characters. The only difference is the more acute base of the American form. The latter feature and the wide geographic separation of the two occurrences have pre- vented the direct reference of the American form to Pleer's species. ' Saporta, G. de. Etudes sur la vi^getation du sudest de la France k lV'i)oque tertiaire, vol. 2, p. 338, pi. 11, fig. 1, isijii. ! Hcer, Oswald, op. cit., p. 190, pi. 153, figs. 40-12. The genus contains al)out 125 existing species of wide geographic distribution, especially in the temperate and boreal regions of the Northern Hemisphere. It occurs, however, at high elevations in the Tropics, both north and south of the Equator, and evident!}' grew in intermediate areas at some past times. The fossil species are numerous and ranged from the Eocene onward, with their maximum display during the Miocene, when they were especially abundant in southern Europe along the shores of the Mediterranean Sea of that epoch. Occurrence: Good Hope Hill, D. C. Family EKICACEJE. Genus PIERIS Don. Pieris scrobiculata Hollick. Plate XII, figure Vi. Pieris scrohindata. Hollick, Miocene, p. 48(5, tf. Ig, Mary land Geol. Survey, 1904. This species was described by Hollick from material obtained at the Good Hope Hill locality. He compared it with the existing Pieris nitida (Bartram) Bentham and Hooker, a shrub of wet woods which ranges from south- eastern Virginia to Florida near the coast and is said to occur also in Cuba. Additional ma- terial from the type locality agrees with the type in outline and texture, but is only 11 millimeters in length and 5 millimeters in maximum width. Occurrence: Good Hope Hill, D. C PLATES. 71 PLATE XI. Figures 1, 2. Quercus chapmani/oUs Berry, (^iood Iloiie Hill, D. C. FiGUUES 3-8. Quernis chnpmani Sargent. A modern form from Santa Rosa Peninsula. Fla. Figures 9-U. Qucrcus khmanni Ilollirk. Good Hope Hill, D. <;'. (Figs. 9 and 11 after Hollick.) 72 U. S. GEOLOGICAL SURVEY PROFESSIONAL PAPER 98 PLATE XI O*. .4h \:>:^>m- ■'L 10 11 FOSSIL PLANTS FROM THE CALVERT FORMATION. U. S. GEOLOGICAL SURVEY PROFESSIONAL PAPER 98 PLATE XII FOSSIL PLANTS FROM THE CALVERT FORMATION. PLATE XTI. Figure 1. Pinus sp. Figure 2. Ulmus basiamtata HoUick. (.After Hnllick.) Figure 3. PhiiUUes cercocarpi/olia Berry. Figure 4. Same, X 3. Figures. Cxsalpinia ovali/olia'Sollick. (After Hollick. Figure 0. Cassia loraformis Berry. Figure 7. Same, X 3. Figure 8. Ilex calverlensis Berry. All but figure 10 from Good Hope Hill, D. C. Figure 9. Figure 10. Figure 11, Figure 12. Figure 13. Figure 14. Figure 15. Rhus milleri Ilollirk. Same, a more lobate form from Richmond \'a. Berchemia priscaformis Berry. Same, X 2. Pieris scrohiculala Hollick. (After Ilollirk.), Vaccinium cf. V. textum Heer. Tip of same to show venation. 73 O