Citation
Upper Cretaceous and Eocene floras of South Carolina and Georgia

Material Information

Title:
Upper Cretaceous and Eocene floras of South Carolina and Georgia
Series Title:
Professional paper (Geological Survey (U.S.) ; no. 84.
Alternate title:
Department of the Interior United States Geological Survey Professional Paper 84
Creator:
Berry, Edward Wilber, 1875-1945 ( author )
Place of Publication:
Washington, D.C.
Publisher:
Department of the Interior, United States Geological Survey, 1914
Government Printing Office
Language:
English
Physical Description:
200 pages

Subjects

Subjects / Keywords:
Palebotany -- South Carolina
Paleobotany -- Georgia
Paleobotany -- Cretaceous
Paleobotany -- Eocene
Genre:
government publication ( marcgt )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
This item is a work of the U.S. federal government and not subject to copyright pursuant to 17 U.S.C. §105.
Resource Identifier:
AAA1670 ( LTQF )
AME1934 ( NOTIS )
002436769 ( Old AlephBibNum )
035114247 ( New AlephBibNum )
Classification:
I 19.16:84 ( sudocs )

Full Text

DEPARTMENT OF THE INTERIOR
UNITED STATES GEOLOGICAL SURVEY
GEORGE OTIS SMITH, DmzECTOR

PROFESSIONAL. PAPZR 84


THE


UPPER CRETACEOUS AND EOCENE FLORAS


OF


SOUTH CAROLINA AND GEORGIA




BY


EDWARD WILBER BERRY





















WASHINGTON,
GOVERNMENT PRINTING OFFICE
1914


I* , ' , ' .
.* .j - -














SI T. 3

Usvo


.*. S.
* S5 * S*
S S
* S * S. *

55*
* S* 5 55 * .S.'
* 55 5 *
S *
















CONTENTS.


Cretaceous flora of South Carolina.............................................................. 5
duction................. .... ............. ............................... . .. ............ 6
to sketch..................................... ........................... ..... ..... .......... 5
y of the Cretaceous deposits of South Carolina .................................................. 7
Lower Cretaceous series...................... ..... .. ................................... ........ 7
Upper Cretaceous series............................................. ...................... ........ 7
Black Creek formation................- ...... ........................... .... ... ...... 7
Middendorf arkose member................................ ................. ............ 7
Other deposits ............................................................................ 8
Peedeesand ....................................................... ........... ... . ........... 8
Bcription of the plant localities.............................................................------------- 8
per Cretaceous geologic history of South Carolina.................................................... 12
thematicc description of the flora.................................................................... 14
inamic character of the flora .......................................................................... 63
ysical conditions indicated by the flora...----....-".................................. 66
elation of the Black Creek formation.............................................................. 68
paper Cretaceous flora of Georgia ..................................................................... 99
introduction ............................................ ..... .................... .. .. . ........... 99
ecology of the Cretaceous deposits of Georgia.......................................................... 99
. Ripley form ation .................................................... ..................... . ....... 100
Cuseta sand member .................................................... ..........o... 100
M iddle division ............................................................................... 100
Providence sand member.............................................. ....................... 100
Eutaw formation....................................... .............................. .... ......--------- 100
Lower division .................................. .... ......................... ....... .... 100
Tombigbee sand member .............................................. .......- ............. 100
Plant localities ......................................................... . ..... .... .............. 100
M cBrides Ford.................................................... ............ .. ............ 101
Broken Arrow Bend............................................... ......- ... .... ............. 101
Chimney Bluff.................................................. .......... .. ... .......... .. 101
. Near Buena Vista ................................................. ........................... 102
Near Byron.................................. .. ........... . .................. .......... 102
tic description of the flora .................................................................... 103
ic character of the flora.......................... ................. . .....--------- .... ................... 121
h cal conditions indicated by the flora ............................................................ 122
of the Mesozoic flora .........................--....... ................. ......................... 122
tion of the flora..................................................................... .......... 125
division of the Eutaw formation ............................................................... 125
y formation.................................. ......---------------------.... ....... ........ ................. 126
OCuseta sand member...........................------.... ----- ----............................... 126
eEocene flora of Georgia ............................................. .......................... 129
duction................................................................. . ................... 129
of the middle Eocene.................---------.. ------------.................................... 129
tic descri tion of the flora.................................................................... 132
florns of Europe and North America............................................................. 147
conditions indicated by the botanic character of the flora.... --.......-------------------------------........... 156
phy............................................ . .... ... . .......... ........... . .............. 161
3




p - - - -- ? 4- -




























ILLUSTRATIONS.

Page.

PLATE I. Plant-bearing clay lens in the Middendorf arkose member of the Black Creek formation:
A, Near Middendorf, Chesterfield County, S. C.; B, Near Langley, Aiken County, S. C ....... 8
II-X1V. Upper Cretaceous plants from South Carolina............................................... 73-98
XV. A, PlantlocalityatChimney Bluff, on Chattahoochee River, Ga.; B, Plantlocalityin theCusseta
sand member of the Ripley formation near Buena Vista, Ga................................ 102
XVI. A, Mangrove swamp along Miami River near Miami, Fla.; B, Habit of growth of mangroves
along M iami River .............................- ........- ... ... ........ ... ............. 160
XVII-XXIII. Upper Cretaceous plants from Georgia .................................................... 165-178
XXIV. Upper Cretaceous and Eocene plants from Georgia ........................................... 180
XXV-XXIX. Eocene plants from Georgia............................................................... 181-190
FIoURE 1. Restoration of Dewalquea smiihi Berry, from the Tuscaloosa formation of western Alabama ......... 41
2. Restoration of the leaf figured in Plate XXII ................--..........................---....... 115
3. Restoration of the leaf figured in Plate XXIII........................... . ................... 115
4. Sketch map showing the Cretaceous and Recent distribution of Tumion .......................... 123
5. Sketch map showing the Cretaceous occurrences of the Araucaries ................................ 123
6. Sketch map showing the existing distribution of the Araucarieve ................................... 124
7. Sketch map showing the Cretaceous and Recent distribution of Cinnamomum ..................... 124
8. Sketch map showing the Cretaceous and Recent distribution of Eucalyptus........................ 125
9. Diagrammatic geologic section of Phinizy Gully............--..............--............-- ...... 130
10. Restoration of Thrinax eocenica Berry..................................--..........--.......----....... 136
11. Sketch map showing the distribution of existing species of Rhizophora............................ 159
12. Sketch map showing the distribution of Nipa and Nipadites..................................... 160












'HE UPPER CRETACEOUS AND EOCENE FLORAS OF SOUTH
CAROLINA AND GEORGIA.


By EDWARD WILBER BERRY.


THE UPPER CRETACEOUS FLORA OF SOUTH CAROLINA.
INTRODUCTION.
rhe following report is the first systematic account of fossil plants from the State of
k Carolina. It describes a considerable flora, which clearly demonstrates the Upper
ceous age of the deposits in which it is found and which serves to correlate these deposits
the Upper Cretaceous of adjacent States both to the north and to the south.
the present study should be regarded as preliminary in nature, for it is probable that
kthe Coastal Plain of the State is exhaustively studied many new localities for fossil plants
discovered and many new species will be added to the Cretaceous flora; not uncommonly
Cretaceous floras consist of two or three hundred species. What is already known of
ra of the Middendorf arkose member of the Black Creek formation renders it certain
additional plant-bearing outcrops are discovered they will yield a large variety of
y preserved leaf impressions.
writer is under obligations to Mt. Earle Sloan, formerly State geologist of South Caro-
d to the United States National Museum for various collections; to Mr. T. W. Vaughan,
United States Geological Survey, under whom the work has been prosecuted; and
y to Mr. L. W. Stephenson, of the United States Geological Survey, for the care and
ce with which he has collected fossil plants in this area.

HISTORICAL SKETCH.
distinction of definitely recognizing the occurrence of the Cretaceous in North America
to Lardner Vanuxem,1 at one time professor of chemistry in the College of South Caro-
announced its presence in 1829, although John Finch five years earlier had pointed
part of the "alluvial formation" of Maclure was probably of newer "Secondary"
the paper announcing Vanuxem's thesis several Cretaceous localities are mentioned
Carolina, including Mars Bluff on Peedee River. A number of references to the Cre-
of the State are contained in the works of Morton, Hodge, and others. Lyell visited
in 1841-42 and recognized the Eocene age of certain calcareous rocks which had
been included in the Cretaceous.
Time of Lyell's visit, or a little later, geologists were well acquainted with the Cre-
a extending from Cape Fear River in North Carolina to Peedee and Lynches rivers
Carolina, as is witnessed by Henry D. Rogers's address before the Association of
Geologists and Naturalists at Washington in 1844.1 All the preceding contributions,
as well as the descriptions of Cretaceous deposits included by Prof. Tuomey in his
pf South Carolina, refer exclusively to the fossiliferous marine Cretaceous, which
cides with the present Peedee sand. The large area of Lower Cretaceous beds,
the deposits of most of the Black Creek formation, including the Middendorf arkose
Si dnlaer, Proc. Acad. Nat. Sie. Philadelphia, vol. 6, 1829, pp. 59-71; Am. Jour. Sei., let aer., vol. 16, 1829, pp. 2%-256.
b*h , Am. Jour. Set., 1st ser., vol. 7, 1824, pp. 31-43.
ES. D., Am. Iour. Sci., lst ser., vol. 47, 1844, p. 252.
5





6 UPPER OBRETACEOUS AND EOCENE FLORAS OF SOUTH CAROLINA AND GEORGIA.

member, were regarded by Tuomey as of Eocene age and are designated on his geologic map
"Sand hills" and "Red Hill area."
South Carolina is not mentioned in Prof. Ward's exhaustive paper on the geographic
distribution of fossil plants, published in 1889; ' nevertheless, a perusal of the older literature
shows that the earlier geologic workers occasionally encountered vegetable fossils. They
were content, however, merely to mention the fact or to identify the fossils with leaves of
trees in the existing flora which theywere thought to resemble. For example, Michael Tuomey,
in his admirable report on the geology of South Carolina,2 speaks of "impressions of leaves,
in a purplish clay" near Fort Motte on Congaree River and on the same page mentions
having encountered-
on the road, near the head branches of Halfway Swamp, thick beds of sand, containing water-worn nodules of marl
and a log of silicified wood, of considerable size. Immediately beneath this is a bed of yellow, tenacious clay, with
partings of fine sand and scales of mica. Between the laminem of the clay we found very distinct impressions of the
leaves of the oak, beech, and willow, with their most minute veins preserved. This, and the locality already
described, are the only ones known where fossil vegetable remains have been found in the Eocene of the United
States, with the exception of silicified wood and lignite, which are everywhere abundant.
This locality is in Orangeburg County and the horizon may possibly represent an Eocene
leaf-bearing layer instead of the Cretaceous, for the Claiborne Eocene transgresses the Upper
Cretaceous in northern Orangeburg County, and there are no recent records or collections of
leaf impressions from this area. On page 154 of the same work, and again on page 211, Tuomey
lists silicified wood, lignite, Quercus leaves, Fagus leaves, and Salix, erroneously classing the
containing deposits as lower Eocene in age.
Lieber in his first annual reports gives considerable attention to what he calls the "brown
coal of the Cheraws." He suggests its economic possibilities and records the bituminizedd
trunk of a tree in a lignite bed," figuring the trunk in his Plate II, figure 8. Lieber also
regards these Upper Cretaceous beds as of Eocene age.
More recently Darton made a brief reconnaissance across the State and in a short paper
published in 1895' records that "plant remains were observed at many points." He made
no collections, however, and mentions no localities, and the remark quoted may be taken
as one of a general nature. In 1904 Earle Sloan, until recently State geologist of South Caro-
lina, published a report on the clays of that State that shows a most intimate acquaintance
with all parts of its Coastal Plain. Fossil leaves are mentioned from three sections of the
State,5 near Aiken, near Middendorf, and at Rocky Point. This report is entirely economic
in character, and the author merely records the presence of "dicotyledonous leaves" or identifies
them from their resemblance to those of modern species as the leaves of the "elm, ash, cypress,
willow, bay, cane, and pine." The containing beds, however, are correctly identified as of
Cretaceous age. In 1907 the present writer5 recorded nine species of characteristic Upper
Cretaceous plants from Rocky Point, in Sumter County, contained in the collections of the
geological department of Johns Hopkins University. At that time the original collector was
not known, but it subsequently developed that the plants had been collected in 1895 by Prof.
L. C. Glenn.
In the spring of 1897 Profs. Ward and Glenn visited South Carolina and made large collec-
tions from the Rocky Point locality and a small collection from the locality near Darlington.
The important localities near Middendorf and in Aiken County had meanwhile been discovered
by Earle Sloan, who sent collections from these areas to Prof. Ward. These collections, how-
ever, were never studied. In 1906 the writer, in company with L. W. Stephenson, made a
canoe trip from Cheraw to the mouth of Peedee River, after which Mr. Stephenson spent con-
siderable time in the eastern part of the State. In 1907 the writer paid a brief visit to the
Congaree-Wateree area and collected from the Rocky Point locality. In a canoe trip down
I Ward, L. F., Eighth Arn. Rept. U. S. Geol. Survey, pt. 2, 1889, pp. 663-960.
2 Tuomey, Michael, Report on the geology of South Carolina, Columbia, 1848, p. 150.
' Lieber, 0. M., Report on the survey of South Carolina, 2d edition, 1858, p. 94.
4 Darton, N. H., BulL Geol. Soc. America, vol. 7, 1896, p. 517.
* Sloan, Earle, A preliminary report on the clays of South Carolina, 6er. 4, Bull. South Carolina Geol. Survey No. 1, 1904, pp. 26, 76, 104.
* Berry, E. W., Johns Hopkins Univ. Cire., new ser., No. 7, 1907, p. 81.






UPPER CRETrACOUS FLOR4 OF SOUTH CAROLINA.

River. during the same year a small collection from the banks of that river was made
Profs. B. L. Miller and M. W. Twitchell. In the spring of 1908 Mr. Stephenson spent some
in the State, making collections from the Black Creek formation of the eastern part qf
State as well as from deposits near Middendorf and Langley. Subsequently Mr. Stephen-
a and, the writer made extensive collections from Cretaceous beds near Middendorf and
,gley.. In 1910 Earle Sloan sent the writer a collection from Miles Mill in Aiken County.
These collections have come into the hands of the writer and are the basis of the present
Port. The identifiable material is deposited in the United States National Museum.

GEOLOGY OF CRETACEOUS DEPOSITS OF SOUTH CAROLINA.
The Cretaceous deposits of South Carolina form a belt of varying width extending entirely
ross the State east-northeast and west-southwest. This belt is narrow toward the Georgia
, being only about 20 miles in width at Savannah River, but widens northeast of Black
ver to nearly 100 miles. Both Lower and Upper Cretaceous deposits are represented.
LOWER CRETACEOUS SERIES.
On the southeastern margin of the crystalline rocks of the Piedmont Plateau, at an eleva-
of 400 to 500 feet or slightly more, lies a series of white and colored clays, in many places
kaolins, and of arkosic, locally micaceous, coarse or fine sands with clay balls and, toward
base, subangular pebbles. The deposits are in a few localities lignitic, but no recognizable
fossils have been collected. The beds dip 50 or 60 feet to the mile toward the southeast
have an estimated thickness of 200 to 300 feet. They constitute the "Hamburg phase"
Sloan, which in the Aiken area is separated by a local unconformity into two divisions,
by Sloan "Lower Hamburg" and "Upper Hamburg." They appear to be continuous
the Patuxent ("Cape Fear") formation of North Carolina and thus to represent the
formation of the Potomac group of the Maryland-Virginia area. To the southwest they
ar to be continuous with the Lower Cretaceous deposits of Georgia, which have been
neously correlated with the Tuscaloosa formation (Upper Cretaceous) of Alabama by the
brgia Geological Survey. According to the evidence afforded by poorly preserved plants,
ever, the Lower Cretaceous deposits of Georgia and Alabama appear to be younger than
SPatuxent formation of the type region.

UPPER CRETACEOUS SERIES.
BLACK CREEK FORMATION.
Middendorf arose member.-A somewhat similar series of cross-bedded, varicolored, mica-
us, and in many places arkosic sands, containing pebbles, clay balls, and local deposits of
Mrly pure kaolin unconformably overlies the Lower Cretaceous deposits. These beds, which
i here called Middendorf arkose member, are between 100 and 200 feet in thickness and
Beist to a large extent of reworked Lower Cretaceous materials. They can be traced inter-
tedly nearly across the State, being transgressed by the Eocene at several points southwest
inches River and either replaced by partly contemporaneous deposits of a different character
gressed by later Cretaceous in the northeastern part of the State. These beds, which
tute the Middendorf formation of Sloan, in places carry a rich and varied fossil flora of
Cretaceous age. Along the landward margin of their outcrop their lithologic features are in
d contrast with the characteristic beds of the Black Creek formation, but farther south-
, for example, in the Congaree River valley, the Middendorf member commonly exhibits
us, lignitic sandy phases, and dark laminated argillaceous phases exactly similar to
of the typical Black Creek but less extensively developed. On the other hand, the
k Creek formation as developed in North Carolina contains a great thickness of light
highly colored cross-bedded sands similar to those of the Middendorf member of South
lina, and it there covers the time interval of the Middendorf member. The Middendorf
ore has. been adopted by the Survey as a member of the Black Creek formation.





8 UPPER CRETACEOUS AND EOCENE FLORAS OF SOUTH CAROLINA AND GEORGIA.

Other deposit-.-In part later than the Middendorf member and in part contemporaneous
with it is a series of dark laminated clays shaless) with sand partings and lenses of micaceous
fine-grained sands. These clays, which constitute the Black Creek formation of Sloan, are
confined to the eastern part of the State, covering the southern half of Marlboro County, the
northwestern part of Marion County, the northern part of Florence County, nearly all of
Darlington County, and the southeastern part of Lee County, their areal outcrop being termi-
nated abruptly by the transgression of the Eocene in south-central Lee County. These beds
are extensively developed in North Carolina, where they attain great thickness. The lower
part of these beds probably represents most, if not all, of the Middendorf sedimentation, as
well as that of the period intervening between the close of the Middendorf and the beginning
of the Peedee: Fossil plants of genera and species similar to those described from the North
Carolina area have been collected from a number of localities in Chesterfield, Florence, and
Darlington counties, but the exposures throughout this area are few and poor, owing to the
flatness of the country, which is uniformly covered with a mantle of Pleistocene in addition
to patches of Eocene and Miocene deposits. The area occupied 'by these deposits of the Black
Creek formation in South Carolina has not been very extensively examined for fossil collections
because there is no reason to believe that their flora differs in any respect from that found in
the Black Creek formation in the North Carolina area, where the exposures are more
plentiful.
As previously mentioned, the Middendorf member in the Congaree Valley contains lenses
of micaceous lignitic sands and dark laminated clays like those of the typical Black Creek, indi-
cating that at times the conditions which led to this great thickness of deposits along the north
border of South Carolina prevailed farther toward the southwest. Like the Black Creek
deposits of North Carolina, these beds of South Carolina are in places glauconitic, and pellets
of amber are widely distributed in them.

PEEDEE SAND.
A series of compact sands, in many places fossiliferous, less commonly calcareous, and
somewhat glauconitic, conformably 1 overlies the Black Creek formation where it is not trans-
gressed by the Eocene. These sands constitute the Peedee sand, or the "Burches Ferry marls"
of Sloan. They extend northeasterly into North Carolina. They are typically marine deposits,
carrying the Belemnitella americana fauna, and are correlated with the Monmouth formation
of the Northern Atlantic Coastal Plain and with the typical Ripley of northeastern Mississippi.
They are not known to contain fossil plants in the South Carolina area and are confined to the
eastern part of the State, where they are widely distributed, though commonly hidden by
various Tertiary or Pleistocene deposits.

DESCRIPTION OF PLANT LOCALITIES.
To supplement the foregoing brief outline of the Cretaceous as developed in South Carolina,
a few words may be devoted to the 11 different exposures where fossil plants have been collected.
Localities referred to the Middendorf arkose member of the Black Creek formation are
first considered, after which those in the other deposits of the Black Creek formation, in the
order of their geographic succession to the southwest, are taken up.
1. Near Middendorf (Field No. 3.7).-This locality is near the Seaboard Air Line Railway,
about 2 miles northeast of the town of Middendorf and immediately west of overhead wagon
bridge 366. It is just south of the Camden wagon road and about 5 miles east of Big Black
Creek. (See Pl. I, A.) The following section is exposed in the railroad cut:
Section exposed in railroad cut 2 miles northeast of Middendorf.
Feet.
1. Surficial sandy loam, sand, and clay .................................................... 4-10
2. Clay lens containing leaves ......................-- ....--- ..........- ..- ............. .. 0-4
3. Cross-bedded arkosic sands, variegated in color........................................... 4-10
1 Stephenson records an unconformity, presumably local, in Darlington County (unpublished notes).











PROFESSIONAL PAPER 84 PLATE I


PLANT-BEARING CLAY IN MIDDENDORF ARKOSIC MEMBER, BLACK
CREEK FORMATION.
A, Near Middendorf, Chesterfield County, S.C.; B, Near Langley, Aiken County, S.C.


U. 8. GEOLOGICAL SURVEY








ORBTAOCEOUS FLORA OF SOUTH CAROLINA. 9

tfoId beneath a ferruginous crust about 3 inches thick, which is local and
th middle portion of the clay lens. They are for the most part replace-
e and make very handsome specimens, the rusty red leaf contrasting against
olight-buff kaolin, for the bright red of the freshly collected material fades
drying out. Toward each end of the cut the Pleistocene-Middendorf contact
and disappears below the level of the track. The Middendorf member in
consists for the most part of sands, with the exception of the one prominent

ie most prolific locality for fossil plants in South Carolina, having furnished 41
b species. Most of these are splendidly preserved. The following species occur
orf locality:


" illeoidee Berry.
iophylloides Berry.
ic re Hollick.
'macrocarpum Newberry..
3ndorfensis Berry.
fdorfenss Berry.
, crenatum Heer.
lddendorfensis Berry.
rum (Lesquereux) Berry.
S'pandurEeformis Berry.
Heer.
tlted (Heer) Heer.
Eiian Berry.

Berry.

BHeer.


Heer.
elegant Hollick.


Laurophyllum nervillosum Hollick.
Leguminosites middendorfensis Berry.
Lycopodium cretaceum Berry.
Magnolia capellinii Heer?
Magnolia obtusata Heer.
Magnolia tenuifolia Lesquereux?
Momisia carolinensis Berry.
Moriconia americana Berry.
Onoclea inquirenda (Hollick) Hollick.
Pachystima? cretacea Berry.
Pinus raritanensis Berry.
Potamogeton middendorfensis Berry.
Proteoides lancifolius Heer.
Proteoides parvula Berry.
Quercus pseudowestfalica-Berry.
Salix flexuosa Newberry.
Salix lesquereuxii Berry.
Salix pseudohayei Berry.
Sequoia reichenbachii (Geinitz) Heer.
Widdringtonites subtilis Heer.


the most abundant form is Sequoia reichenbachii (Geinitz) Heer, of which whole
es, many with their cones attached, are present.
* Point (Field No. 3.1).-This locality is one-half mile northeast of Sumter Junction,
division of the Southern Railway, along the eastern scarp of the Wateree Swamp,
east of the river and a short distance east of Beach Creek. The following section


Section exposed at Rocky Point.
ene:
4d, coarse, reddish, argillaceous, more or less cemented, with iron crusts below.........
:k and, fine, light gray, stratified, becoming more ferruginous below..................


Feet.


brownish, laminated, with partings of light-gray sand ........................... 6-8
taceous (Middendorf arkose member of Black Creek formation):
coarse, white and yellow, ferruginous, becoming very coarse below ................ 5-6
', lens, brownish, laminated .... .......... .................................... 1
ad, very coarse, ferruginous ...... ........................................... 4-5
WWr materials, more or less indurated .......................................... 3-4
y, interlaminated, brownish, and clay ironstone with abundant leaf impressions ...... 2
6d, coarse, brown............ ...... ......................................... 2
y, laminated, brown, and clay ironstone crusts .......... .......-............. 3-4
, more massive, weathering to reddish and purple ............................... 4-5
, light drab, fine, with some clay laminme and a few thin ferruginous crusts ........... 10
, light drab............ .................... .... ................. .............. 10
toad at this point is 115 feet above sea level.






10 UPPER CRETACEOUS AND EOCENE FLORAS OF SOUTH CAROLINA AND GEORGIA.


The following 26 species of fossil plants occur at this locality:


Andromeda grandifolia Berry.
Andromeda parlatorii Heer.
Arundo grenlandica Heer?
Carex clarkii Berry.
Celastrophyllum elegans Berry.
Cinnamomum newberryi Berry.
Crotonophyllum pandurmformis Berry.
Cunninghamites elegans (Cords) Endlicher.
Diospyroe primava Heer.
Diospyroe rotundifolia Lesquereux.
Eucalyptus geinitzi (Heer) Heer.
Ficus atavina Heer.
Ficus crassipes Heer.


Ficus krausiana Heer.
Hamamelitee (?) cordatus Lesquereux.
Heterolepis cretaceus Berry.
Illicium watereensis Berry.
Magnolia capellinii Heer?
Phragmites pratti Berry.
Podozamites knowltoni Berry.
Proteoides lancifolius Heer.
Protophyllocladus lobatus Berry.
Quercus pseudoweetfalica Berry.
Quercus sumterensis Berry.
Salix lesquereuxii Berry.
Widdringtonites subtilis Heer.


Of these the various species of Ficus are the most abundant, Ficus crassipes Heer and
Ficus krausiana Heer being especially common and well preserved. Andromeda grandifolia
Berry and Protophyllocladus lobatus Berry are also common, the other species being represented
by only a few individuals. This locality stands next to that near Middendorf in the variety
of forms represented, but a good many of the specimens were in a rather fragmentary condition
before entombment.
3. About 25 miles below Columbia (Field No. 3.3).-The plants from this locality, which is
on the right bank of Congaree River, in Lexington County, are represented by faint impressions
in a buff, almost white, micaceous, finely arenaceous, thinly laminated clay, and only the
following forms, the majority tentatively identified, are recognizable:
Cinnamomum newberryi Berry (?). Ficus krausiana Heer (?).
Diospyros primeva Heer (?). Salix flexuosa Newberry (?).
Ficus crassipes Heer. Salix lesquereuxii Berry (?).
The two species of Ficus are the most common. Dark laminated lignitic sandy clays
along Congaree River in this vicinity show the typical lithology of the Black Creek formation
farther to the northeast.
4. Near Langley (Field No. 3.8).-This locality is just east of the old Augusta road, 1 mile
north of the town of Langley and one-half mile west of Langley Dam, on Bighorse Creek, in
Aiken County, in a shallow gully which has cut into clay of the Middendorf member to a depth
of about 8 feet. The material is a massive, nearly white kaolin ("chalk"), and the leaf impres-
sions, which are not common, are replacements by ferric oxide and show as a rich coffee color
against the background of nearly white clay. (See Pl. I, B.)
The following 17 species have been identified from this locality:


Andromeda parlatorii Heer.
Araucaria jeffreyi Berry ?
Celastrophyllum carolinensis Berry.
Crotonophyllum pandureformis Berry.
Dewalquea smithi Berry.
Diospyros primnva Heer.
Eucalyptus geinitzi (Heer) Heer.
Ficus crassipes Heer.
Ficus stephensoni Berry.


Laurus plutonia Heer.
Leguminosites robiniafolia Berry.
Myrsine gaudini (Lesquereux) Berry.
Sabalites carolinensis Berry.
Salix flexuosa Newberry.
Salix lesquereuxii Berry.
Salix sloani Berry.
Sapindus morrisoni (Lesquereux MS.) Heer,


All are common, the Dewalquea, Diospyros, Sabalites, and the species of Salix and Ficus
being most common and best preserved.
4a. Miles Mill (Field No. 8.11).-This locality is in northern Aiken County and has not
been visited by the writer. The materials and preservation are the same as at the Langley
locality, and the collection was made by Earle Sloan, former State geologist. The following
species are present in the collection:


Citrophyllum aligerum Berry ?
Crotonophyllum pandurmformis Berry.
Dewalquea smith Berry.
Diospyros primava Heer.


Quercus pseudowestfalica Berry ?
Salix lesquereuxii Berry.
Salix sloani Berry.







.' PPB CRETACEOUS FLORA OF SOUTH CAROLINA. 11

Species the Dewalquea is by far the most common at this locality. The
i not rare, but the others are each represented by only one or two specimens.
M. iaie below Cheraw (Field No. 8.10).-This locality is in Chesterfield County
d by characteristic Black Creek materials carrying leaf remains, not in place,
a river sand bar from some undiscovered Black Creek outcrop in the vicinity. '
Shave been carried far because of their character and because this point is at the
margin of the Black Creek formation. Only one recognizable species, Hedera
Saporta, is represented.
8 miles east of Darlington (Field No. 38.).-This exposure, on the west bank of
e, one-half mile below the ferry, at a spring, shows about 18 feet of dark car-
ated clay with micaceous and locally glauconitic sand partings. About 1
top there is a 1-foot layer of lignitic sand and clay balls, and a foot or two below
perfect leaf remains are plentiful. These include a Phragmites-like form and a'
cus suggesting Ficus craseipes Heer, but the materials are too incomplete for
. Amber in small drops is present at this outcrop. Louthers Lake is an oxbow
dee, and the above exposure is about 6 miles west of the present river channel and
tern boundary of Darlington County.
at bank of Black Creek (Field No. 3.6).-This locality, 1J miles east of Darlington
t south of the Cashua Ferry road bridge, shows a low exposure of characteristic
materials with poor plant remains. The only recognizable forms are Araucaria
* Berry, Ficus krausiana, and Strobilites anceps Berry, although Mr. Stephenson
of Cephalotaxospermum, which were not found in the material sent in, though
bly they occur here, as they are found along Black Creek in the vicinity.
hAua Road (Field No. 3.4).-A low, poor exposure of a light chocolate-colored, poorly
clay of Black Creek age, containing abundant but much-macerated and rather faint
ons appears just northeast of Darlington, near the foot of the slope leading down
Creek and about 2 rods south of the road. This locality was visited by Ward and
1897, and the small collection 1 made by them at that time was sent to Dr. Arthur
the New York Botanical Garden. The writer examined this collection in New York
-to find any identifiable forms. The collection made by Stephenson in 1908 contains
species:
arialis Saporta ? Proteoides lancifolius Heer ?
nervillosum Hollick ? Rhus darlingtonensis Berry.
wberryi Berry ? Salix lesquereuxii Berry ?
Berry.
-last-mentioned form is the most abundant represented in the collection. In addition
forms indeterminable casts of invertebrates were found.
eight bank of Black Creek (Field No. 8.9).-About 25 feet of dark carbonaceous clays
* ated with yellowish, locally indurated, lignitic sand are exposed about 2 miles below
ns Bridge in Darlington County. Amber is present, and comminuted plant remains
iful and well distributed. The plant material is best preserved within a few feet of
f of the section, and the following forms have been recognized:
ancalyptus angusta Velenovsky
'icu krausiana Heer.
Myrica brittoniana Berry.
Ashby's place, 83 miles northeast of Florence (Field No. 3.5).-This exposure is beside a
branch flowing into Black Creek. Characteristic Black Creek materials at this point
the following species:
Algites americana Berry.
Araucaria bladenenais Berry.
Cephalotaxoepermum carolinianum Berry.
I The exact locality of this collection was along a ditch north of the road, from the same outcrop.







12 UPPER CRETACEOUS AND EOCENE FLORAS OF SOUTH CAROLINA AND GEORGIA.

Lignite is abundant, and the clays are full of fragments of indeterminable leaves of dico-
tyledons. They also contain obscure pelecypod impressions. The materials are dark, lami-
nated, lignitic clays with micaceous sand partings and locally small clay lenses. They are
exposed along the branch to a total thickness of 25 or 30 feet, separated into two nearly equal
divisions by an intercalated lens of greenish-gray or yellowish rather fine sand with here and
there small lenses of dark clay, the sand member being some 20 feet in thickness. Fossil plants
occur both above and below the sand lens, detached leaves of the Araucaria being recognizable
most commonly.

UPPER CRETACEOUS GEOLOGIC HISTORY OF SOUTH CAROLINA.
The geologic history of that portion of the Upper Cretaceous of South Carolina which has
yielded fossil plants is not varied, although it apparently represents a considerable portion of
Upper Cretaceous time as a whole. The data upon which the following brief sketch is based
are both physical and biologic, and in order to avoid ambiguity of statement the Middendorf
arkose member and the other deposits of the Black Creek formation are accorded more or less
separate treatment. The deposits of the Middendorf member furnish satisfactory evidence of
their origin, the method of their deposition, and the attendant physical conditions. Histori-
cally they represent the initial advance of the Upper Cretaceous sea over the coastal lands
made up of the Lower Cretaceous arkosic sands and kaolin deposits. This advance was in
point of time somewhat later than the initial sedimentation of the Tuscaloosa formation of
Alabama, but was probably slightly earlier than the initial deposits of the typical Black Creek
sedimentation.
Middendorf time was not of great duration, for the deposits are not thick and were for the
most part formed rapidly, as evinced by the current bedding and coarseness of the sands.
They represent littoral or beach deposits combined with delta deposits and with estuarine
deposits in shallow bays or sounds. Geographically the conditions may be comparable with
those in Albemarle and Pamlico sounds of eastern North Carolina, or less closely with those in
the quiet estuaries of Chesapeake Bay, where fine-grained muds are being formed in certain areas
at the present time. Topographically, however, the parallel is not so close, for the present-day
topography is more mature and the relief is slight. Sections drawn across a number of recent
bays and sounds along the South Atlantic and Gulf coasts show in some degree what must have
been the topography in South Carolina at the beginning of Middendorf sedimentation, except that
the modern shores are low. In Middendorf time the Coastal Plain was represented by a com-
paratively narrow strip of Iower Cretaceous deposits and the eastern flank of the Piedmont was
considerably more elevated than it is to-day, so that the earlier deposits of the Middendorf,
at least, are more largely sands than the deposits now forming along the seaward margin of
the Coastal Plain. These conditions for the Middendorf member may be most simply illus-
trated by the probable history of a single valley. The Lower Cretaceous surface may be
imagined to represent the emerged land surface before the depression or warping or other
process that inaugurated the Upper Cretaceous cycle of deposition as it is known to-day. Ter-
restrial aggradation deposits would be laid down on both valley slopes, and fluviatile or lacus-
trine deposits would be laid down at some points along its floor. With the changes in relative
level the valley would become a bay, estuary, or sound, and the quickened erosion would
rapidly build out deltas or alluvial fans along its western margin. This would segregate the
coarser sediments, and the finer particles, as well as a large part of the floating carbonaceous
matter and minute flakes of mica, would be carried beyond the coarser materials along the
shore and ultimately deposited in laminated clays of the Black Creek type. A brief study of
similar physiographic deposits of the present confirms the essential correctness of this sup-
position. The interaction of forces would surely develop some relief of the surface of the sand
flats; bars or atolls would be formed from the coarser materials and within the lagoons inclosed
by them the kaolins would be deposited. That the sea did not have free access to the area is
shown by the complete absence of evidence of marine life in the Middendorf member and other






U nPPE CRETACEOUS FLORA OP SOUTH CAROINA.' 13

sediments. The fact that remains of terrestrial, fluviatile, or estuarine
found must be attributed to the nature of the sedimentation'just outlined.
rd numerous and more or less torrential in character, piling up their sediments
-alluvial fans which finally merged and were completely united by the sediments
sea. The kaolins were meanwhile deposited in the sheltered basins. Deposits
rarely fossiliferous, as is evinced by the lack of fossils in other analogous series.
taceous of the Coastal Plain outcrops for several hundred miles along the eastern
'Piedmont, but it contains very few fossil plants, and after several decades of
not more than half a dozen scarcely determinable shells and a single fragmentary
found in it.
the Piedmont river gradients were lessened and river and coastal swamps became
us. With continued subsidence such barriers as existed, whether sand bars or
arable to the North Carolina "banks," were submerged, and the sediments began
sa typical marine character and to consist of laminated sands and clays with some
and much vegetable debris derived for the most part from the coastal swamps.
entation of the typical Black Creek type commenced considerably earlier toward
Carolina border than to the southwest-in fact, there are no traces of such sedi-
e Aiken area, a condition partly explained by subsequent subsidence in this area
e profound Eocene overlap. Nevertheless, the much more extensive deposits of
the Aiken area give evidence of longer-continued, comparatively quiet nonmarine

typical Black Creek sedimentation begins a series of deposits which in their origin,
[of materials, and contained flora, are strictly comparable with the Magothy and
formations of the northern Coastal Plain, the Black Creek formation of North Caro-
tutaw formation of Georgia, and the upper half of the Tuscaloosa formation of western
Probably also the physiography of the lands adjacent to the places where these
i were laid down was the same in its general character. All the facts available point
tical synchroneity of these deposits.
'ef at the beginning of Middendorf sedimentation was considerable; the Piedmont
ply weathered during long ages, had been extensively stripped of the decayed.rock
ed the Lower Cretaceous deposits and of the deposits representing the subsequent
al. This removal was not, however, nearly so extensive as many persons have
A considerable elevation and erosion interval which follows is represented by the
atapsco, and Raritan formations of the more northern Coastal Plain and by the
ess of marine sediments of at least the upper part of the Lower Cretaceous section
as region. A subsidence or warping at the close of Raritan time inaugurated the
tic cycle of Upper Cretaceous time.
of the Piedmont surface upon which the Lower Cretaceous rests is between 50 and
the mile, and though the amount of subsequent warping which may have occurred is
able, these figures, however inadequate, furnish some basis for calculating the
ovation of the Piedmont land surface at the commencement of the Lower Cretaceous
ough the Lower Cretaceous surface suffered considerable erosion before Middendorf
beginning of Middendorf sedimentation indicates a second warping with more or less
ubsidence to the eastward and elevation to the westward, so that the general slope
been nearly as great as it was at the beginning of Lower Cretaceous sedimentation

lora as well as the physical conditions show that the Middendorf member and the
of the typical Black Creek deposits were formed, in part, at least, at practically the
e. The Middendorf, representing locally the initial basement sands, was probably
part on land, or at the point where the Cretaceous rivers reached sea level, or in the
bays which are supposed to have existed. Meanwhile the typical Black Creek sedi-
being deposited seaward. As subsidence continued any barriers which may have
submerged, especially in the Peedee area, so that practically the whole sedimentation







14 UPPER CRETACEOUS AND EOCENE FLORAS OF SOUTH CAROLINA AND GEORGIA.

became of the typical Black Creek type and continued long after the close of the deposition
of the Middendorf member.
This report does not treat of the subsequent geologic history, which includes the depositiou
of the immediately succeeding Peedee sand into which the Black Creek formation finally merged
and the subsequent reworking of the deposits in the gradual withdrawal of the sea to its various
Tertiary and more modem levels.

SYSTEMATIC DESCRIPTION OF THE FLORA.
Phylum TRALLOPHYTA.
Genus ALGITES Seward.
ALGITES AMERICANA 8p. nov.
Description.-The thallus is preserved in the form of dichotomously divided branches
ranging in width from 2 to 5 millimeters, thin and undulating, but evidently rather coriaceous
in life, with slightly waved margins. These branches are not preserved for lengths of more
than a few centimeters, in which space they are observed to divide only once or not at all. In
some specimens they appear to radiate from a common center, but as their proximal parts are
invariably missing this supposition can not be verified.
This species is also present in the Magothy formation of Maryland and in the Black Creek
* formation of North Carolina. The Maryland remains are rare and are in the form of impressions
along which recent rootlets have commonly penetrated, giving some specimens the appearance
of having midribs. The North Carolina remains, which are abundant in the Black Creek
formation at certain localities along Black River, show considerable carbonaceous residuum,
indicating that in life the thallus was of considerable consistency.
The name of the genus Algites, to which this form is referred, was proposed by Seward 1
for those fossil remains which are in all probability those of algae but which from their nature
can not be decisively assigned to any established genus.
Fossil algae are common at some geologic horizons, but their characters are generally ill
defined, especially when preserved as impressions, so that comparisons with modern genera
altogether lack certainty. As pointed out by Seward for the type of this genus, Algites
valdensis of the English Wealden, these forms suggest various modem genera, such as Chondrus,
Zonaria, Dictyota, and others.
Occurrence.-Black Creek formation, 3 or 4 miles northeast of Florence (Ashby place),
Florence County. (Collected by L. W. Stephenson.)
Collections.-U. S. National Museum.
Phylum PTERIDOPHYTA.
Order FILICALES.
Family POLYPODIACEMI.
Genus ONOCLIA Linn.
ONOCLEA INQUIRENDA (Hollick) Hollick.
Plate II, figures 7 and 8.
Osmunda obergiana Heer, Flora foesilis arctica, vol. 3, Abth. 2, 1874, p. 98 (parn), Pl. XXVI, fig. 9d (non figs. 9-9b;
Pl. XXXII, fig. 7a).
Caulinites inquirendus Hollick, Bull. New York Bot. Garden, vol. 3, 1904, p. 406, PI. LXX, fig. 3.
Onoclea inquirenda Hollick. The Cretaceous flora of southern New York and New England: Mon. U. S. Gepl. Survey,
vol. 50, 1907, p. 32, Pl. I, figs. 1-7.
Deacription.-Remains of this form appear in fragments of fertile fronds, not showing
any of the laminse, which appears to be reduced to short pinnate branches bearing one or more
1Seward, A. C., Wealden flora, pt. 2, 1894, p. 4.






" 'ONEaQOUS FLORA OF SOUTH CAROLINA. 15

interpreted as sori. These are uniformly 1.5 millimeters or slightly

originally described by Hollick and referred to the genus Caulinites,
Removed to the ferns because of its close resemblance to the modern
figured forms of the same character were associated by Heer with his
ia because they were found in the same beds with the fronds of this
Were not found in organic union with the fronds. These fruits are
of the modern forms of Onoclea than they are like those of Osmunda,
with the type form of the species to which the writer has referred them.
and Marthas Vineyard forms have these sori in a single row on each side
of the South Carolina specimens seem to have a similar arrangement;
ntely in threes, one terminal and two lateral. This latter arrangement
vely in the Greenland specimens and in similar material from the Magothy
. This variation is of minor importance. and is mentioned simply because
grouping in threes is the normal arrangement and that it has been obscured
in the specimens where it is not clear.
, this species ranges from the Atane beds of Greenland southward in
on of Marthas Vineyard, Long Island, and Maryland, to the locality in

ddendorf arkose member of Black Creek formation, near Middendorf,
y. (Collected by E. W. Berry.)
U. S. National Museum.

Order LYCOPODIALES.
Family LYCOPODIACES.
Genus LYCOPODIUM Linnm.
LYCOPODIUM CRETACEUM Berry.
Plate II, figures 1-6.
Berry, Am. Jour. Sci., 4th ser., vol. 30, 1910, pp. 275, 276, figs. 1-6.
These remains consist of fruiting spikes, which are common at the Middendorf
having been collected. Spikes loosely imbricated, of modified foliage leaves
et spike, which is nearly complete, is 5 centimeters in length and 5 millimeters
probably somewhat flattened, the bulk of the specimens indicating some-
. ons. Axis stout. Bracts several ranked, peduncled, with a cordate or
abruptly narrowed acute recurved apex, with an entire margin, each bract
spheroidal sporangium which may possibly be reniform, though in the
in the clays of the Middendorf member it appears to be globular.
of foliage resembling that of the modern club mosses have been frequently
Lycopodium or Lycopodites Brongniart, but the majority of such determi-
ty in that they show neither anatomical nor fruiting characters, so that the
great interest as the only post-Paleozoic fossil known to the writer which
absolute certainty to the genus Lycopodium. No remains of foliage have
these clays which can be correlated with these fruiting spikes.
ddendorf arkose member cf Black Creek formation, near Middendorf, Ches-
- (Collected by E. W. Berry.)
SU. S. National Museum.





16 UPPER CRETACEOUS AND EOCENE FLORAS OF SOUTH CAROLINA AND GEORGIA.
Phylum SPERMOPHYTA.
Class GYMNOSPERKM.
Order CYCADALES (?).
Family CYCADACES (?).
Genus PODOZAMITES F. Braun.

PODOZAMITES KNOWLTONI Berry.
Plate IV, figtire 5.
Zamites angstifolius Eichwald, LethEea roesica, vol. 2, 1860, p. 39, Pl. II, fig. 7.
Podozamites angustifolius Schimper, Pal6ontologie veg6tale, vol. 2, 1872, p. 160 (non Schenk).
Podozamites angustifolius Heer, Flora foesilis arctica, vol. 4, Abth. 1, 1876, p. 36, Pls. VII, figa. 8-11, and VIII, figs.
2e, 5.
Podozamites angustifolius Heer, idem, Abth. 2, 1876, p. 45, Pl. XXVI, fig. 11.
Podozamites angustifolius Heer, idem, vol. 5, Abth. 2, 1878, p. 22, Pl. V, figs. lib, 12.
Podozamites angustifolius Lesquereux, The Cretaceous and Tertiary floras, 1883, p. 28.
Podozamites angustifolius Lesquereux, The flora of the Dakota group: Mon. U. S. Geol. Survey, vol. 17, 1892, p. 27,

w.. JiiH fJg� _5 Newberry, The flora of the Amboy clays: Mon. IT. S. Geol. Survey, vol. 26, 1896, p. 44.
-. .- ,yrt Po Msiaenusztolwius Moller, Kgl. Svensk. Vetensk. Akad., Handl., vol. 9, 1903, PI. I, figs. 8-12, 17b.
Podozamites knowltoni Berry, Bull. Torrey Bot. Club, vol. 36, 1909, p. 247.
Podozamites knowltoni Berry, idem, vol. 37, 1910, p. 182.
Description.-Leaflets elongate, linear-lanceolate in outline, falcate in the single nearly
complete South Carolina specimen, 5 to 15 centimeters in length by 0.6 to 1.3 centimeters in
greatest width. Base narrowed to a short or obsolete petiole. Apex long pointed. Veins
straight and parallel, gradually dying out as they are encroached upon by the narrowing mar-
gins, that is, running to the margins and not converging appreciably toward the tip. At 4
centimeters above the base they are 20 in number, and at 2 centimeters below the tip of the
specimen figured they are 16 in number.
Fragments of the leaf tissue preserved show that the leaf was comparatively thick and
smooth on one surface, presumably the upper, with veins prominent on the other surface.
Epidermis seems to be composed of small cells. Holes in a single, more or less imperfect row
between the veins may represent stomata that are apparently confined to the lower (?) surface.
The outlines of the internal tissue mesophylll) show large cells longitudinally elongated. This
species is suggestive of Nageiopsis longifolia Fontaine, of the Lower Cretaceous of Maryland
and Virginia, and it may also be compared with a variety of so-called species of Podozamites
founded upon detached leaflets of doubtful attribution.
In 1872 Schimper referred the Zamites angustifolius of Eichwald to the genus Podozamites,
overlooking the fact that four years earlier Schenk had described and named a Podozamites
angustifolius. The natural impulse would be to dedicate the species here described to Eich-
wald, but Eichwald has already had a species of Podozamites named for him, in consequence of
which the name Podozamites knowltoni has been proposed in honor of Dr. F. H. Knowlton, of
the United States National Museum. This species has a wide range, both geologically and
geographically. It is common in the Jurassic of high latitudes in Russia, Siberia, Bornholm,
and Spitzbergen, and it is found in the Upper Cretaceous as indistinguishable remains widely
distributed in America, being common in the Dakota sandstone of the West, the Raritan
formation of New Jersey, and the Black Creek formation of North Carolina.
Occurrence.-Middendorf arkose member of Black Creek formation, Rocky Point, Sumter
County. (Collected by L. W. Stephenson.)
Collections.-U. S. National Museum.
I Although the date on the title-page of this work is 1895, it was not actually published until 1896.







UPPBR CRETACEOUS FLORA OF SOUTH CAROLINA. 17
Order CONIFERALES.
Family TAXACEB.
Genua PBOTOPHYLLOCLADUS Berry.
PROTOPHYLLOCLADUS LOBATUS Berry.
Plate II, figures 9-13.
nov. Berry, Johns Hopkins Univ. Cire., new ser., No. 7, 1907, p. 81.
lobatus Berry, Bull. Torrey Bot. Club, vol. 38, 1911, p. 403.
' .-Leaves (phylloclads) of large size, lanceolate or oval in general outline, either
crenate margins, rounded apex, and narrowly cuneate base or compound through
ent of opposite lateral lobes. Axial vascular strand very stout below, becoming
and finally disappearing by repeated branching apically. When the leaves are lobate,
te opposite vascular strands form the axis of the lobes, and these also are generally,
variably, lost before reaching the tips of the lobes by giving off innumerable secondary
Margins in all specimens are rather remotely undulate crenate, and the tips are all
Secondaries numerous and thin, diverging from the main axis of the phylloclad or
of the lobes at very acute angles, curving outward, some simple but many dichoto-
ked, and a few several times forked. Lobes when present are separated by cuneate,
rounded sinuses which terminate some distance from the main axis. The largest
which is still incomplete both at the apex and at the base, measures 8 centimeters
and 5 centimeters from tip to tip of the lower lobes, the entire upper portion meas-
ut 1.5 centimeters in width.
remains are superficially like fern fronds, especially in specimens which are com-
and were it not for the presence in the Cretaceous of other phyllocladus like remains
demonstrated gymnospermous structure (for example, Androvettia) their reference to
us would seem hazardous. The entire specimens are strikingly like some of the forms
pyllocladus subintegrifolius (Lesquereux) Berry of the Raritan and Magothy formations,
Protophyllocladus polymorphus (Lesquereux) Berry from higher western American
, and even the compound specimens have an unlobed apical portion of comparable
:which is also similar in appearance to the two species just mentioned. The compound
are superficially like Thinnfeldia rhomboidalis Ettingshausen,' the type of the genus
eldia,*whose systematic position has been the occasion of so much controversy and which
variously regarded as a fern, as a cycad, and as a conifer. The present species shows
t differences, however, aside from being much younger, and it is confidently believed
.unrelated to the various older Meso:oic species of Thinnfeldia which have been described.
It may also be compared with various forms from the Upper Cretaceous of Dalmatia
based at great length by Kerner,2 who refers them to the genus Pachypteris, which he
as cycadaceous in nature.
e present species is believed to be closest to ProtophyUocladus subintegrifolius, a species
is abundant in the Atane beds of Greenland, the Dakota sandstone of Kansas and
a, the Raritan formation of New Jersey, and the Magothy formation from Marthas
d to New Jersey, and which commonly assumes a sublobate form. This. is especially
in unreported collections made by the writer in the Magothy formation of New Jersey.
tthCarolina form is common, mostly as fragmentary specimens, at Rocky Point, to which
it appears to be confined in South Carolina. Regarding the systematic position the
is confident, despite certain criticisms, that Protophyllocladus is referable to the
e. The species is also found in the Magothy formation of Maryland.
ecurrence.-Middendorf arkose member of Black Creek formation, Rocky Point, Sumter
. (Collected by L. C. Glenn, Ward and Glenn, and E. W. Berry.)
eons.-U. S. National Museum; Johns Hopkins University.
I Ettingshausen, C. von, Abhandl. K.-k. geol. Reichsanstalt, vol. 1, pt. 3, No. 3, 1852, p. 2, P1. I, figs. 4-7.
!-. 'I Kerner, F. von, Jahrb. K.-k. geol. Reichsanstalt, vol. 45, 1896, p. 39.






18 UPPER CRETACEOUS AND EOCENE FLORAS OF SOUTH CAROLINA AND GEORGIA.
Genus CEPHALOTAXOSPERSMUM Berry.

CEPHALOTAXOSPERMUM CAROLINIANUM Berry.
Plate III, figure 4.
Cephalotazospermum carolinianum Berry, Bull. Torrey Bot. Club, vol. 37, 1910, p. 187.
Description.-Drupaceous fruits, solitary (?), sessile or with an extremely short and stout
peduncle, ovoid, somewhat pointed apically and inclined to become slightly cordate below,
consisting of an outer fleshy layer and an inner bony layer, as in the Cycadales and Gingkoales;
its surface mammillated much as in Podocarpu elongata but less markedly so. Bony endocarp
ovate-acuminate, immersed in the apical part of the exocarp. Evidently the drupaceous
fruit of some Cretaceous member of the Taxai6em which finds its closest homology in the recent
flora in the fruits of Cephalotaxus and certain species of Podocarpus. These drupes have the
following dimensions as preserved in a much flattened condition: Length 6 millimeters to 13
millimeters, averaging about 10 millimeters; breadth 5 millimeters to 10 millimeters, averaging
about 8 millimeters; thickness about 3 millimeters; fruit in life probably almost circular in
cross section. Peduncle short and stout or wanting. Stone ovate-acuminate, lying in the apical
part of the fleshy exocarp with the beaked micropylar end reaching almost or quite to the
apex. As preserved in a much flattened condition in the clays, these fruits tend to split into
two parts, disclosing the bony endocarp or merely a cast of its cavity. The fleshy part of
the fruit is carbonized and fails to show any histological details. There is some evidence, or at
least a suggestion, in some specimens of the remains of a micropylar canal. Away from the
pointed apex the exocarp is 1 to 2 millimeters in thickness, reaching a thickness of 3 millimeters
at the chalazal end.
These fruits are very abundant at certain localities in the Black Creek formation in North
Carolina, and have also been collected in the extension of this formation near Florence, S. C.,
and in the lower part of the Eutaw formation in Hale County, Ala.
Fruits referable to the Taxacee are extremely rare in the fossil state, as are also remains of
foliage which can be referred with certainty to this family. Both Tumion and Cephalotaxopsis
from the Lower Cretaceous of Maryland and Virginia are founded upon foliage which seems
referable with considerable certainty to this family, and the same strata in those States abound
in the foliage referred to the genus Nageiopsis, which seems to be closely related to Podocarpus,
so that there is considerable reason for expecting to find Upper Cretaceous representatives of the
family in this same general region. Heer1 describes from the Patoot beds (Senonian) of Green-
land a leafy twig with a large solitary fruit, which he calls Cephalotazites insignia, an identifi-
cation which Solms-Laubach2 seems to consider probable. Bertrand3 has described carbonized
seeds from the Aachenian of Tournay, Belgium, under the name of Vesquia tournaisii, which he
considers, because of the arrangement of the vascular bundles, as intermediate between Tumion
and Cephalotaxus. It certainly seems to be not without significance that remains of this sort
occur at nearly homotaxial horizons in America, Europe, and Greenland.
None of the foregoing, however, are comparable with the present forms, although certain
indefinite remains described by Lesquereux as Inolepis sp.,' are remotely suggestive of them.
It is not believed, however, that they are congeneric.
The general features which seem to indicate a closer relation with Cephalotaxus than
with Podocarpus are the absence of the thickened peduncle of the latter and the presence in the
same beds with these seeds of foliage described by the writer as Tumion carolinianum,5 which is
of the same type as that of Cephalotaxus and may not improbably have been the foliage of the
tree which bore the very abundant fruits named Cephalotaxospermum.
I Heer, Oswald, Flora fossils arctic, vol. 7, 1883, p. 10, PI. I1III, fig. 12.
* Solms-Laubach, Fossil botany, 1891, p. 61.
' Bertrand, C. E., Bull. Soc. beot. France, vol. 30, 1883, p. 293.
4 Lesquereux, Leo, in Hayden's Ann. Rept. for 1874, 1876, p. 337, PI. IV, fig. 8; The Cretaceous and Tertiary floras, 1883, p. 33, PI. I, fig. 8.
6 Herry, E. W., Am. Jour. ScL., 4th ser., vol. 25,1908, pp. 382-386 figs. 1-3.





-�

UPPER CRETACEOUS FLORA OF SOUTH CAROLINA. 19
_
e modem genus Cephalotaxus Siebold and Zuccarini, with four species, is confined to
ina-Japan region, although it seems evident that it was much more widespread in former
* times, and to it should probably be referred some of the leafy twigs included in the
Taxites Brongniart. Fruits of three species of Cephalotaxus, apparently identified cor-
y, are described by Kinkelin 1 from the upper Pliocene deposits of the Main Valley in
any.
Occurrence.-Black Creek formation, 3j miles northeast of Florence (Ashby place), Flor-
County. (Collected by L. W. Stephenson.)
llections.-U. S. National Museum.
Family ARAUCARIACEB.
Genus ARAUCARIA Jussieu.
* ABRAUCARIA BLADENENSIS Berry.
Plate III, figures 6 and 7.
"-a bladenesi Berry, Bull. Torrey Bot. Club, vol. 35, 1908, p. 255, Pls. XII, XIII, and XIV, figs. 1-3.
Description.-Foliage dense; phyllotaxy spiral; leaves decurrent, coriaceous, ovate-
l!ate, about 1.6 by 0.8 centimeters, the base rounded, apex thickened, cuspidate; veins
sed, averaging 16 in number, straight, parallel; stomata small, in rows on ventral surface.
-. ranging from 1 to 2.8 centimeters in length by 0.5 to 1.2 centimeters in width, averaging
y 0.8 centimeters, obovate in outline, with a broad, rounded base narrowing abruptly and
u'rent; the blade broadest about one-third of the distance from the base, above which point
oTws in a short distance to a thickened cuspidate tip; phyllotaxy spiral; leaf substance
.nted by a sheet of lignite about 0.5 millimeter thick, in which the veins are immersed.
veins average 14 to 16 in number, although some specimens may have as many as 20.
are stout, incurved at the base (forking not observed), becoming parallel and running
ly upward until they abut against the leaf margin; that is, they are not convergent toward
, of the leaf. Although their megascopic appearance is lifelike, their microscopic structure
preserved.
From one or two places where the specimens are in a more argillaceous matrix it has been
le to get rather inferior specimens showing the arrangement and outlines of the stomata.
are broadly ovate in shape with very thin guard cells, at least when viewed on the surface.
are arranged in somewhat irregular rows on the ventral surface of the leaf, the number
a between the two veins being generally four. Aside from these facts no other details
made out.
-is species is most remarkably similar to the recent Araucaria bdwid i of the Australian
The resemblance is even closer than the reproductions indicate, a dried herbarium
en of the latter and a twig of the former when preserved as a brownish impression being
lly indistinguishable.
-is resemblance in form, habit, and stomatal characters, reenforced by the occurrence of
-tristic araucarian cone scales in the same beds at certain localities, renders the identifi-
reasonably conclusive.
e most nearly related form seems to be Araucarites ovatus, described by Hollick2 from the
"od clay" of New Jersey. The leaves of this form differ merely by their larger size,
of basal characters, and much less pointed tips; in fact, if the two were found in closer
'on or if in the abundant material of Araucaria bladenesis any specimens had approached
es ovats in size, the writer would be disposed to consider them variants of a single
As the case stands, it seems better to keep them distinct, for the leaves in the material
a southern Coastal Plain are sufficiently and uniformly different to be readily recognized,
-eis the further possibility that the New Jersey species may be more or less closely related
modern genus Dammara rather than Araucaria.
- dathrdt, Hermann, and KinkelUn, F., AbhandL Senckenb. naturf.'GeselL, vol. 29, No. 3, 1908, p. 194, PL XXIII, figs. 9-13.
MHolk, Arthur, Trans. New York Acad. Sci., vol. 16, 1897, p. 128, P1. XII, figs. 3a, 4.
-�

:-^




UPPER CRETACEOUS AND BOOBEN FLORAS OF SOUTH CAROLINA AND C


EORGIA. .


A European form which must surely be considered as a nearly related congener of Araucaria
bladenensis is Saporta's Araucaria toucasi, described from the Turonian of Beausset, near
Toulon, France.1 This is strikingly similar to the American species in every respect, and is
likewise closely allied, in appearance at least, to the recent Araucaria bidwili of Australia.
Kerner I records PachyphyUum (Pagiophylum) rigidum Saporta and PachyphyIlum (Pagio-
phyUum) araucarinum Saporta from the Cenomanian of Lesina, an island in the Adriatic off the
coast of Dalmatia, both being originally Jurassic species from the French Corallian of Verdun.
Both are very similar to the American species and are of about the same age. The probable
identity of Cenomanian and Corallian species seems to the writer extremely doubtful, and in his
opinion both Kerner's species should undoubtedly be considered new species of Araucaria, nearly
related to, if not identical with, such Cretaceous forms as Araucaria bladenesis or Araucaria
toucasi.
The present species is exceedingly common in and characteristic of the Black Creel; formation
in North Carolina. In South Carolina it is found in the extension of these beds at the single
locality cited. So far as observed, the leaves are always found detached at this outcrop, indi-
cating a large amount of maceration and trituration. They are, however, entirely characteristic.
This species has also been found in the upper part of the Tuscaloosa formation of western
Alabama, in the Eutaw formation of western Georgia, and at a somewhat younger horizon nbar
Buena Vista, Ga.
Recently Wieland ' has described a distinct but comparable species, Araucaria hatcheri,
from the "Ceratops beds" of Wyoming.
Occurrence.-Black Creek formation, 31 miles northeast of Florence (Ashby place), Florence
County. (Collected by L. W. Stephenson.)
Collections.-U. S. National Museum.

ARAUCARIA DARLINGTONENSIS sp. nov.
Plate III, figure 1.
Description.-Seed obovate in outline with broadly rounded apex, straight lateral margins,
and somewhat narrowed rounded base, 1.25 centimeters in length and 0.75 centimeter in width
across the widest part, 0.50 centimeter wide at base.
This species is based upon the single detached seed figured, which is undoubtedly that of
an araucarian conifer. From its size and geologic position it seems probable that it may be
a seed of the cone scales described as Araucaria jeffreyi Berry, which in turn are probably the
cone scales of the leafy twigs described as Araucaria bladenensis Berry. No other araucarian
remains are, however, associated with this seed at this locality.
Occurrence.-Black Creek formation, right bank of Black Creek, 1 miles east of Darlington,
Darlington County. (Collected by L. W. Stephenson.)
Collections.-U. S. National Museum.

ARAUCARIA JEFFREYI Berry (?).
Araucaria jefreyi Berry, Bull. Torrey Bot. Club, vol. 35, 1908, p. 258, Pl. XVI.
Description.-Cone scales deciduous, rhomboidal, straight sided, thin margined, the apex
broadly rounded, with long central apical spur; scales divided by transverse furrow into "ligule"
and scale proper; single seeded.
This species was based upon a considerable number of large single-seeded cone scales
preserved as impressions and associated with Araucaria bladenensis at Big Bend and near the
Atlantic Coast Line Railroad bridge over Black River, at points 92 and 871 miles above Newbern
on Neuse River, and at Parker Landing on Tar River, all localities in the Black Creek formation
of North Carolina. The specimens from Tar River differ somewhat from the others and
I Saporta, G. de, Le monde des planted, 1879, p. 198, fig. 27.
2 Kerner, F. von, Jahrb. K.-k. geol. Reichsanstalt, vol. 45, 1896, p. 49, Pl. IV, figs. 1, 3.
a Wieland, G. R., Bull. Geol. Survey South Dakota No. 4, Rept. for 1908, 1910, p. 80, fig. 2.


4
1
:!





S PPRm CRETACEOUS FLOBA OF SOUTH CAROLINA. 21

nearly the shape of the foliage leaves of Araucaria bladnenais, but the
are somewhat variable, as indeed they are from different positions on a single
cone, and it seems likely that they all belong to one species.
rae rhomboidal, the thin lateral margins straight to the point of greatest width,
rounded, produced medially into a long and narrow point. This point is
ntieter long in two specimens which still lack .the terminal portion. In some
es are ob-ously divided by a transverse furrow into the scale proper and the
-e." They are all preserved as impressions with fragments of lignite representing
. Except that they do not appear to have been as thick, they are strictly
the typical scales of Araucaria bidwilli. In general outline they are also
h the scales of Araucaria coolcii of the Eutacta section of the genus. Seeds
found in the Carolina material at this level. From the structure as disclosed
Impressions it seems obvious that the scales were single-seeded, as in the modern
n in conjunction with the foliage just described as Araucaria bladenensis, they
nsive evidence of the abundant presence in the middle part of the Cretaceous of
. America of true Araucariese, thus still further increasing the parallel between the
us floras of this country and those of Europe. Many remains of cones and
ve been described as species of Araucariese,.but it seems scarcely worth while to
r| Chere.
Carolina occurrence is based on a single poorly preserved and doubtfully deter-
. This species also occurs in Georgia.
r material is neither abundant nor well preserved, but the identifications are
ble. A single scale was collected from the Eutaw formation at Chimney Bluff on
Ree iver, where it was associated with the abundant leafy twigs of Araucaria
jierry. A single entirely characteristic specimen obtained near Byron, Georgia, is
m a higher horizon in the Cretaceous.
Sefreyi is extremely close to a form from the Turonian of Priesen, Bohemia,
d named Araucariafrid by Velenovsky.1
ree.-Middendorf arkose member of Black Creek formation near Langley, Aiken
elected by E. W. Berry.)
je.-U. S. National Museum.
Family BRACHYPHYLLACEB.
Genus BRACHYPHYLLUM Brongniart.
BRACHYPHYLLUM MACROCARPUM Newberry.
Plate III, figure 2.
eton m Debey and Ettingshausen, Heer, Flora fossilis arctica, vol. 7, 1883, PI. LIV, fig. Ic (non Heer's

pl equereux; The Cretaceous and Tertiary floras, 1884, p. 32.
Ocrauum Lesquereux, Proc. U. S. Nat. Mus., vol. 10, 1887, p. 34.
t esesum Lesquereux, The flora of the Dakota group, Mon. U. S. Geol. Survey, vol. 17, 1892, p. 32, Pl.

macrocwrpum, Newberry, MSS. name mentioned in footnote. The flora of the Amboy clays: Mon. U. S.
, vol. 26, 1896, p. 51.
lp., Knowlton, Bull. Geol. Soc. America, vol. 8, 1897, pp. 137, 140.
bacrocorpum Knowlton, Bull. t. S. Geol. Survey No. 163, 1900, p. 29, Pl. IV, figs. 5, 6.
Wsomsarpun Hollick, Bull. New York Bot. Garden, vol. 3, 1904, p. 406, Pl. VII, figs. 4, 5.
.uarocarpum Berry, Bull. Torrey Bet. Club, vol. 32, 1905, p. 44, Pl. II, fig. 9.
hsidearprum Berry, idem, vol. 33, 1906, p. 168, PI. IX.
unwrocarpuqm Berry, Ann. Rept. State Geologist, New Jersey, for 1905, 1906, p. 139.
Smaerocarpum Hollick and Jeffrey, Am. Naturalist, vol. 40, 1906, p. 200.
imacroarpum Hollick, The Cretaceous flora of southern New York and New England: Mon. U. S. Geol.
_vol. 50, 1907, p. 44, Fl. III, figs. 9 and 10.
1 In Frie, Archiv natuv.Lan lteeturid rro rflci55ulame,'O.'\%,NC,.L,'L,fl.1?2 ,te.It%.i- ..





22 UPPER CRETACEOUS AND EOCENE FLORAS OF SOUTH CAROLINA AND GEORGIA.

Description.-Stout twigs with club-shaped, privately arranged branches, covered with
large, thick, rhomboidal squamate, densely crowded, appressed leaves, attached by practically
their whole ventral surface. Phyllotaxy spiral. Leaf surface more or less striated, the string
converging toward the obtuse apex. Cones not positively determined.
Brachyphyllum is chiefly an older Mesozoic type, but it remains abundant through the
Lower Cretaceous, two species having been described from the Potomac group of Maryland and
Virginia. It is a waning type in the Upper Cretaceous, where it is represented by the species
here discussed and by a variety (see p. 106) which persists as high as the lower Senonian. It is
widely distributed and is recorded from Long Island, Staten Island, New Jersey, Delaware,
Maryland, North Carolina, Georgia, and Alabama in the East and from the Dakota sandstone
of Kansas and the Montana group of Wyoming in the West. It is probably represented in the
Patoot beds of Greenland by the material which Heer erroneously refers to Moriconia. Though
it is not recorded from Europe, Velenovsky has described remains from the Cenomanian of Bo-
hemia which appear to be identical with the American representatives, referring them to the
Jurassic genus Echinostrobus of Sohimper.2 Hollick and Jeffrey have recently shown, from
specimens from Staten Island with structure preserved, that this species is related to the
subfamily Araucariese.
This species is extremely common in the upper part of the Raritan formation at South
Amboy, N. J., and in its eastward extension on Staten Island, but has not been collected from
any of the plant-bearing horizons of the lower part of the Raritan. Newberry describes large
cones which he found associated with these twigs and which he thought were related to them,
although this seems improbable. The cones are poorly preserved and their affinities can not
be made out. They are very different from previously described cones of Brachyphyllum, and
the work of Hollick and Jeffrey would seem to indicate that the present species had small
cones. The cones described by Newberry, though here retained in the synonymy of this species,
are comparable to the abundant cones from the older Potomac of Maryland which are referred
to the form genus Abietites. The single characteristic fragment figured is all that represents
this species in the South Carolina Cretaceous, but as it is common in homotaxial deposits in
Georgia and Alabama and has also been found in the Black Creek formation of North Carolina
no uncertainty is attached to the identification of even such meager material.
Occurrence.-Middendorf arkose member of Black Creek formation, near Middendorf,
Chesterfield County. (Collected by E. W. Berry.)
Collections.-U. S. National Museum.
Family PINACEX.
Genus PINUS Linn6.
PINUS RARITANENSIS Berry.
Pinus sp. Newberry, The flora of the Amboy clays: Mon. U. S. Geol. Survey, vol. 26, 1896, p. 47, Pl. IX, figs. 5-8.
Pinus raritanensis Berry, Bull. Torrey Bot. Club, vol. 37, 1910, p. 189.
Description.-This species was discovered in the upper part of the Raritan formation of
South Amboy, N. J. The remains consist of slender leaves in fascicles of threes and of poorly
preserved winged seeds. Similar remains occur in the Magothy formation of New Jersey and
in the Black Creek formation of North Carolina. They are too indefinite to have much strati-
graphic value and are of slight botanic interest beyond showing the presence of a pinelike form
along the Upper Cretaceous Atlantic coast. In this connection attention should be called to
structural material of Pinus from the Raritan formation on Staten Island, N. Y., described by
Hollick and Jeffrey 3 as Pinus triphylla, which may be identical with the present form..
Occurrence.-Middendorf arkose member of Black Creek formation, near Middendorf,
Chesterfield County. (Collected by E. W. Berry.)
Collections.-U. S. National Museum.
I A number of these records are unpublished.
* Velenovsky, J., Die Gymnospermen der bShmischen Kreideformation, 1885, p. 16, Pl. VI, figs. 3, 6-8; Kv6tena beskebo cenomanu, 1889. p. 9,
PI. I, figs. 11-19; Pl. II, figs. 1, 3.
a Mem. New York Bot. Garden, vol. 3, 1909, p. 14, Pls. III, figs. 6 and 7 (?), and XXII, fig. 1.





" :- UPPER CRETACEOUS FLORA OF SOUTH CAROLINA. 28

Family TAXODIEACES.
Genus SEQUOIA Endlicher.

SEQUOIA REICHENBACHI (Geinitz) Heer.'
Plate IV, figures 1-4.

reichenbachi Geinitz, Characteristik der Schichten und Petrefakten des sachs.-blhmiechen Kreidegebirges,
No. 3, 1842, p. 98, Pl. XXIV, fig. 4.
primtra Corda, in Reuss, Versteinerungen der b6hmischen Kreideformation, Abth. 2, 1846, p. 89, PI.
* XLVIII, figs. 1-11.
zertacea Endlicher, Synopsis coniferarum, 1847, p. 281.
K reichenbachi Heer, Flora fossilis arctica, vol. 1, 1868, p. 83, Pl. XLIII, figs. ld, 2b, and 5a.
a reichenbachi Heer, idem, vol. 3, 1874, pp. 77, 101, and 126; Pls. XII, figs. 7c and 7d; XX, figs. 1-8; XXVIII,
.. fig. 2; XXXIV, fig. 1; XXXVI, figs. 1-8; and XXXVII, figs. 1 and 2.
t reichenbachi Lesquereux, The Cretaceous flora, 1874, p. 51, Pl. I, figs. 10-10b.
a reichenbachi Heer, Flora fossilis arctica, vol. 6, Abth. 2, 1882, p. 52, PI. XXVIII, fig. 7.
* reichenbachi Fontaine, The Potomac or younger Mesozoic flora: Mon. U. S. Geol. Survey, vol. 15, 1890, p. 243,
Pls. CXVIII, figs. 1 and 4, and CXIX, figs. 1-5, etc.
couttsix Hollick, Trans. New York Acad. Sci., vol. 12, 1892, p. 30, Pl. I, fig. 5.
reichenbachi Lesquereux, The flora of the Dakota group: Mon. U. S. Geol. Survey, vol. 17, 1892, p, 35, PI. II,
fig. 4.
a reichenbachi Hollick, Trans. New York Acad. Sci., vol. 12, 1892, p. 30, Pl. I, fig. 18.
ia reichenbachi Nathorst, in Felix and Lenk, BeitrAige zur Geologie und Paleontologie der Republik Mexico, 1893,
pt. 2, no. 1. -
uot a reichenbachi Newberry, The flora of the Amboy clays: Mon. U. S. Geol. Survey, vol. 26, 1896, p. 49, Pl. IX,
fig. 19.
reichenbachi Berry, Bull. New York Bot. Garden, vol. 3, 1903, p. 59, P1. XLVIII, figs. 15-18.
reichenbachi Berry, Bull. Torrey Bot. Club, vol. 31, 1904, p. 69, Pl. IV, fig. 8.
ua reichenbachi Berry, idem, vol. 32, 1905, p. 44, Pl. I, fig. 3.
reichenbachi Knowlton, Bull. U. S. Geol. Survey, No. 257, 1905, p. 157, P1. XIV, figs. 3-5.
reichenbachi Hollick, The Cretaceous flora of southern New York and New England: Mon. U. S. Geol. Survey,
vol. 50, 1907, p. 42, Pla. II, fig. 40, and III, figs. 4, 5.
u reichenbachi Knowlton, Smithsonian Misc. Coll., quarterly issue, vol. 4, 1907, p. 126, Pl. XII, figs. 7 and 8.
finitzia reichenbachi Hollick and Jeffrey, Mem. New York Bot. Garden, vol. 3, 1909, p. 38, Ple. V, figs. 7-10; VIII,
figs. 3 and 4; XVI, figs. 2-4; XVII, figs. 1-4; XVIII, figs . 1-4.

Description.-The following description was written by Heer in 1868: "S. ramis elongatis,
liis decurrentibus, patentibus, falcato-incurvis, rigidis, acuminatis."
SThis species has a recorded range on this continent from the Neocomian of Mexico to the
ivingston formation of Montana, being very abundant at numerous horizons, and it has like-
ise been identified from Greenland and Europe. Many investigators have held that some,
least, of these identifications are erroneous, which is probable enough, although the Tertiary
quoia langsdorfii has an almost equally wide range, both geologic and geographic.
In a remarkable memoir recently published Hollick and Jeffrey 2 have described the anatomy
1 some twigs of the Sequoia reichenbachi type from the Raritan formation of Staten Island.
Lcording to these authors their results indicate that these remains are Araucarian in their
finity, a view which has been tentatively suggested by numerous students since the days of
initz, who referred them to the genus Araucarites. In order to make out a good case these
ors were under the necessity of finding Araucarian characters in certain associated cone
es of the sequoia type, as these supposed Araucarian twigs are frequently found with Sequoia-
cones attached to them. They refer these cone scales to new genera which they term
einitzia and Pseudogeinitzia, although the evidence for an Araucarian affinity is in the
writer's opinion extremely slender.
As might be expected from their great range, fossils of the Seguoia reichenbachi type are of
ht stratigraphic value. Nevertheless, the remains are very abundant at the Magothy-
ddendorf-late Tuscaloosa horizon, apparently identical in character and commonly cone-
I Only representative citations, chiefly American, of this widespread and persistent specees are given.
2 Mere. New York Dot. Garden, vol. 3, 1909, p. 38, PI. V et seq.





24 UPPER CRETACEOUS AND EOCENE FLORAS OF SOUTH CAROLINA AND GEORGIA.

bearing, the cones being small prolate spheroids consisting of relatively few peltate umbilicate,
sequoia-like scales.
This species is confined to the Middendorf locality in South Carolina, where it is excessively
abundant and commonly cone-bearing. Sequoia twigs are very resistant to maceration and
are in many places about the last vegetable remains to disintegrate in marine waters; never-
theless, the excellent preservation at Middendorf of leafy branches of large size with cones
attached indicates quiet water and nearness to place of growth. The species is rare in the
Raritan formation but common in later Upper Cretaceous outcrops in New Jersey, Delaware,
Maryland, North Carolina, Georgia, and Alabama.
Occurrence.-Middendorf arkose member of Black Creek formation, near Middendorf,
Chesterfield County. (Collected by E. W. Berry.)
Collections.-U. S. National Museum.

Genus CTNNINGIAITEBS Preal.

CUNNINGHAMITES ELEGANS (Corda) Endlicher.
Ounninghamia elegant Corda, in Reuss, Versteinerungen der bbhmischen Kreideformation, pt. 2, 1846, p. 93, Pl.
XLIX, figs 29-31.
Cunninghamites elegant (Corda) Endlicher, Synopsis Coniferarum, 1847, p. 305.
Cunninghamites elegans Heer, Die Flora von Moletein in Mfihren, 1869, p. 12, Pl. I, fig. 14.
Cunninghamites elegans Schimper, Paldontologie v6g4tale, vol. 2, 1870, p. 259.
Cunninghamites squamous Hosius and Von der Marck (non Heer), Die Flora der Westfalischen Kreideformation
Palheontographica, vol. 26, 1880, p. 179, Pl. XXXVII, figs. 137, 138.
Cunninghamites squamous var. densifolia Hosius and Von der Marck, idem, figs. 139-141.
Cunninghamites squamosus var. linearis IIosius and Von der Marck, idem, p. 180, fig. 142.
Cunninghamites elegant Heer, Flora fossilis arctica, vol. 7, 1883, p. 17, Pl. LIII, fig. 1.
Cunninghamia elegans Velenovsky, Die Gymnospermen der b8hmischen Kreideformation, 1885, p. 14, P1. IV, fig. 5;
PIl. V, figs. 1, 7; Pl. VI, fig. 5.
Cunninghamites elegans Hosius and Von der Marck, Nachtrag, Paheontographica, vol. 31, 1885, p. 227.
Cunninghamia elegant Velenovsky, Sitzungsber. K. b6hmischen Gesell., 1886, 1887, p, 634, figs. 1-5.
Cunninghamia elegans Kerner, Jahrb. K.-k. geol. Reichsanstalt, vol. 45, 1896, pt. 1, PI. IV, fig. 4.
Cunninghamites elegans Newberry, The flora of the Amboy clays: Mon. U. S. Geol. Survey, vol. 26, 1896, p. 48, P1. V.
figs. 1-7.
Cunninghamites elegant Hollick, Trans. New York Acad. Sci., vol. 16, 1897, p. 129, Pl. XI, fig. 2.
Cunninghamites (?) sp., Knowlton, Bull. U. S. Geol. Survey No. 163, 1900, p. 29, PI. V, fig. 3.
Cunninghamites elegans (?) Fliche, Bull. Soc. sci. Nancy, 1900, p. 10, PI. I, fig. 1.
Cunninghamites elegans Hollick, Bull. New York Bot. Garden, vol. 2, 1902, p. 402, rl. XLI, fig. 11.
Cunninghamites elegans Berry, idem, vol. 3, 1903, p. 64.
Cunninghamites elegans Zeiller, Annales des mines, March, 1905, p. 15, PIl. VII, fig. 4.
Cunninghamites elegans Knowlton, Bull. U. S. Geol. Survey No. 257, 1905, p. 135, PIl. XV, fig. 1.
Ounninghamites elegans Hollick, Mon. U. S. Geol. Survey, vol. 50, 1907, p. 41, Pl. III, fig. 1.
Cunninghamites elegans Berry, Bull. Torrey Bot. Club, vol. 31, 1904, p. 70, Pl. III, figs. 7-9, 11.
Description.-The following is the description by Corda in 1846:
C. ramis gracilibus teretibus, pulvinulis foliorumr homboideo-hexagonis, longitudinaliter carinatis; cicatricibus
terminalibus oblique-transversis; foliis hamato-arrectis, attenuatis, integerrimus, acutis, medio nervo tenui simplici.
The type locality for this handsome species was the " Untern Quader von Masseno bei Schlan
in BRhm." Since its description by Corda it has been recorded from a number of European
and American localities. Abroad it is found from the Cenomanian to the Emscherian in Bohemia,
Moravia, Westphalia, Dalmatia, and Bulgaria. In America it occurs in the Magothy formation
from Marthas Vineyard to New Jersey, ranging northward to the Patoot beds of Greenland.
In the West it has been found in the Montana group, both in Montana and in Wyoming. Excep-
tionally large forms of this species are abundant in the upper beds of Black Creek formation of
North Carolina, and the species is also present at a somewhat higher Upper Cretaceous horizon
near Byron, Ga. Abroad the species has been recorded by Schenk from the Urgonian and by
Fliche from the Barremian, but the latter identification is probably and the former almost
certainly incorrect.






SPPlR CRETACEOUS FLORA OF .SOUTH CAROLINA. 25

9The South Carolina material is scanty and fragmentary. It comes from the Rocky Point
ity and is queried because the ferruginous replacement renders the identification uncertain.
Remains are those of long, rather slender, curved leave of the Cunninghamite elegant
the single specimen and its counterpart representing the distal portion of a twig about
-normal size for this species but somewhat smaller than the specimens from the Black Creek
aon of North Carolina.
.The genus Cunninghamites was proposed in 1838 by Presl 1 in Sternberg's great work, and
ngamites oxycedrus from the Quader of Niederschoena in Saxony was designated the
-by Brongniart 2 in 1849. Several fossil species of Cunninghamites have been described,
recently structural material of a cone very close to that of the existing Cunninghamia has
described 3 from the Upper Cretaceous of Japan. The existing species of Cunninghamia,
in number, inhabit the subtropical uplands of the Orient.
Oceurrence.-Middendorf arkose member of Black Creek formation, Rocky Point, Sumter
ty. (Collected by L. F. Ward and L. C. Glenn.)
Collections.-U. S. National Museum.

Family CUPRESSEACEE.
Genus WIDDRINGTONITES Endlicher.

WIDDRINGTONITES SUBTILIS Heer.
Plate II, figures 14-17.

ites subtilis Heer, Flora fossils arctica, vol. 3, Albth. 2, 1874, p. 101, Pl. XXVIII, fig. 1, b.
*tonites subtilis Heer, idem, vol. 6, Abth. 2, 1882, Pis. VII, figs. 13 and 14, and XXVIII, fig. 4, b.
nites subtilia Newberry, The flora of the Amboy clays: Mon. U. S. Geol. Survey, vol. 26, 1896, p. 57, Pl. X,
figs. 2-4.
items reichii Hollick, Annals New York Acad. Sci., vol. 11, 1898, p. 58, Pl. III, fig. 8.
lonites subtilis Hollick, The Cretaceous flora of southern New York and New England: Mon. U. S. Geol.
Survey, vol. 50, 1907, p. 45, Pl. IV, figs. 2-5.
Srtonites reichii Berry, Johns Hopkins Univ. Cire., new ser., No. 7, p. 81, 1907.
tngtonites subtilis Berry, Bull. Torrey Bot. Club, vol. 39, 1912, pp. 341-348, Pls. XXIV, XXV.
Description.--This species was described from the Atane beds of Greenland by Heer in
4. His material was, however, extremely limited. Subsequently it was found in consid-
-le abundance in the Raritan formation of New Jersey and still more recently Hollick has
rded it from Marthas Vineyard and Block Island (Magothy formation). It may be questioned
some of the coniferous material described by Velenovsky from the Bohemian Cretaceous
pder other names should not be compared with the present form. It is even more slender
1an Widdringtonites reichii, with much shorter twigs, which have the appearance of having
een somewhat lax in habit. The leaves are s ally more elongated, close set, and appressed,
prrowly lanceolate, straight and scalelike; thel re said by Heer to be somewhat spread and
Hcate proximad. Remains of this latter sort occur in the Tuscaloosa formation of Alabama
Ipd the Magothy formation of Maryland.
: Newberry mentions a vague cone about 1 centimeter in diameter as included in the Raritan
trial. The writer has not seen this specimen but has found a number of poorly preserved
ached cones among the abundant remains of this species in the Cretaceous beds of South
ina and a number of specimens with attached cones from the Tuscaloosa formation in
t Alabama, where this species is exceedingly abundant at certain localities. These
are terminal, roughly spheroidal in outline, and apparently consist of four thick scales
wide blunt tips and somewhat extended bases. They are closely comparable to the cones
the Cretaceous of eastern Europe ascribed to Widdringtonites reichii by Velenovsky and
r. (See Pl. II, figs. 18 and 19.) These authors refer this form directly to the genus
I Sternberg, Kaspar, Flora der Vorwelt, vol. 2, Nos. 7 and 8, 1838, p. 203.
Brongniart, Tableau, 1849, p. 68.
3 Stopes and Fujii, Phil. Trans. Roy. Soc. London, vol. 201 B, 1910.






I .. -�.






26 UPPER CRETACEOUS AND EOCENE FLORAS OF SOUTH CAROLINA AND GEORGIA.
Widdringtonia, and it would seem that the cones attached to the Alabama specimens of Wid--
dringtonites subtilis conclusively demonstrate its relationship with some of the modern species
included in the genus Callitris by Eichler.
Occurrence.-Middendorf arkose member of Black Creek formation, near Middendorf,
Chesterfield County; Rocky Point, Sumter County. (Collected by E. W. Berry.)
Collectione.-U. S. National Museum.
Genus MORICONIA Debey and Rttingahausen.
MORICONIA AMERICANA Berry.
Plate VII, figures 1-4.
Moriconia cyclotoxon Berry, Bull. New York Bot. Garden, vol. 3, 1903, p. 65, Pls. XLIII, fig. 4, and XLVIII, figs. 1-4
(non Debey and Ettingshausen).
Moriconia cydotoxon Berry, Bull. Torrey Bot. Club, vol. 31, 1904, p. 70; vol. 33, 1906, pp. 165-167.
Moriconia americana Berry, idem, vol. 37, 1910, pp. 20, 180.
Description.-Leafy twigs, apparently deciduous in habit, bifacial, phylloclad-like, consist-
ing of cyclically arranged reduced leaves. Along the main axis on each flat face of the branch
these leaves are relatively and closely appressed, with a narrow base and a broad semicircular
apex. The coiresponding lateral pairs of leaves are thin and pointed and transversely compressed.
In the axis of each of these marginal leaves is a reduced branch flattened in the same plane as the
main branch, so that the whole arrangement is strictly opposite and distichous. These reduced
lateral branches have leaves of the same character and arrangement as those of the main branch.
The bifacial leaves are, however, somewhat smaller and blunter and the marginal leaves are
broader and less acute. In a short distance they become smaller distad, generally not more
than five or six pairs being required to complete the blunt lateral reduced twigs. The main
vascular axis is stout and in some specimens a vascular axis can be made out in the lateral
branches. The leaves do not show any veins. The texture was apparently coriaceous, but
from the appearance of the majority of specimens the leaves were thin. No structural material
or indications of fruits or fruiting characters have been discovered. This species, formerly
confused with Moriconia cyclotoxon of Debey and Ettingshausen, differs from the latter, which
is the type and only other known species of the genus, in being more phylloclad-like and strictly
comparable to a cupressineous genus like Libocedrus. It is also about twice as large, the lateral
twigs are more reduced, and the main axis is invariably leafy. It differs also in its geologic
range, the two species not being anywhere contemporaneous in America, although the type in.
Europe extends as high as the later larger form of America.
Superficially these remains closely resemble fragments of fern fronds. In fact, Debey, the
original discoverer, always insisted that they were ferns and Heer described the earliest collected
and poorly preserved remains from Greenland as a species of Pecopteris. There can be no
doubt, however, of their gymnospermous naS. For stratigraphic determinations they are
one of the most characteristic fossil plants knOn, as the geometrically arranged outlines of the
leaves is recognizable with certainty in the smallest fragment.
They are strikingly like the curious genus Androvettia, which was recently described by
Hollick and Jeffrey 1 and referred by them to the Araucariese, although Moriconia has, on the
evidence of the foliar characters, been invariably referred to the Cupressinese. The present
species is common at the Midd.endorf locality in South Carolina and is a characteristic post-
Raritan species in the Atlantic Coastal Plain, having been recorded by the writer from numerous
localities in the Magothy formation of New Jersey, Delaware, and Maryland, and from the
Black Creek formation in North Carolina. Moriconia cyclotoxon, the type of the genus, is
confined to the Raritan in this country, although it is found in both the Atane and Patoot
beds of western Greenland and came originally from the Senonian of Prussia.
Occurrence.-Middendorf arkose member of the Black Creek formation, near Middendorf, -
Chesterfield County. (Collected by L. W. Stephenson and E. W. Berry.)
Collections.-U. S. National Museum.
I Hollick, Arthur, and Jeffrey, E. C., Mem. New York Bot. Garden, vol. 3, 1909, p. 22, P1. III, figs. 1-5, etc.






UPPER CRETACEOUS FLORA OF SOUTH CAROLINA. 2Z
CONIFEBM INCEBRT SBDIS.
Genus STROBIELITES Lindley and Hutton.
STROBILITES ANCEPS sp. nov.
Plate III, figure 5.
cripton.--Cone a prolate spheroid in shape, about 3 to 4 centimeters in length by 2.4
in diameter, made up of spirally arranged, rather thick, flat, obovate scales. The
n of preservation of these cones, of which two were collected, is such that their generic
tribal affinity can not be determined; hence they are referred to the comprehensive
us Strobilites. They are not of the Sequoia or Widdringtonia type and resemble
cones of the Pinus more than they do any of the other coniferous genera represented by
in the South Carolina Cretaceous. They are similar to the much more elongated cones
e Lower Cretaceous of the Atlantic Coastal Plain, which have been referred to the form
iAbietites, and are not at all like the assorted conelike remains described by Hollick and
From the Raritan formation as forms of Strobilites. As regards size the South Carolina
Share comparable to Strobilites inquirendus Hollick from the Magothy formation of New
, but the preservation of the latter is so poor that the specimens are almost valueless.
comparison which is entitled to some weight may be made with the cones described by
ausen from the Upper Cretaceous of Saxony and identified as Cunninghamites stern-
.1 Whatever may be the true botanic affinity of these fossils, they are almost certainly
eric with those from South Carolina with which they are in close agreement.
Occurrence.-Black Creek formation, right bank of Black Creek, 1 miles east of Darlington,
ington County. (Collected by L. W. Stephenson.)
Collections.-U. S. National Museum.
Genus HETEROLEPIS gen. nov.
HETEROLEPIS CRETACEUS sp. nov.
Plate III, figure 3.
Description.-Cone scale of large size, about 3 centimeters in length, with a stout, woody
5 millimeters wide at the base and 1.3 centimeters wide where it expands to form a bosslike,
tly umbilicate tip which is evenly rounded, forming a right angle with the axis; in outline
*st circular, with a slightly irregular striated margin, 1.7 centimeters in diameter.
'This species is based upon the single specimen figured, which is so characteristic that it is
to be a valuable stratigraphic fossil, although its botanic affinity can not be determined.
nay be cycadaceous or it may belong to a large-coned species comparable with the remains
nmonly referred to Sequoia or Geinitzia.
Occurrence.-Middendorf arkose member of Black Creek formation, Rocky Point, Sumter
mty. (Collected by L. C. Glenn.)
C Oollections.-U. S. National Museum.
-. Class ANGIOSPERMS.
Subclass MONOCOTYLEDONS.
Order NAIADALES.
Family NAIADACEE.
- Genus POTAMOGETON Linn6.
POTAMOGETON MIDDENDORFENSIS sjp. nov.
Plate IV, figure 6.
Description.-Leaves of small size, entire, obovate lanceolate or spatulate in outline, about
pitimeters in length by 1.45 centimeters in greatest width, which is in the apical half of the
. Petiole short and broad. Venation fine, immersed, acrodrome.
I Ettingshausen, C. von, Die Kreideflora von Niederschoena in Sachsen, 1867, p. 12, PI. I, figs. 4-6.
4*-





28 UPPER CRETACEOUS AND EOCENE FLORAS OF SOUTH CAROLINA AND GEORGIA.

This species is based upon the single specimen figured and its counterpart, which may be
interpreted as the remains of two leaves in juxtaposition or of a single leaf in which the upper
and lower cuticle has parted company. The latter appears to be the most reasonable explana-
tion. The species was evidently aquatic, and it is very close to certain modem species in this
genus. It also suggests the genus Pistia, of which a typical species is common in the Black
Creek formation of North Carolina. It is quite distinct from the latter, however, and seems
to be an immersed leaf of a Potamogeton in the sum of its characters.
Occurrence.-Middendorf arkose member of Black Creek formation, near Middendorf,
Chesterfield County. (Collected by E. W. Berry.)
Collections.-U. S. National Museum.

Order POALES.
Family POACEM.
Genus ARTUNDO Ldnn6.
AROUND GRdZBLANDICA Heer (1).
Plate IV, figure 7.
Arundo grcenlandica Heer, Flora fossilis arctica, vol. 3, Abth. 2, 1874, p. 104, Pl. XXVIII, figs. 8-11.
Arundo grenlandica Heer, idem, vol. 6, Abth. 2, 1882, p. 57, PIl. XVII, fig. 10.
Arundo granlandica Heer, idem, vol. 7, 1883, p. 18, Pl. LIV, figs. 1-3.
Description.-Fragments of long, linear, pointed leaves, about 2.5 centimeters in greatest
width. Primaries numerous, parallel, fine, about 2.5 millimeters apart, separated by numerous
very fine parallel secondaries; the central primary apparently somewhat enlarged to form a
midrib.
In the Greenland material this species included culms as well as leaf fragments. The
South Carolina material is queried because it differs from the type in having a midrib, which is
not, however, so prominent a feature as the drawings indicate, having been accentuated by the
method of preservation, replacement by iron oxide. It may be compared with various described
fragments of monocotyledonous leaves referred to this genus and to Phragmites and a variety
of other genera of the Poales. There is the further possibility that these remains may represent
fragments of the rays of some palm. They are of slight botanic interest but are present in
well-marked specimens at the Rocky Point locality and may be found to possess considerable
stratigraphic value.
Occurrence.-Middendorf arkose member of Black Creek formnnation, Rocky Point, Sumter
County. (Collected by L. C. Glenn.)
Collections.-t. S. National Museum.
Genus PHRAGMITES Trinius.
PHRAGMITES PRATTII Berry.
Phragmites sp. Berry, Bull. Torrey Bot. Club, vol. 34, 1907, p. 190, Pl. XI, fig. 5.
Phragmites prattii Berry, idem, vol. 37, 1910, p. 191.
Description.-Parallel-veined, monocotyledonous leaf fragments, indicating grasslike leaves
1 centimeter or more in width, having about ten equal parallel veins and a few fine transverse
veinlets. Leaf substance thin.
This species is strictly comparable with a large number of fossils that have been referred
to Phragmites, which must be regarded as purely a form genus for the reception of fragments
of the leaves of grasses or sedges. The present species, which has been detected at a number
of localities in the Black Creek formation of North Carolina, has been collected only at the
Rocky Point locality in South Carolina, where small fragments are common. It occurs also in
the basal beds of the Eutaw formation along Chattahoochee River.
Occurrence.-Middendorf arkose member of Black Creek formation, Rocky Point, Sumter
County. (Collected by L. F. Ward and L. C. Glenn.)
OcIlections.-U. S. National Museum.





S FFs CRXTACEOUS FLORA OF SOUTH CAROLINA. 29

" *Family CYPERACES.
- Genus CAREX Linn6.
CARE CLARKII Berry.
, Am. Naturalist, vol. 39, 1905, p. 347, fig. 1.
, Ann. Rept. State Geologist New Jersey for 1905, 1906, pp. 138-141.
, Bull. Torrey Bot. Club, vol. 33, 1906, p. 169.
. , Johns Hopkins Univ. Circ., new ser., 1907, No. 7, p. 81.
*o.-Leaf fragments, the largest being 6 centimeters in length, varying in width
eters to 4 millimeters, averaging between 2 millimeters and 3 millimeters, slightly
ming thicker and narrower proximad. Midribmoderately prominent. Lateralveihs,
parallel with it, very fine and scarcely discernible except in the larger specimens.
mon with other fossil remains of grasses and sedges this species has no botanic
pt as an indication of the presence of plants of these families. It has, however, like
fossils of vague botanic affinities, considerable stratigraphic value, as it is found to
oe the Magothy formation at a large number of outcrops from New Jersey to Mary-
e South Carolina remains are not abundant and are confined to the Rocky Point

ence.-Middendorf arkose member of Black Creek formation, Rocky Point, Sumter
. (Collected by L. C. Glenn.)
.-U. S. National Museum.
Order ARECALES.
Family PALMX.
Genus SABALITEB Saporta.
SABALITES CAROLINENSIS sp. nov.
Plates V and VI.
pion.-A fan palm with very large flabellate leaves. Rays numerous, keeled
d, becoming nearly flat distad, ultimately splitting more or less. Petiole not pre-
in any of the collected material, which includes abundant but fragmentary specimens.
es stout and prominent, as much as 1.2 centimeters apart in the largest specimen.
ries usually seven in number between each pair of primaries, with which they are
imately parallel; thin, connected by numerous ill-defined cross veinlets.
Swas evidently a very large Sabal-like leaf, of which the largest collected specimens
o figured. All the material comes from the single locality near Langley, though very
agents of what appears to belong to a palm have been detected at several other locali-
Sthe South Carolina Cretaceous. These may represent this species, but they are too
tentative to be even tentatively identified with it.
he enormous number of existing palms, which includes considerably more than one
ad species, is about equally divided between the oriental and occidental tropics, with
onotypic genera, showing well the marked effects of geographic distribution and isola-
the formation of species. There are no outlying forms, the highest northern latitude
Being about 430 in Europe and the highest southern latitude about 450 in New Zealand.
luereux writing in 1878 1 records fossil palms in 520 north latitude in both America
" Since then remains have been described from as far north as 800 (Grinnell Land,
gen), and two fine species are recorded from the earlier Tertiary of Greenland (latitude
Variety of Paleozoic remains has been referred to the Palmse, these remains ranging
Sbotanic affinities from Stigmaria trunks to Cordaitean leaves and fruits. The nature
ter was first rightly conjectured by Brongniart in 1828. With the marvelous increase,
Last 25 years, in knowledge of the vegetation of the Paleo-zoic, it can now be posi-
ed that palms are unknown from pre-Mesozoic formations.
I The Tertiary flora, 1878, p. 109.






80 UPPER CRETACEOUS AND EOCENE FLORAS OF SOUTH CAROLINA AND GEORGIA.

Stenzel, who has recently monographed I the fossil palm wood of the world, finds the
oldest known wood to come from the Turonian of France (one species), the succeeding Senonian
terrane yielding six species. At the beginning of the Tertiary period the species became
numerous.
Undoubted remains of palm leaves occur somewhat earlier, the middle part of the
Cretaceous, in the light of present -knowledge, marking the introduction of the type. The
Cenomanian of Europe has furnished undoubted palm leaves (upper Cenomanian of Tiefenfurth
in Silesia), and Stur = has described fruit from that horizon in Bohemia. Fliche has described
three species in two genera from a similar horizon in France.' The Senonian deposits show
species in a variety of genera. It is in the Tertiary, however, that palms become greatly devel-
oped and widespread, and ,the numerous species found on evidence afforded by stems, leaves,
petioles, fruits, and even flowers are referable to a large number of genera (Geonoma, Manicaria,
Calamopsis, Thrinax, Phoenix, Nipa, Chamaerops, Oreodoxites, Sabal, Iriartea, Latanites, and
the like). 'In this country the earliest known remains are those small fragments of striated
leaves, of a rather doubtful nature, which Lesquereux described 4 as Flrellaria minima from
the Dakota sandstone.'
The Montana group of Senonian age has furnished Knowlton ' with the undoubted remains
of a large palmetto-like form (Sabalites),7 and the Laramie formation furnishes a number of
species, some of which, represented by both leaves and fruit, continue into the Eocene.
The present species may be compared with Sabalites magothiensis Berry,' which is found
in the Magothy formation from Raritan Bay in New Jersey to Severn River in Maryland. The
two species are entirely distinct, however, and the South Carolina form is much better charac-
terized and represented by more complete material.
Occurrence.-Middendorf arkose member of Black Creek formation, near Langley, Aiken
County. (Collected by Earle Sloan, E. W. Berry, and L. W. Stephenson.)
Collections.-U. S. National Museum.

Subclass DIOOTYLEDON&.
Order JUGLANDALES.
Family JUGLANDACES.
Genus TUGLANB Linni6.
JUGLANS ARCTICA Heer.
Plate VIII, figures 1 and 2.
Juglans arctica Heer, Flora fossilis arctica, vol. 6, Abth. 2, 1882, p. 71, Pla. XL, fig. 2; XLI, fig. 4c; XLII, figs. 1-3,
and XLIII, fig. 3.
Juglans arctica Lesquereux, The flora of the Dakota group: Mon. U. S. Geol. Survey, vol. 17, 1892, p. 68, Pls. XIX,
fig. 3, and XXXIX, fig. 5.
Juglans arctica Newberry, The flora of the Amboy clays: Mon. U. S. Geol. Survey, vol. 26, 1896, p. 62, Pl. XX, fig. 2.
Juglans arctica Hollick, Annals New York Acad. Sci., vol. 11, 1898, p. 58, P1. III, fig. 7.
Ficus atavina Hollick, Trans. New York, Acad. Sci., vol. 11, 1902, p. 103, Pl. IV, fig. 5.
Juglans arctica Berry, Ann. Rept. State Geologist New Jersey for 1905, 1906, p. 139, Pl. XXI, fig. 1.
Juglans arctia Berry, Bull. Torrey Bot. Club, vol. 33, 1906, p. 170.
Juglans arctica Hollick, The Cretaceous flora of southern New York and New England: Mon. U. S. Geol. Survey, vol.
50, 1907, p. 54, Pl. IX, figs. 6-8.
Description.-The following is the description of Heer, written in 1882:
I. Nuce ovali, 34 mm. long, 17 mm. lata; foliis magnis, foliolis ovalibus, basi inaequilateralibus, integerrimis,
nervo medio valido, nervis secundariis angulo semirecto egredientibus, curvatis.
I Beitr. Pal. v. Geol. Oesterr.-Ungarn., 1904, pp. 1-182, Pis. I-XXII.
I Verhandl. K.-k. Geol. Rleichsanstalt Wien, 1873.
Siltudes sur la flore fossile de I'Argonne, 1896.
* Cretaceous flora: 1874, p. 56, PI. XXX, fig. 6.
* It is now definitely decided that HRollick's supposed palm, Serenopsis, from the CretacecJus of Long Island, is a Nelumbo.
* Bull. U. S. Geol. Survey No. 163, 1900, p. 32.
7 Dawson has also described a Sabal from the Upper Cretaceous at Nanaimo.
a Torreya, vol. 5, 1905, p. 32.






UPPER CRETACEOUS FLORA OF SOUTH CAROLINA. 81

The leaves of this species vary considerably in size and outline, as might be expected in
this genus. Heer's type material is somewhat imperfect and some of it is difficult to distinguish
from some of the forms referred to the same author's Jugladn crassipes, although the latter is
on the whole a much larger less oblong form with a narrower base. Juglana arctica is oblong-
ovate in outline, with an obtusely pointed apex and a rounded, generally inequilateral base.
The petiole and midrib are stout. Secondaries numerous, well marked, parallel, camptodrome.
Size varies in complete specimens from 9 to 15 centimeters in length and from 3 to 6 centimeters
in width.
A nut and catkins are associated with the leaves at the type locality in the Atane beds of
Greenland, which confirm their reference to this genus. A single leaf is recorded from the
Raritan formation of New Jersey, and the species also occurs in beds of this age on Staten Island.
With these exceptions the species is confined to later horizons, occurring in the Magothy forma-
tion on Marthas Vineyard, Block Island, and in New Jersey, in the Black Creek formation of
North Carolina, and in the Middendorf arkose member of the Black Creek formation of South
Carolina. In the West this species occurs in the Dakota sandstone of Kansas. Its occurrence
in South Carolina is based on the basal halves of characteristic inequilateral leaves of rather
small size.
Occurrence.-Middendorf arkose member of Black Creek formation, near Middendorf,
Chesterfield County, Rocky Point, Sumter County. (Collected by E. W. Berry.)
Collections.-U. S. National Museum.
Order MYRICALES.
Family MYRICACEAE.
Genus MYRICA De Candolle.
MYRICA BRITTONIANA Berry.
Plate VII, figures 17 and 18.
Myrica heerii Berry, Am. Naturalist, vol. 37, 1903, p. 682, figs. 7, 8 (non Boulay).
Myrica brittoniana Berry, Bull. Torrey Bot. Club, vol. 32, 1905, p. 46.
Description.-Leaves elongate-lanceolate in outline, 13 to 14 centimeters long by 2.7
'centimeters in greatest width, which is in the middle part of the leaf. Apex elongated, narrowed,
bluntly pointed. Base attenuated. Margin entire, or entire below and undulate or distantly
and obtusely toothed above. Texture coriaceous. Petiole and midrib fairly stout. Secondaries
thin, immersed, generally obsolete, branching from the midrib at rather large angles, com-
rparatively straight, camptodrome.
This striking species was described by the writer in 1903 from the Magothy formation of
'New Jersey, to which it has been heretofore confined. Fragmentary leaves are, however,
resent in the deposits along Black Creek in South Carolina. The form is somewhat similar
species of the Dakota sandstone Myrica aspera Lesquereux.1
Occurrence.-Black Creek formation, right bank of Black Creek, 2 miles below Williamson's
idge, Florence County, (Collected by L. W. Stephenson.)
Collections.-U. S. National Museum.
MYRICA ELEGANT Berry.
Plate IX, figure 4.
Selegans Berry, Bull. Torrey Bot. Club, vol. 34, 1907, p. 191, PIl. XI, figs. 1-4, 6.
? Description.-Leaves ovate-lanceolate in outline, variable in size, ranging from 3.2 to 9.5
meters in length by 1.5 to 3 centimeters in greatest width, which is midway between the
and the base. Like the living species of Myrica the margin is variable, ranging from forms
*ch it is rather angularly crenate with an approach to serrate in some of the teeth, through
Sin which the crenations become more and more rounded, until the extreme of variation
S Leequereux, Leo, The flora of the Dakota group: Mon. U. S. Geol. Survey, vol. 17, p. 66, 1892, Pl. II, fig. 11.





32 UPPER CRETACEOUS AND EOCENE FLORAS OF SOUTH CAROLINA AND GEORGIA.

in this direction shows relatively broad leaves with deeply scalloped subcrenate margins.
Midrib straight and fairly stout. Secondaries numerous, 5 to 17 pairs, relatively stout and
prominent on the lower surface of the leaf, subopposite or alternate, equidistant and parallel,
branching from the midrib at angles of more than 45�, approaching 90� in some specimens.
They are nearly straight and craspedodrome, running to the marginal teeth. Tertiaries usually
not seen because of the coarse matrix. Some transverse veinlets are discernible, as well as
curved branches from the distal part of the secondaries, which run to the subordinate teeth of
the marginal scallops when these are developed. Apex pointed. Base cuneate and generally
with an entire margin. Texture subcoriaceous.
This species is markedly distinct from any other Cretaceous Myrica. It is very abundant
in the Black Creek formation of the upper Cape Fear River in North Carolina, but has only
been found at Darlington in the South Carolina area, and there in a much macerated condition.
Although the resemblance is not close, Myrica elegans is more like Myrica prwecoz, described
by Heer from the Patoot beds of Greenland, than it is like any other species with which it has
been compared. The latter is a smaller leaf with a rounded apex, and the wide marginal
crenations lack the crenulations of the Carolina species.
Oceurrence.-Black Creek formation, near Darlington, Darlington County. (Collected by
L. W. Stephenson.)
Collection.-U. S. National Museum.
Order SALICALES.
Family SALICACRA.
Genus SALIX Linn6.
SALIX FLEXUOSA Newberry.
Plates VII, figures 14-16, and XI, figure 1.
Salixflexuosa Newberry, Notes on the later extinct floras of North America: Ann. Lyc. Nat. Hist. New York, vol. 9,
1868, p. 21.
Salixflexuosa Newberry, Illustrations of Cretaceous and Tertiary plants, 1878, P1. I, fig. 4.
Salix protewafolia linearifolia Lesquereux, The flora of the Dakota group: Mon. U. S. Geol. Survey, vol. 17, 1892, p. 49,
PI. LXIV, figs. 1-3.
Salix proteefoliafleruosa Lesquereux, The flora of the Dakota group: Mon. U. S. Geol. Survey, vol. 17, 1892, p. 50, Pl.
LXIV, figs. 4, 5.
Salix protexfolia flexuosa Hollick, Bull. Torrey Bot. Club, vol. 21, 1894, p. 50, Pl. CLXXIV, fig. 5.
Salir protewfoliajflexuosa Hollick, Annals New York Acad. Sci., vol. 11, 1898, p. 59, Pl. IV, fig. 5a.
Salixflexuosa Berry, Ann. Rept. State Geologist, New Jersey, for 1905, 1906, p. 145.
Salizflexuosa Berry, Bull. Torrey Bot. Club, 1906, vol. 33, p. 171.
Salix proteefolia flezuosa Berry, Bull. New York Bot. Garden, vol. 3, 1903, p. 67, PI. XLVIII. fir 12, Pl T.IT fig 9.
Salix protexfolia linearifolia Hollick, The Cretaceous flora of southern New York and New England: Mon. U. S. Geol.
Survey, vol. 50, 1907, p. 52, Pl. VIII, fig. 12.
Salix protexfolia flextuosa Hollick, The Cretaceous flora of southern New York and New England: Mon. U. S. Geol.
Survey, vol. 50,1907, p. 51, Pls. VIII, figs. 5, 6a, XXXVII, fig. 8b.
Description.-Leaves narrow, linear-lanceolate in outline, equally pointed at both ends,
short petioled, ranging from 5 to 10 centimeters in length and from 8 to 13 millimeters in greatest
width. Margin entire. Midrib stout below, tapering above, commonly somewhat flexuous.
Secondaries more or less remote, about ten alternate pairs, branching from the midrib at angles
varying from 350 to 450, camptodrome, of, ine caliber, in many specimens often obsolete.
This species was described by Newberry from the Dakota sandstone in 1868. Lesquereux
subsequently made it one of the varieties of his Salix protea�folia, although it is obviously
entitled to independent specific rank. It is of rare occurrence in the Raritan formation, where
it first appears in the uppermost beds at South Amboy, N. J., and is preeminently a species
which characterizes the Magothy and homotaxial horizons to the southward. It is recorded
in beds of Magothy age from Marthas Vineyard to the Potomac. It occurs in the Black Creek
formation of North Carolina and at a large number of localities in the Tuscaloosa formation in
the western part of Alabama.
In South Carolina it has been found in considerable abundance in the Middendorf arkose
member of the Black Creek formation entirely across the State, and fragments which have been







wPPa OBETACBOUS rLOnA or SOUTH CABOLUNA. 88
as this species are present well toward the top of the Cretaceous on

Middendorf arkose member of Black Creek formation, near Middendorf,
; near Langley, Aiken County; right bank of Congaree River, about 25
bia, Lexington County. (Collected by E. W. Berry, L. W. Stephenson,
M. W. Twitchell.)
-0U. S. National Museum.

SALIX LESQUEREUXII Berry.
Plate VII, figures 11-13.
Lesquereux, Am. Jour. Sci., 2d ser., vol. 46, 1868, p. 94 (non Forbes).
L ux, The Cretaceous flora, 1874, p. 60, PI. V, figs. 1-4.
afquereux, in Cook and Smock, Report on the.clay deposit of Woodbridge, South Amboy, and
in New Jersey, 1878, p. 29.
. squereux, The Cretaceous and Tertiary floras, 1883, p. 42, PIe. I, figs. 14-16, and XVI, fig. 3.
.Leaquereux, The flora of the Dakota group: Mon. U. S. Geol. Survey, vol. 17, 1892, p. 49.
Iengifolia Lesquereux, idem, 1892, p. 50, Pl. LXIV, fig. 9.
longfolia Bartsch, Bull. Lab. Nat. Hist., Univ. Iowa, vol. 3, 1896.
' Newberry, The flora of the Amboy clays: Mon. U. S. Geol. Survey, vol. 26, 1896, p. 72 (pars),
fig. 11.
- Newberry, idem, p. 129 (pars), Pl. XLI, fig. 12.
SNewberry, idem, p. 66, Pl. XVIII, figs. 3, 4.
Kurtz, Revista Mus. La Plata, 1902, vol. 10, p. 51.
Berry, Ann. Rept. State Geologist, New Jersey, for 1905, 1906, p. 139.
f Berry, Bull. Torrey Bot. Club, vol. 33, 1906, p. 171, P1. VII, fig. 2.
Berry, idem, vol. 36, 1909, p. 252.
'ption.-Leaves ovate-lanceolate in outline, somewhat more acuminate above than
able in size, ranging from 6 to 12 centimeters in length and from 1.1 to 2.2 centi-
greatest width, which is usually slightly below the middle. Petiole stout, much
in Salix flexosa, the largest being 1.2 centimeters long. Midrib stout below,
above. Secondaries numerous, sometimes as many as twenty pairs; they branch from
at angles of about 450 and are parallel and camptodrome.
se an exceedingly variable species, as might be expected in a Salix, and Lesquereux
several varieties, of which at least one, linearifolia, is referable to Saliz fleuosa
Some of Lesquereux's forms are distinguishable with difficulty from the latter, as
cown" by examination of the leaves which he figures on Plate I of his "Cretaceous and
floras." These leaves are, however, larger and somewhat more robust, of a thicker
and broadest near the base, from which they taper upward to an exceedingly acuminate
general, Salix lesquereuii is a relatively much broader, more ovate form with more
Sand easily seen secondaries and a longer petiole.
ies is an exceedingly abundant Cretaceous type in both the East and the West.
tjo Coastal Plain it ranges from the base of the Raritan formation to the top of the
nation and is abundant in the Magothy, Black Creek, and Tuscaloosa formations.
it is common in the Dakota sandstone. It is one of the forms recorded by Kurtz
Cretaceous of Argentina, indicating, if the identification is correct, a very con-
tion during the early Upper Cretaceous.
dant in South Carolina in both the Middendorf member and other deposits of
Creek formation.
.-Middendorf arkose member of Black Creek formation, near Middendorf, Ches-
nty, near Langley and Miles Hall, Aiken County; Rocky Point, Sumter County;
of Congaree River, about 25 miles below Columbia, Lexington County. Other
the Black Creek formation near Darlington, Darlington County. (Collected by E. W.
:W. Stephenson, B. L. Miller, M. W. Twitchell, and Earle Sloan.)
.-U. S. National Museum.
1P-14-----





84 UPPER CETACEOUS AND OC MNE FLORAS OF SOUTH CAROLINA AND GEORGIA.

SALIX PSEUDOHAYEI Berry.
Plate X, figure 8.
Salix sp. Newberry, The flora of the Amboy clays: Mon. U. S. Geol. Survey, vol. 26, 1896, p. 68, Pl. XLII, figs. 6-8.
Salix pseud&hayei Berry, Bull. Torrey Bot. Club, 1909, vol. 36, p. 251.
Description.-Leaves small, relatively short and broad, ovate-lanceolate, uniformly about
3 centimeters in length by 1.1 to 1.4 centimeters in greatest width, which is about halfway
between the apex and base or lower apex acuminate. Base rounded. Margin entire. Petiole
short. Midrib slender and slightly curved. Secondaries fine, obscured in some specimens, five
or six pairs, alternate, camptodrome, branching from the midrib at an angle of about 450 and
curving upward.
This species is not uncommon in the Raritan formation of New Jersey, although Prof.
Newberry fails to mention the exact localities from which he collected it. Later material has
come from the lower part of the Raritan at Milltown, N. J. It has been compared with the
Dakota sandstone species, Salix hayei Lesquereux, and with the Arctic Tertiary Salix raeana
Heer, both of which it resembles in general appearance. The leaf from the Dakota sandstone,
however, is coriaceous, with a coarse venation, blunt apex, and more narrow pointed base, and
is seen to be quite different from the eastern species when careful comparisons are made.
The single specimen figured is all that represents this species in the South Carolina
Cretaceous. It is in every respect identical with the type material from New Jersey.
Occurrence.-Middendorf arkose member of Black Creek formation, near Middendorf,
Chesterfield County. (Collected by L. W. Stephenson and E. W. Berry.)
Collections.-U. S. National Museum.

SALIX SLOANI sp. nov.
Plate VIII, figures 10-12.
Description.-Leaves lanceolate in outline, large and elongated, 13 centimeters or more in
length by 1.6 to 3 centimeters in greatest width, which is about halfway between the apex and
the base, the leaf tapering gradually and equally distad and proximad to the narrowly acute
and extended apex and base. The margin is entire for a short distance below, above which it
is beset with short, triangular, generally obtusely pointed, somewhat aquiline teeth, separated
by rounded sinuses. Midrib stout. Secondaries very numerous, thin, approximately parallel,
branching from the midrib at wide angles, 600 to 700, at intervals of 2 to 3 millimeters, extending
outward with only a slight upward curve to a point near the margin and then curving upward
and extending for a greater or less distance almost parallel with the latter, giving off short
tertiary branches to the marginal teeth. Tertiaries fine, exceedingly numerous, transverse,
parallel at nearly right angles to the midrib.
This very striking and exceedingly distinct and characteristic species is not rare at the
locality near Langley, though few complete leaves are found. It has not been detected elsewhere
in the Atlantic Coastal Plain. It is much like certain modern species of Grevillea and other
antipodean members of the family Proteaccie, but may be compared with a number of
existing species of Salix, as, for example, Salixfluviatilis Nuttall. It is believed to be referable
with certainty to this genus and is one of the most ancient forms known wlich show decisive
characters identical with those of modern willow leaves. The most nearly related fossil species,
one perfectly distinct, however, is Salix eutawensis Berry,' which has been recorded from the
Black Creek formation in North Carolina and from the Eutaw formation in western Georgia.
Occurrence.-Middendorf arkose member of Black Creek formation, near Langley and
Miles Hall, Aiken County. (Collected by L. W. Stephenson, E. W. Berry, and Earle Sloan.)
Collections.-U. S. National Museum.
I Berry, Bull. Torrey Hot. Club, vol. 37, 1910, p. 193, Pl. XXII.






UPBB COEDTACDOUS ? MA OF SOUTH CAROLINA. 85
Order PAGAL.UB.
S Family FAGACEMJ.
Genus QUBBROUS Linn6.

QUERCUS SUMTERENSIS sp. nov.
Plate X, figures 9 and 10.

--Leaves large, linear, acuminate, coriaceous, entire, with a broadly cuneate
about 15 to 16 centimeters. Greatest width 3 centimeters. Petiole short and
stout. Secondaries thin, branching from the midrib at angles of 45� or more,
their straight course almost to the margin and then turning sharply upward. Ter-
te, as are also the secondaries in some specimens.
ecies is clearly referable to the willow or laurel oaks and is markedly distinct from
hitherto described, although it resembles in general outline some of the lanceolate
species of Ficus. It is not uncommon at the Rocky Point locality, to which it is
ed. It is not unlike some of the leaves of the modern Quercus rudkini, a hybrid
'narylandica and phellos..
.-Middendorf arkose member of Black Creek formation, Rocky Point, Sumter
(Collected by L. F. Ward and L. C. Glenn.)
.-U. S. National Museum.

QUERCUS PSEUDOWESTFALICA sp. nov.
Plate IX, figure 5.

ton.-Leaves of medium size, relatively small for this genus, ovate in general
with an acuminate apex and a cuneate base. Texture coriaceous. Length about 9.5
es. Greatest width about 4 centimeters near the middle of the leaf. Margin with
ate teeth separated by broadly rounded sinuses. Midrib stout. Secondaries thin,
lately parallel, about seven subopposite pairs, branching from the midrib at angles
it450, slightly curved, craspedodrome, ending in the marginal teeth, of which there is
ah secondary. Petiole not preserved.
1 species evidently belongs to the group of the chestnut oaks, not differing materially
rtain leaves which may be collected from the existing Quercus michauxii, prinus, or
he of eastern North America. Among fossil species it is very similar to some of the
a of Quercus westfalica figured by Hosius and von der Marck I from the Senonian of
jalia, and it is especially close to one of the Bohemian leaves which Velenovsky 2 refers
species. Heer 3 records this same species from the Atane beds of Greenland, but his
are too fragmentary for any great certainty of identification. Another leaf which
Similar is one from Kieslingswalde in Silesia, which Velenovsky identifies with his
Spseuddrymeja,4 although it is more probably referable to Quer u westfalica.5 No
Coastal Plain species is closely similar to Quercus pseudowestfalica, although Quercus
Berry 6 from the Magothy formation in New Jersey is remotely related to it, as is
raritanensis Berry 7 from the Raritan formation of that State.
form which is closely related to the Carolina species, however, is Quercus dakotensis
uax." It is of about the same size and outline, but differs in having less prominent
d in the details of its tertiary venation.
. Palmontographlea, vol. 26,1880, p. 161, Pl. XXIX, figs. 52-63; XXX, figs. 64-75.
2 Velenovaky, J., Die Flora der bShmischen Kreideformation, pt. 2,1883, p. 17, PI. II, fig. 23.
a Flora foesllis arctca, vol. 6, Abth. 2,1882, p. 67, Pl. XV, figs. 5-7.
4 Op. cit., figs. 21, 22.
* Op. cit., pt. 4, 188, p. 13, PIl. VII, fig. 10.
I Berry, E. W., Bull. Torrey Bot. Club, vol. 31, 1904, p. 74, Pl. IV, fig. 11.
S * Idem, vol. 36, 1909, p. 249.
V Loquernx, Leo, The flora of the Dakota group: Mon. U. 8. Geol. Survey, vol. 17,1892, p. 56, PI. Vn, fig. 4.





86 UPPEB CErITACEOUS AND BOOmmNE PRA OF SOUTH CAROUNA AND GEORGIA.

Quercus pseudowestfalica has been found in only the Middendorf member of the Black
Creek formation of South Carolina. It has also been doubtfully determined from the Black
Creek formation at Court House Bluff in North Carolina.
Occurrence.-Middendorf arkose member of Black Creek formation, near Middendorf,
Chesterfield County; Rocky Point, Sumter County, and Miles Mill, Aiken County. (Collected
by E. W. Berry, L. W. Stephenson, and Earle Sloan.)
CoUection.--U. S. National Museum.

Order ULTICALEBS.
Family LMAC.OA.
Genus MOMIBIA F. G. Dietrich.
MoMxxas cAnouwazaWS sp. Rov.
Plate XII, figure 5.
Descripiown.-Leves entire below, more or lem dentately toothed spically, ovate in outline,
6 or 7 centimeters in length by 2.5 oentimeters in greatest width. Apex narrowed and pointed.
Base cuneate, not quite equilateral, finally decurrent on the stout petiole, whih is short, 6 or 7
millimeters in legth. Midrib stout below, becoming thin distad. Basal secondaries, eensti-
tuting the lateral primaries, subopposite, branching from a point at or near the base of the
midrib at acute angles, long, nearly straight, ascending, camptodrome. Upper secondaries
remote, thin, camptodrome.
This species of a quasi triple-veined leaf seems allied to the existing species of Momisia, of
which no fossil forms are known except a single species from the Eocene of Georgia described
recently by the writer. Leaves of this character are commonly referred to the genus Cinna-
momum or at least to the Lauracese, but the botanical affinity of this form seems to be with
certain tropical Ulmacee. The existing species of Momisia number a score or more forms of
the American tropics. They are closely related to Celtis and are even made a subgenus of Celtis
by Engler. These leaves differ from the described species of Cinnamomum and its allies in
their toothed margins and in the character of their tertiary venation, and may be considered
the Cretaceous representative of the Claiborne species of Momisia.
Occurrence.-Middendorf arkose member of Black Creek formation, near Middendorf,
Chesterfield County. (Collected by L. W. Stephenson and E. W. Berry.)
Collections.-U. S. National Museum.

Family ARTOCARPACER.
Genus FICUS Linn6.
FiCUS ATAVINA Heer.
Plate X, figure 11.
Picus protogma Heer, Flora foesilis arctica, vol. 3, Abth. 2, 1874, p. 108, Ple. XXIX, fig. 2b, and XXX, figs. 1-8. (von
Ettinsghausen.)
Ficus protogmea Schenk, Palsontographica, vol. 23, 1875, p. 169, P1. XXIX, fig. 12.
Fieus atavina Heer, Flora fossilis arctica, vol. 6, Abth. 2, 1882, p. 69, PIS. XI, figs. 5b, 7b, 8b; XVII, fig. 8b; XIX,
fig. lb; and XX, figs. 1, 2.
Fieus atavina Heer, idem, vol. 7, 1883, p. 26.
Ficus atavinat Leequereux, Proc. U. S. Nat. Mus., vol. 10, 1887, p. 40.
Ficus protogra Hollick, Bull. Torrey Bot. Club, vol. 21, 1894, p. 51, Pl. CLXXV, fig. 4.
Fic atavina Berry, Bull. Torrey Bot. Club, vol. 31, 1904, p. 75, Pla. I, figs. 8, 9, and III, fig. 6.
Ficus atavina Hollick, The Cretaceous flora of southern New York and New England: Mon. U. S. Geol. Survey, vol.
50, 1907, p. 58, Pl. X, figs. 5, 6 (not fig. 4, which is referable to Juglas elongata).
Description.-Leaves entire, broadly oblong-lanceolate in outline, large in size, the South
Carolina leaves measuring about 20 centimeters in length by 3.5 centimeters in greatest width,
which is about halfway between the apex and the base. Apex and base equally acute. Midrib






I MS CEAAOUiOJB PFLO OF SOUT CAROUinA. 6W

. Texture coriaceous. Secondaries numerous, parallel, branching from the midrib
,of more than 450, camptodrome. Tertiaries rarely seen, forming small four or five
les.
large fig leaf stands about midway between Ficu krausiana and Ficu craaipes, and
y represent the somewhat narrower, more elongated leaves of the former species.
rbed originally from Greenland as Ficus protogra Heer, a preoccupied name, and
ently renamed by its describer. It has been recorded by Hollick from Marthas
d Long Island, from what are apparently Magothy deposits, and by the writer from
thy formation of New Jersey. Lesquereux recorded it from the far West (Utah), but
has not seen his material and queries the citation, for the species appears to be con-
t hemisphere, at least, to eastern North America. The range was possibly extended
, because Schenk identifies it in the Gosau beds of the northern Tyrol. It has not
been discovered in Alabama or in deposits of the southern Atlantic Coastal Plain.
.-Middendorf arkose member of Black Creek formation, Rocky Point, Sumter
SoMar Middendorf, Chesterfield County. (Collected by L. F. Ward, L. C. Glenn, L. W.
o, and E. W. Berry.)
*.-IU. S. National Museum.
SFicus CELTIFOLIUS sp. nov
Plate XII, figure 4.

s ption.-Leaves small, entire, elliptical-ovate in outline with an obtusely pointed apex
y rounded base. Petiole short, about 0.6 centimeter in length, relatively stout.
Secondaries stout, remote, three or four pairs, the lower pair opposite, diverting
midrib just below its base and functioning as lateral primaries, the balance alternate.
from the midrib at angles of about 450 and are camptodrome some distance from
in. Tertiaries distinct, the inner transverse, the peripheral camptodrome in uniform

Bispecies, which is obviously unlike any previously described Ficus, is the sole repre-
of the palmately veined figs found in South Carolina, although the somewhat similar
Ficus wooleoni Newberry and Ficus ovatifolia Berry occur in either the Black Creek
o of North Carolina or the Tuscaloosa formation of Alabama. It is especially charac-
by the prominence of its secondary and tertiary venation, and is very similar to certain
ga species of Celtis with entire margins, as, for example, Celti cramafoli Lamarck, and
reicuata Torrey. It is also similar to a number of Tertiary species of Ficus and not
a very much larger undescribed species of Ficus present in clays of the Tuscaloosa for-
of western Alabama.
nce.-Middendorf arkose member of Black Creek formation, near Middendorf,
d County. (Collected by L. W. Stephenson and E. W. Berry.)
oS.-U. S. National Museum.
Ficus CRASSIPES (Heer) Heer.
Plates X, figure 12, and XII, figures 8-10.

species is described in the section of this report dealing with the Upper Cretaceous
of Georgia, pages 110-111. It is common in South Carolina, especially at the Rocky Point
ty, from which the figured specimens were collected. It is especially characteristic of
~oet-Raritan and pro-Montana horizons of eastern North America.
Octurrence.-Middendorf arkose member of Black Creek formation, near Middendorf,
terfield County; near Langley, Aiken County; Rocky Point, Sumter County; right bank
legaree River about 25 miles below Columbia, Lexington County. (Collected by E. W.
f, L. F. Ward, L. C. Glenn, B. L. Miller, and M. W. Twitchell.)
Voiections.-U. S. National Museum.





88 UPPER CRETACEOUS AND EOCENE FLORAS OF' SOUTH" CAROLINA AND GEORGIA.

SFicus KRAUSIANA Heer.
Plate XI, figures 4-7.
ieus krausiana Heer, Neue Denkschr. Schw. Gasell., vol. 23, 1869, p. 15, PI. V, figs. 3-6.
FPius beckwithii Leequereux, The Cretaceous and Tertiary floras, 1883, p. 46, PIs. XVI, fig. 5, and XVII, figs. 3 and 4.
Ficus suapecta Velanovsky, Die Flora der bahmischen Kreideformation, pt. 4, 1885, p. 10, PI. V, figs. 6, 9.
Pietaatavina Heer? Hollick, Trans. New York Acad. Sci., vol. 2, 1892, p. 103, Pl. IV, figs. 4, 6.
Ficus krauaina, Lesquereux, The flora of the Dakota group: Mon. U. S. Geol. Survey, vol. 17, 1892, p. 81, Pl. I, fig. 5.
Picus beekwithii Leequereux, idem., vol. 17, 1892, p. 80.
Ficus krausiana Hollick, Bull. Geol. Soc. America, vol. 7, 1895, p. 13.
Fiurawasia Hollick, Annals New York Acad. Sci., vol 11, 1898, p. 59, Pl. III, fig. 1.
Picua krausana Bary, Bull. Torrey Bot. Club, 1906, vol. 33, p. 172.
Picus rauiana Hollick, The Cretaceous flora of southern New York and New England: Mon. U. S. Geol. Survey,
vol. 50, 1907, p. 58, Pls. IX, fig. 9, and X, figs. 1-3.
Description.-The following is Heer's description of this species, written in 1869:
F. foliis lanceolatis, integerimus, utrinque attenuatim, nervo medio valido, nervi secundis numerous, campto-
dromis, subtillimimi.
Leaves of large size, ovate lanceolate in outline, broadest at or below the middle. Apex
and base acutely pointed, the apex often extended and attenuated. Petiole and midrib stout.
Secondaries regular, open, thin, ascending, camptodrome, branching from the midrib at angles
of 45� or more. Texture coriaceous or subcoriaceous. Length about 17 centimeters. Greatest
width about 4 centimeters.
This species was described by Heer from Moletein in Moravia (Cenomanian), and it has been
subsequently identified from a large number of American localities. In the West it occurs
in the Dakota sandstone of Kansas and Colorado; in the East it is common from Marthas
Vineyard and Block Island to Alabama and is present between these limits in Maryland, North
Carolina, and South Carolina. These occurrences are all in beds of the Magothy or homotaxial
formations, the only uncertain occurrence being that at Tottenville, Staten Island, where the
morainic material containing this species may possibly have been derived from the Raritan,
although this is considered doubtful.
Associated with this species at the type locality in Moravia are similar leaves which were
described by Prof. Heer as a different species, Ficus mohliana. These are somewhat larger
with a more sparse secondary venation. It seems probable that these merely represent the
somewhat larger leaves of Ficus krausiana, but they are not united with it in view of the lack
of positive evidence, because Ficus mohliana has priority and this would involve the change of
name of this well-known type and horizon marker. In both North and South Carolina fruits
are found associated with this species, but whether they are related to it or to some of the
other rather numerous species of Ficus which occur at the same localities can not be determined.
Occurrence.-Middendorf arkose member of Black Creek formation, near Middendorf,
Chesterfield County (?); Rocky Point, Sumter County; right bank of Congaree River, about
25 miles below Columbia, Lexington County (?). Other deposits of the Black Creek formation,
Black Creek, Florence County. (Collected by L. F. Ward, L. C. Glenn, E. W. Berry, L. W.
Stephenson, and M. W. Twitchell.)
Collections.-U. S. National Museum.

Ficus STEPHENSONI Berry.
Plate XII, figures 1-3.
Ficuas tephensoni Berry, Bull. Torrey Bot. Club, vol. 37, 1910, p. 194, Pl. XXIII, figs. 2, 3.
Description.-Leaves variable in size, ranging from 6 to 18 centimeters in length and from
2.3 to 6.4 centimeters in greatest width, broadly lanceolate-ovate, tapering equally from the
middle toward both ends but more fully rounded at the base and more slender toward the tip,
especially in the smaller leaves. Midrib broad. Secondaries very slender, leaving the midrib






^"' ' PPME CR ETACEOUS FLORA OF SOUTH CAROLINA. 39
angles which become as great as 90� in some of the larger specimens, very numerous,
S lmeters apart, parallel, almost straight to the marginal vein, which is well marked and
i millimeter distant from the margin, with which it is parallel. Veinlets largely at right
to the secondaries and not especially well shown. Petiole stout.
s is an exceedingly well-marked species of Ficus and is very close to various existing
in form and venation characters, as is well shown by a comparison with the leaves of
lastica Roxburg. It is probable, however, that the texture of the fossil species was less
us, for all the larger leaves are considerably macerated.
It is believed that the larger forms represent the normal size of the leaves in this species and
the smaller leaves, which occur only in material from South Carolina, represent abortive
*which fell before reaching maturity, as is so common in the modern allied species.
me authors refer leaves of this type to the genus Eucalyptus, with which genus the vena-
has much in common. In point of size the Carolina leaves are comparable with those of
a species as Eucalyptus latifolia Hollick from Glen Cove, Long Island, but the secondaries
'lees regular and only about half as numerous in the latter species, and there seems to be
Doubt of the propriety of referring the present species to the genus Ficus.
iIt is very similar to a variety of closely related Upper Cretaceous species of Ficus of the
of the existing Ficus elastic Roxburg and its allies, commonly cultivated as ornamental
B bs and trees under the name of "rubber plants." The comparable fossil forms include
glascoeana Lesquereux,1 with which there is a possibility that the present species may be
tical, as it is very similar in outline and venation, except that the figures of the Kansas
yes (types 478 and 532a, Museum of Comparative Zoology) do not show any marginal vein,
ough Lesquereux mentions one in his description. The latter species has been detected south-
along the western shore of the Mississippi embayment in the Woodbinesand of Texas and is
a more coriaceous texture, with more obtuse tip, and with the secondaries joining the midrib
an angle of 60*. Another very similar species is Ficus atavina Heer,' which ranges from the
e and Patoot beds of western Greenland southward along the Atlantic Coastal Plain to
as Vineyard, Glen Cove, Long Island, and Cliffwood, N. J. All the occurrences are
bably of Magothy age.
The North Carolina leaf has full-rounded basal margins, which are rather straight in Ficus
tna; its secondaries, which are twice as numerous as in Ficus atavina, are less ascending.
marginal vein is also closer to the margin. Another species, perhaps identical with the
vious one, is Ficus peruni Velenovsky,3 from the Cretaceous of Bohemia, which differs from
North Carolina leaf in the same respects in which Ficus atavina Heer differs. Velenovsky
Iints out the great similarity between Ficus peruni and Eucalyptus geinitzii Heer, a similarity
ich is more striking in the forms he has referred to this species of Eucalyptus than in the
ves usually so identified by other paleobotanists.
Ficus stephensoni was recently described by the writer4 from material collected in the
k Creek formation at Court House Bluff on Cape Fear River in North Carolina. It is
und to be equally abundant in clays of the Middendorf member of the Black Creek formation
the State of South Carolina, but has not been certainly detected elsewhere in the Atlantic
astal Plain, though doubtful remains which may represent this species are present in the
elections from clays of the Tuscaloosa formation of western Alabama.
Occurrence.--Middendorf arkose member of Black Creek formation, Middendorf, Chester-
d County; near Langley, Aiken County. (Collected by L. W. Stephenson and E. W. Berry.)
- Collections.-U. S. National Museum.
' Lmquereux, Leo, The flora of the Dakota group: Mon. U. S. Geol. Survey, vol. 17,1892, p. 76, Pl. XIII, figs. 1, 2.
2 Flora foselis arotica, vol. 3, Abth. 2, 1874, p. 108, PIs. XXIX, fig. 2b, and XXX, figs. 1-8.
SDie Flora der bahmischen Kreideformation, pt. 3,1884, p. 16 (41), P1. IV (12), figs. 1-3. Compare his fig. 2 with Berry, Bull. Torrey Bot. Club,
S1,1904, PL. Il, fig. 6.
- LoC. cit.





40 UPPER CRETACEOUS AND EOCENE FLORAS OF SOUTH CAROLINA AND GBORGIA.
Order PBOT3ALEB.
Family PROTBACB2.
Genus PROTBOIDBmb8 eer.
PROTEOmiES LANCIFOLjUS Heer.
Plate X, figure 1.
Proteoides lancfolius Heer, Zur Kreideflora von Quedlinbrtg, 1871, p. 12, PI. III, figs. 5, 6.
Proteoids landifolius Lesquereux, The Cretaceous and Tertiary forau, 1808, p. 50.
Prwteoidua anfolUu Leequereux, The flors of the Dakota group: Mon. U. 8. Geol. Survey, vol. 17, 1892, p. 90, Ple.
XV, fig. 6, and I, fig. 8.
Description.--Leaves entire, lanceolate, commonly falcate in outline, with a narrowed
attenuated apex and a narrowed decurrent base, about 10 centimeters in length by 1.4 centi-
meters in greatest width, which is in the basal half of the leaf. Petible stout. Midrib stout
below, rapidly narrowing upward. Secondaries very numerous, approximately parallel, many
obsolete, branching from the midrib at acute angles of less than 45�. These are connected by
branches or tertiaries in the irregular manner characteristic of the venation of existing Proteacem.
Texture coriaceous.
This species was originally described from the Cretaceous of Blankenburg, Saxony, by IHeer.
Lesquereux recorded it from the Dakota sandstone and it is not uncommon in the South Carolina
Cretaceous, the latter remains being identical with the Dakota sandstone specimens and identical
with the type material except that in some of the leaves of the South Carolina specimens the
upper secondaries are depicted as open and camptodrome, subtending a wider angle with the
midrib, a character different from that shown in the lower secondaries of the same specimens.
All the observed characters are identical with those of various existing genera of the Proteacem,
as, for example, the genus Protea. The present species does not differ to any extent from the
Dakota sandstone species, Proteoides acuta Heer and Proteoides grevilliformis Ifeer, the latter
the type of the genus. The genus Proteoides does not contain many described species and is
confined to pre-Montana deposits in North America, although it ranges upward into the Senonian
of Europe.
Occurrence.-Middendorf arkose member of Black Creek formation, near Middendorf,
Chesterfield County; Rocky Point, Sumter County. Other deposits of the Black Creek forma-
tion, near Darlington, Darlington County. (Collected by E. W. Berry, L. C. Glenn, and
L. W. Stephenson.)
Collections.-U. S. National Museum.

PROTEOIDES PARVULA sp. nov.
Plate X, figure 5.
Description.-Leaves of small size, lanceolate in outline, entire, somewhat falcate, with
equally acute apex and base, 2 centimeters in length by 4.5 millimeters in greatest width,
which is about halfway between the apex and the base. Midrib thin, somewhat flexuous.
Secondaries fine, immersed, and mostly obsolete, numerous, ascending, irregularly anastomosing.
This species is one of the smallest which has been referred to this genus, and except for its
size is very similar to Proteoides lancifolius Heer, which occurs at the same locality-in fact, it
may simply be a juvenile form of the latter. This can only be determined by a larger amount
of comparative material than is available at the present time.
Occurrence.-Middendorf arkose member of Black Creek formation, near Middendorf, Ches-
terfield County. (Collected by L. W. Stephenson and E. W. Berry.)
Collections.-U. S. National Museum.






* W amBrABmous FLORA OF SOUTH CAROuNA. 41
Order BANALUS.
ai-l" y 3ANUNOrL&ACBA (P).
Qdbt D A3T*4UAA laporta and Marion.
D DwALQUA MrrTHI Berry.
Plate VIII, figures 3-9.
SToney, vol. 10, 1910, p. 36, fig. 1.
--Leaves palmately decompound, the petiole dividing into three principal
I gl of divergence varying from 20* to 600 and the two lateral branches forking
1 to 2 centimeters above their base. The middle leaflet is lanceolate in out-
in its central part and tapering almost equally to the acute apex and base.
'10 centimeters. Greatest width 2 to 4 centimeters. Margin entire or serrate,
-below and serrate in the apical three-fourths, some specimens having large



























FIGURE 1.-Restoration of Dewalqew smith Berry, from the Tuecalooes fonnrmation of western Alabama
rate teeth. Midrib stout. Secondaries regular, subopposite, parallel; about 20
ching from the midrib at angles varying from 450 to 700, generally about 50�, curving
and running to the marginal teeth or camptodrome. The base of the leaflet extends
within 2 or 3 millimeters of the forks of the petiole. Lateral leaflets more or less
Literal, usually somewhat smaller than the middle leaflet. The internal lateral leaflet is
llate, the outer lamina starting at or very near the point where the lateral branch of the
forks. The inner lamina, however, extends downward almost to the base of the lateral
i, making the base markedly inequilateral. In general outline and in marginal and
lion characters it is identical with the middle leaflet. The outer lateral leaflet is also
bt inequilateral, but less so than the internal lateral leaflet, the internal lamina starting
near the fork and its outer lamina extending more or less below the fork. Marginal and
ion characters as in the outer leaflets. This handsome species, of which a restoration is
s in figure 1, is abundantly represented at the locality near Langley, mostly by terminal





42 UPPER CRBTAOEOUS AND EOCENE FLORAS OF SOUTH COAOLINA AND GEORGIA.

leaflets, a number of which are reproduced in Plate VIII, figures 3-9. It is common in the
Tuscaloosa formation at Whites Bluff, on the right bank of Warrior River 309 miles above
Mobile, Ala., a small collection of fossil plants from this outcrop containing no less than 27
specimens of this form. Several of these were complete and were sketched at the time they
were collected, which proved fortunate, as the extremely arenaceous matrix did not withstand
shipment very well. The museum material, though considerably broken, shows several entire
detached leaflets and three or four basal parts of the leaf showing the mode of division of the
petiole.
The genus Dewalquea was founded by Saporta and Marion in 18741 upon remains from
the Senonian of Westphalia communicated by Debey and named by him in manuscript Aralio-
phyllum, and on additional remains collected by those authors from the Paleocene of Gelinden,
Belgium mariness heersiennes, 6tage Than6tien). Three species were enumerated, Dewalguea
haldemiana and Dewalque aqtuigraenais from the Westphalian Senonian and Dewlquea
gelindeneneis from the basal Eocene. In the past 35 years several additional species have been
referred to this genus, including another species from the German Senonian, Dewalquea irsignis,
described by Hosius and Von der Marck;2 two species from the Cenomanian of Bohemia,
Dewalque coriacea and Dewalquea pentaphylla, described by Velenovsky; * two American species
from the .Dakota sandstone, Dewalguea dakoteniia and Dewalquea primordialis, described by
Lesquereux,' both of which are fragmentary and of uncertain relationship; a species from the
Raritan formation of New Jersey, Dewaluea trifoliata, described by Newberry; 5 and a species
described by Heer 6 from Greenland, Dewalquea groenlandica, and subsequently recorded from
Staten Island, New Jersey, North Carolina, and Alabama.
Hosius and Von der Marck 7 record the Eocene species from the Senonian of Westphalia, but
the remains are not of this species, being fragments of Dewalguea mlademiana, which is common
at that horizon. The European species Dewalguea insignia is recorded by Heer 8 both from
the Atane and Patoot beds of Greenland and by Hollick 9 from the Cretaceous of Staten Island,
but both of these determinations are based upon fragments of single leaves and are, in the
writer's judgment, entirely untrustworthy. Attention should also be called to the possibility
of Celastrus arctica Heer10 representing the leaflets of a Dewalquea. This species was described
from the Patoot beds of Greenland, where it is sparsely represented. It is abundant, however,
in the upper part of the Raritan formation of New Jersey, but of some scores of specimens
examined by the writer all were detached and failed to show their habit of growth.
The botanic relationship of Dewalquea has always remained obscure and no better discussion
of it is extant than that given by Saporta and Marion," who, after comparing these leaves with
those of Ampelopsis, Arissema, Anthurium (Aracese), and other genera, arrive at the conclusion
that they are prototypes of the tribe Helleboreae of the Ranunculaces.
The present species is markedly distinct from the American species of Dewalquea previously
described, all of which were apparently tripartite. Among the European species it is quite
similar to the Senonian species Dewalquea insignia Hosius and von der Marck, which is, however,
entirely distinct. It is also similar to Dewalquea coriacea and Dewalquea pentaphylla described
by Velenovsky from the Cenomanian of Bohemia.
As previously mentioned, this species shows both entire and serrated forms. It is remark-
able that where this genus has been found in any abundance, two species are usually described,
I Saporta, 0. de, and Marion, A. F., Essal sur I diat de la vfgdtatlon & I'dpoque des maries heersiennes de Gelinden: Mm. court. et des Say.
dsang. Acad. roy. Belgique, vol. 37, p. 55. 1874.
2 Hoslus and Von der Marck, Paleontographica, vol. 26, 1880, p. 172, Pl. XXXII, figs. 111-113; XXXIII, fig. 100; XXXIV, fig. 110; and
xxxV, fig. 123.
a Die Flora der bdhmlschen Kreldeformation, pt. 3, 1884, pp. 11, 14, Pis. I, figs. 1-9; II, fig. 2; and VIII, figs. 11, 12.
SThe flora of the Dakota group: Mon. U. S. GeoL Survey, vol. 17, 1892, p. 211, PI. LIX, figs. 5, 6. Geol. and Nat. Hist. Survey Minnesota,
vol. 3, 1893, p. 18, PL A, fig. 10.
* Newberry, J. S., The flora of the Amboy clays: Mon. U. S. Geol. Survey, vol. 26, 1896, p. 129, Pl. XXII, figs. 4-7.
* Heer, Oswald, Flora fosllia arctica, vol. 6, Abth. 2, 1882, p. 87, PIs. XXIX, figs. 18, 19; XLII, figs. 5, 6; XLIV, fig. 11.
T Op. cit., p. 50.
* Op. cit., voL 6, Abth. 2, 1882, p. 86, Pla. XXV, fig. 7; XXXIII, figs. 14-16; idem, vol. 7, 1883, p. 37, Psl. LVIII, fig. 3; LXII, fig. 7.
* The Cretaceous flora of southern New York and New England: Mon. U. S. Geol. Survey, vol. 50, 1907, p. 106, PL VIII, fig. 24.
10 Op. cit., vol. 7, 1883, p. 40, PI. LXI, figs. 5d, 5e.
n Op. eit., pp. 56-61.





- - UPPER CRETACEOUS FLORA OF SOUTH CAROLINA. 48

one with toothed margins. Thus in Germany Dewalquea haldemiana is entire,
insignia is toothed, and probably both are the leaves of the same plant.
Dewaluea pentaphylla is entire and Dewa quea coriacea is toothed. As for the
t, it is believed that the entire and serrate leaves are specifically identical, for the
aws great many gradations in the size of the teeth and great variability regarding
n which the entire part of the margin bears to the toothed part on single leaflets.
.--Middendorf arkose member of Black Creek formation, near Langley and Miles
County. (Collected by L. W. Stephenson, E. W. Berry, and Earle Sloan.)
.-U. S. National Museum.

Family MAGNOLIACE H.
Genus MAGNOLIA Linn6.
MAGNOLIA CAPELLINII Heer (1).
Plate X, figure 3.
species, Which is confined to the Middendorf arkose member of the Black Creek forma-
South Carolina, is described in the section of this report dealing with the Upper Cretaceous
Georgia, pages 112-113.
rrence.-Middendorf arkose member of Black Creek formation, near Middendorf,
eld County; Rocky Point, Sumter County. (Collected by L C. Glenn, L. W. Stephen-
and E. W. Berry.)
olection.-U. S. National Museum.

MAGNOLIA NEWBERRYI Berry (1).
" longifolia Hollick, Trans. New York Acad. Sci., vol. 12, 1892, p. 36, P1. III, fig. 9.
j lonfifolia Smith, On the geology of the Coastal Plain of Alabama, 1894, p. 348.
a lonaifolia Hollick, Annals New York Acad. Sci., vol. 11, 1898, p. 422, PI. XXXVII, fig. 3.
" longfWolia Hollick, The Cretaceous flora of southern New York and New England: Mon. U. S. Geol. Survey,
vol. 50, 1907, p. 66, PI. XX, figs. 2, 3 (non Sweet, 1826).
Slongi(olia Newberry, The flora of the Amboy clays: Mon. U. S. Geol. Survey, vol. 26, 1896, p. 76, Pls. LV,
figs. 3 and 5, and LVI, figs. 1-4.
neerrayi Berry, Bull. Torrey Club, vol. 34, 1907, p. 195, PI. XIII, fig. 6.
Description.-Leaves mostly of large size, ovate to oblong in outline, about 20 centimeters
by 9 to 10 centimeters in width, broadest toward the base. Apex subacute or obtuse.
varying from obtusely rounded, almost truncate, to somewhat cuneate. Petiole and
bb stout. Secondaries comparatively thin and open, about 12 pairs, camptodrome.
eiars forming four, five, or six sided areoles, quite prominent in some specimens.
'This is the largest magnolia of the Raritan formation, the leaves of which are said by
berry to reach a length of 30 centimeters or more. It is common at the Woodbridge
lty and has also been reported from Staten Island and Marthas Vineyard, from the Tuscaloosa
tion in Alabama, and from the Black Creek formation in North Carolina. In a general
it resembles an immense leaf of Magnolia woodbridgensis, and it also approaches somewhat
longipes, but the petiole is only about one-third the length that it has in the latter
es. Its occurrence in the South Carolina Cretaceous is based upon doubtfully determined
from the locality near Darlington, but it must almost certainly have been a member
e South Carolina flora, for it has been detected in homotaxial deposits both north and south
area. (Collected by L. W. Stephenson.)
Occurrence.-Black Creek formation, near Darlington, Darlington County.
Collectiona.-U. S. National Museum.
i.,





44 UPPER COBTACEOUS AiD BOO NE FLORA OP SOUTH OARouNA AND GEORGIA.

MAeNOLIA OBTUSATA Heer.
aanlia ecapdeli Heer, Flora Fcesili Arctica, vol. 3, Abth. 2,1874, P1. XXIII, fig. 4 (non other citations of
species).
Magnuia obtsaus HeAr, idem, vol. 6, Abtk. 2, 1882, p. 90, Ph. XV, fig. 12, and XXI, fig. 3.
JisMoke6iaUo ts Lesquereux, The floor of the Dakota&goup: Mon. U. L. GeoL Survey, vol. 17, 1892, p. 201, Pl. LX,
Afg. 5,6.
Magnolia obtuhta Berry, Bull. New York Bot. Garden, vol. 3, 1908, p. 76, PL XLVII, fig. 4.
Magnolia obtusa5a Berry, Idem, vol. 37, 1910, p. 23.
Description.-Leaves of variable size, oblong-ovate or obovate in outline, entire, with a
broadly rounded apex and a narrowed cuneate base, ranging from 7 to 14 centimeters in length
by 2.4 to 7 centimeters in greatest width, which is above the middle. Petiole and midrib stout.
Secondaries few in number, ascending, curved, camptodrome. Texture coriaceous.
This species was described from the Atane beds of Greenland by Heer, and was based upon
rather fragmentary material. Subsequently Lesquereux recorded some fine specimens from the
Dakota sandstone of Kansas. It is present in the Magothy formation from New Jersey to
Maryland and is also a member of the Tuscaloosa flora in western Alabama. The South Carolina
specimens are few in number and come from but a single locality.
Ocurrene.-Middendorf arkose member of Black Creek formation, near Middendorf,
Chesterfield County. (Collected by L. W. Stephenson and E. W. Berry.)
Coetions.-U. S. National Museum.

MAGNOLr TENUrotiA Lesquereux (?).
Plate IX, figures 2 and 3.
Magnolia teuifolia Lesquereux, Am Jour. Sci., 2d ser., vol. 46, 1868, p. 100.
Magnolia tenuifolia Lesquereux, The Cretaceous flora, 1874, p. 92, PI. XXI, fig. 1.
Magnolia tenuifolia Lesquereux, The flora of the Dakota group: Mon. U. S. Geol. Survey, vol. 17, 1892, p. 198, PI.
XXIV, fig. 1.
Magnolia tenuifolia Berry, Bull. New York Bot. Garden, vol. 3, 1908, p. 77, Pl. XLVII, fig. 10.
Magnolia tenujfolia Hollick, idem, vol. 3, 1904, p. 413, Pl. LXXIII, fig. 2.
Magnolia tenuwfolia Berry, Bull. Torrey Bot. Club, vol. 31, 1904, p. 76, Pl. I, fig. 7.
Magnolia tenuifolia Berry, idem, vol. 33, 1906, p. 174, PIl. VII, fig. 1.
Magnolia enuufolia Hollick, The Cretaceous flora of southern New York and New England: Mon. U. S. Geol. Survey,
vol. 50, 1907, p. 64, Ple. XVII, fig. 1; XVIII, figs. 4, 5.
Description.-Leaves large, entire, oblong-ovate in outline, with a very stout petiole and
midrib. Length, about 20 centimeters. Greatest width, which is about halfway between
the apex and the base, about 8 centimeters. ApeW. bluntly pointed. Base cuneate, pointed.
Secondaries open, approximately parallel, inequidistaut, camptodromae.
The presence of this species in South Carolina Cretaceous is baaed upon the doubtfully
determined fragments figured. The leaf in life was such a large one that most of the occurrences
are based upon mere fragments. It was described originally from the Dakota sandstone and
has subsequently been recorded from the Magothy formation on Marthas Vineyard and Long
Island and in New Jersey and Delaware.
Occurrence.-Middendorf arkose member of Black Creek formation, near Middendorf,
Chesterfield County. (Collected by L. W. Stephenson and E. W. Berry.)
Colections.-U. S. National Museum.

Genus ILLICIrUM ULnd.
ILLICIUM WATEREENSIS ap. nov.
Plate XIV, figure 8.
Description.-Leaves entire, lanceolate in outline, with the apex and the base acutely and
equally pointed. Length about 9.5 centimeters. Greatest width 1.9 centimeters, in the
middle part of the leaf. Midrib narrow but prominent. Secondaries numerous, parallel,
branching from the midrib at angles of about 45�, camptodrome.







UPM ORETACrOUS FLORA OF SOUTH CAROLINA.


This speciw may be compared with a variety of described species in unrelated genera,
M, for example, Nyesa, Daphne, Apocynum, Andromeda, and various lauraceous genera, but
is believed to have more in common with the Ranalian genus Illicium, in which only-.ne
her retaeeous species, lUicium deetwum Velenovsky,' from the Cenomanian of Bohemia, is
own. The latter is very similar to the South Carolina leaf, differing merely in having4ewer
eondaries and more open secondary venation. A number of Tertiary species of Ilicium
known, and the modern forms, which are seven or eight in number, inhabit the warmer
of eastern North America and eastern Asia.
Occurrence.-Middendorf arkose member of Black Creek formation, Rocky Point, Sumter
unty. (Collected by E. W. Berry.)
Couections.-U. S. National Museum.
Order R08ALEB.
Family 1HAWAELMDAGIB.
Genus HAWAMBrUTRS Saporta.
HAMAMELITES? CORDATUS Lesquereux.
Plate X, figua 2.
ite to eodatu Leequereux, The Cretaceous and Tertiary floras, 1883, p. 71, PI. IV, fig. 3.
itesr cordatus Lesquereux, The flora of the Dakota group: Mon. U. S. Geel. Survey, vol. 17, 1892, p. 109.
Description.-Leaves large, elliptical in outline. Length about 10 centimeters. Greatest
th 6 to 7 centimeters; near the middle part of the leaf. Apex unknown. Base cordate.
undulate to dentate or serrate with shallow teeth. Midrib stout. Secondaries stout,
numerous, approximately parallel, branching from the midrib at angles of 45� or more, nearly
straight, at right angles to the secondaries. Texture coriaceous.
The genus Hamamelites was founded by Saporta in 1865 upon foreign early Tertiary
material, and to it Lesquereux referred five species from the Dakota sandstone, which exhibited
a combination of the characters of Hamamelis, Alnils, Viburnum, and other genera. Sub-
equently one of these was referred to Quercus. All were founded upon rather sparse material,
particularly Hamamelites? cordatus. Its occurrence in South Carolina is based upon the
igle imperfect specimen figured, which is very similar to the figured type from the West. It
ows part of the margin, the characteristic venation, and half of the typical cordate base.
t Hamamelis-like forms were present at this time is shown by the presence of wood of this
described by Lignier in 1907 as Hanramelidoxylon from the Cenomanian of France.
Occurrence.-Middendorf arkose member of Black Creek formation, Rocky Point, Sumter
Snty. (Collected by L. C. Glenn.)
Cobectione.-U. S. National Museum.
Family MIMOSACEI.
Genus ACACIAPHYLLITB8 gen. nov.
ACACIAPHYLLITE8 GREVILLEOIDES sp. nov.
Plate IX, figures 9 and 10.

Description.-Leaves or leaflets of small size, entire, oblong-elliptical in outline, about 2
ters in length by 0.5 to 0.65 centimeters in greatest width, which is at the middle part
leaf. Apex and base equally rounded. Petiole very short and relatively stout, about
, meters in length. Midrib thin, very much attenuated distad. Secondaries fine,
us, approximately parallel, branching from the midrib at acute angles; long, ascending,
ly camptodrome, connected at irregular intervals by cross branchlets of the same


I Velenovsky, J., Die Flora der bShmischen Kreideformation, pt. 3, 1884, p. 4, PL. IU, Ag. 5.







46 UPPER CEETACBOUS AND NOCEN FLORAS OF SOUTH CAROLINA AND GEORGIA.

This species is believed to show undoubted characters which ally it with the genus Acacia.
and is certainly referable to the families Mimosacese or Csealpiniacese, although suggesting the
family Proteacese. It is totally unlike any known fossil American species but resembles certain
of the Bohemian Cretaceous leaves referred by Velenovsky I to Grevillea or Grevilleophyllum,
although, as has been said, it is here considered leguminous.
These small leaves are not uncommon at the locality near Middendorf, to which they are
COfiied. Th&ir smUll ze has snhimftl^-np ta fmineaib'w-preserveh, ana 'ihough they
are thin they appear to have had a resistant epidermis of thick-walled cells, greatly resembling
a number of typical existing acacias.
Occurrence.-Middendorf arkose member of Black Creek formation, near Middendorf,
Chesterfield County. (Collected by L. W. Stephenson and E. W. Berry.)
Collectione.-U. S. National Museum.

Genus CSALPINIA Linn&.
CzsALPINIA MIDDENDORFENSIS sp. nov.
'late X, figure 7..
Description.-Leaves compound. Leaflets entire, elliptical in outline, with a broadly
rounded apex and a slightly broader, rounded, markedly inequilateral base, 1.5 centimeters in
length by 0.9 centimeter in greatest width, which is at the middle or below. Midrib slender
and curved in the single complete leaflet collected. Secondaries few and thin, four or five pairs,
branching from the midrib at a wide angle, more than 45*, camptodrome.
This is a well-marked species, clearly referable to some member of the Cmsalpiniaceme with
compound leaves, usage sanctioning the reference of leaflets of this sort to the genus Cmsalpinia,
with the leaflets of which they agree closely. Only two other Cretaceous species are known,
both from the Raritan formation in New Jersey. With the smaller of these, Cksalpinia cookiana
Hollick,2 the South Carolina form is closely comparable, but it differs in its more elongate inequi-
lateral outline and more numerous secondaries.
Occurrence.-Middendorf arkose member of Black Creek formation, near Middendorf,
Chesterfield County. (Collected by L. W. Stephenson and E. W. Berry.)
Collections.-U. S. National Museum.

Genus .LEGUMINOSITE8 Bowerbank.
LEGUMINOITrrES MIDDENDORFENSIs sp. nov.
Plate VIII, figarb 13.
Description.-Leaves or leaflets ovate-lanceolate in outline, with a pointed apex and
rounded base, 3 centimeters in length by 1.4 centimeters in greatest width, which is in the
middle part of the leaf. Midrib thin. Secondaries thin, open, five or six pairs, slightly curved,
ascending at angles of 45 to 50�.
This species is of rare occurrence and vague relationship, although it seems to be most
closely related to the leguminous leaves usually referred to the form-genus Leguminosites.
Occurrence.-Middendorf arkose member of Black Creek formation, near Middendorf,
Chesterfield County. (Collected by L. W. Stephenson and E. W. Berry.)
Collections.-U. S. National Museum.

LEGUMINOsrTES ROBINIIFOLIA Berry.
Plate IX, figure 11.
Leguminostes robiniifolia Berry, Bull. Torrey Bot. Club, vol. 37, 1910, p. 196.
Description.-Leaflets sessile, ovate to ovate-elliptical in outline. Length about 2.5
centimeters. Greatest width, 1.3 to 1.5 centimeters, at a point slightly nearer the base than
I Die Flora der b5hmischen Kreideformation, pt. 2, 1883, p. 3, PL I, flga. 6-10.
2 Newberry, J. S., The flora of the Amboyclays: Mon. U. S. Geol. Survey, vol. 26, 186, p. 94, .PL XLII, figs. 49, 50.








' UPPER CRBTACEOUS FLORA OF SOUTH CAROLINA. 4-

the apex and the base are obtusely rounded, the base broadly and the apex
* Midrib fairly stout. Secondaries consist of about five alternate to opposite
regularly curved, and approximately parallel, branching from the midrib at
45� or slightly more, camptodrome.
marked species is entirely distinct from previously described forms and is
to the leaves of the existing Robinia pseudacacia Linn6 of the eastern United
fact has led to the specific name chosen for it. It was described recently by the
material collected from the Black Creek formation at Court House Bluff on Cape
Sin North Carolina.
.-Middendorf arkose member of Black Creek formation, near Langley, Aiken
elected by L. W. Stephenson and E. W. Berry.)
.-U. S. National Museum.

Order GERANIALES.
Family BUTACEM.
Genus CITROPHYLLUM Berry.
CrITBOPHYLLUM ALIGERUM (Lesquereux) Berry.
4Leaquereux, Mon. U. S. Geol. Survey, vol. 17, 1892, p. 84, PI. X, figs. 3-6.
Baeny, Ann. Rept. State Geologist, New Jersey, for 1905, 1906, p. 139.
Berry, Bull. Torrey Bot. Club, vol. 33, 1906, p. 172.
aligerum Berry, idem, vol. 36, 1909, p. 258, P1. XVIIIa, figs. 1-8.
ption.-The following description is given in the reference last cited:
mB ail, elliptical to ovate or ovate-lanceolate, coriaceous, varying from 2.5 to 6 centimeters in length by
timetersin breadth. Margin entire, occasionally slightly undulate. Apex rounded or obtusely acuminate-
, subtruncate, or cuneate. Petiole stout, 0.7 to 2 centimeters in length, conspicuously alate. The
may be oblong-lanceolate or obovate; together they are from 2.5 to 5 millimeters in width, averaging
eters. Midrib stout. Secondaries fine, more or less obscured by the coriaceous leaf substance, about
rote pairs, branching from the midrib at angles of from 45 to 50*, parallel, camptodrome.
se curious leaves (morphologically leaflets) were described by Lesquereux in 1892 from
6ta sandstone as a species of Ficus and compared with Ficus bumelioides Ettingshausen
Smudge Lesquereux, neither which has alate petioles and the first has an emarginate
subsequently the same leaves were found in the Magothy formation of New Jersey,
7 recently a single small leaf was found in the upper part of the Raritan formation of
mboy, N. J. They exhibit considerable variety in outline, but all have exactly the
|et and conspicuous, more or less late petioles. They appear to be closely related to
B of the modern genus Citrus. The latter have exactly the same texture and venation,
variability in outline and marginal undulations, the same stout midrib, and conspicu-
ite petioles. In examining a suite of specimens of the latter and comparing them with
Sthe conclusion seems to be irresistible that they are related, and the writer consequently
the fossils to a new genus with a name that emphasized this relationship to the modern
the more complete leaves of this species were figured by the writer in 1909, as well
ent Citrus leaves for comparison with them. Possible arguments against the present
.be based on the theory that the modern alate petioles are derived from ancestors
pound leaves; in fact, some modern species still have trifoliate leaves, and if this were
fossils as well it would require considerable rapidity of evolution in the genus previous
dle part of the Cretaceous. The modern leaves abscise from the top of the petiole
be unlikely to occur as fossils with the petiole attached; neither can any indication
abscission line be made out in the fossils. This is the most difficult argument to
'However, modern leaves are sometimes shed in their entirety, and we are justified
the occasional fall of leaves before maturity when the abscission layer of cells had
me weakened. The cause might be violent winds, the passage of large animals





48 UPPER CRETACEOUS AND EOCENE FLORAS OF SOUTH CAROLINA AND GEORGIA.

like some of the Cretaceous dinosaurs, or weakened conditions resulting from attacks of insects,
or of fungous diseases. The species is also found in the Magothy, Tuscaloosa, and Dakota
formations.
The South Carolina occurrence of this species is based upon the incomplete but character-
istic leaves and tends to confirm the doubtful identification of this form in the Black Creek
formation of North Carolina.
Occurrence.-Middendorf arkose member of Black Creek formation, near Middendorf,
Chesterfield County; Miles Mill, Aiken County. (Collected by E. W. Berry and Earle Sloan.)
Collections.-U. S. National Museum.

Order RUPHORBIALUg.
Family lU KMORBIACZ.
Genus OCOTONOPTYLLUX Velenov ay.
CROTONOPHY.LUM PANDURarOrMIs sp. nov.
Plate VII, figures 5-10.
Description.-Leaves of variable size, entire and ovate-lanceolate or irregular panduriform
in outline, about 8 to 10 centimeters in length by 3 to 4 centimeters in greatest width, which is
in the basal half of the leaf. General outline ovate, separated by a sharp lateral sinus on each
side into a broad basal portion with full rounded margins and an upper narrower portion which
is more or less rounded or elongated. In some specimens the sinus is wanting on one side; it
may be wanting on both sides and the leaf be ovate-lanceolate in outline, as it often is in the
only other known species of the genus. This is the habit of a number of specimens from the
Miles Mill locality. Apex bluntly pointed. Base slightly recurrent to the stout petiole, which
is of considerable length. Texture boriaceous. Midrib stout. Secondaries numerous, rather
stout, branching from the midrib at angles of about 450, parallel, camptodrome below and in
some specimens also in the apical portion of the leaf, in which case they pursue an upwardly
curved course. In other specimens they are straight in the apical half of the leaf, and their
ends are connected by a nearly straight marginal vein, which is the continuation of some lower
secondary; in fact, the regularly camptodrome lower secondaries are parallel with the margin
before they finally inosculate.
These curious leaves are not uncommon in the South Carolina Cretaceous, though they
are generally incomplete. They are wholly unlike any known American Cretaceous leaves,
although they suggest the leaf described from the Upper Cretaceous of Vancouver Island by
Dawson as Liriodendron succedens.' In the writer's opinion Dawson's leaf is not a Liriodendron,
but as Dawson's figures are inaccurate, and as the writer has not seen the original material,
his opinion is not conclusive. The genus Crotonophyllum was proposed by Velenovsky for
leaves from the Cenomanian of Vyserovic, Bohemia, which are very similar to the present
species. A single species, CrotonophyUlum cretaceum,2 was described and compared with the
existing species of Croton, but as the discussion is in Bohemian the description is not readily
accessible to English readers. The illustrations, however, are ample and depict a leaf which
is surely congeneric with the South Carolina fossils.
Occurrence.-Middendorf arkose member of Black Creek formation, near Middendorf, Ches-
terfield County; near Langley and Miles Mill, Aiken County; Rocky Point, Sumter County.
(Collected by L. W. Stephenson, E. W. Berry, and Earle Sloan.)
Collections.-U. S. National Museum.


1 Trans. Roy. Soc. Canada, vol. 11, see. 4,1894, p. 62, P1, VIII, fig. 26.
s Velenovsky, J., Kvtena ftdskdho cenomanu, 1889, p. 20. P1. V, figs. 4-11.







UPPER CRETACEOUS FLORA OF SOUTH CAROLINA. 49

Order BAPINDALUS.
Family SAPINDACEB.
Genus *SAPINDUB Idnn6.

SAPINDUS MORHISONI Heer.
Plate IX, figure 6.
Smorrisoni Heer (Lesquereux MS.), Flora fossils arctica, 1882, vol. 6, Abth. 2, p. 96, Pls. XL, fig. 1; XLI,
g. 8; XLIII, figs. la, lb; and XLIV, figs. 7 and 8.
n morrisoni Heer, idem, vol. 7, 1883, p. 39, Pl. LXV, fig. 5.
Smorrisoni Lesquereux, The Cretaceous and Tertiary floras, 1883, p. 83, Pl. XVI, figs. 1 and 2.
Smorrisoni Leaquereux, The flora of the Dakota group: Mon. U. S. Geol. Survey, vol. 17, 1892, p. 158, PI.
XXXV, figs. 1 and 2.
Smorrisoni Hollick, Annals New York Acad. Sci., vol. 11, 1898, p. 422, Pl. XXXVI, fig. 4.
morrisoni Knowlton, Twenty-first Ann. Rept. U. S. Geol. Survey, 1901, p. 317, pt. 7.
morrisoni Berry, Bull. New York Bot. Garden, vol. 3, 1903, p. 83, Pl. XLVII, figs. 2 and 3.
Smortisoni Berry, Bull. Torrey Bot. Club, vol. 31, 1904, p. 78.
'temosoi Berry, Ann. Rept. State Geologist New Jersey for 1905, 1906, pp. 138 and 139.
uorrisoni Hollick, Mon. U. S. Geol. Survey, vol. 50, 1907, p. 90, PIl. XXXIII, figs. 16-20.
cription.-Leaves pinnately compound. Leaflets large, subcoriaceous, entire, lanceo-
Outline, about 10 to 15 centimeters in length or in a few specimens larger, by 2 to 3
er in greatest width. Base cuneate or rounded, generally inequilateral. Midrib stout.
stout, bowed, rather open, approximately parallel, camptodrome.
these leaflets are usually found detached, as in the few specimens found near Langley,
although some of the material from the west shows the habit. This species was described
lquereux and named in allusion to the type locality. As Lesquereux's report on Creta-
9 and Tertiary floras was not published promptly, it happened that Heer had meanwhile
ihed this species, based on Greenland material and identified by means of some of the
pof Lesquereux's work which the latter had sent to him, so that the species must be credited
er. It occurs at both the Atane and Patoot horizons in Greenland and is a common form
6 Woodbine sand of Texas, which is regarded as the southern equivalent of the Dakota
tone. In the east it is probably confined to post-Raritan deposits, although there is some
Pt regarding the age of the morainic material on Staten Island from which it has been
pded by Hollick. It is common in the Magothy formation of New Jersey, but has not
tofore been recorded from the southern Coastal Plain, although it will probably be found
St member of the Tuscaloosa flora when the Alabama collections are thoroughly studied.
t tcsrrence.-Middendorf arkose member of Black Creek formation, near Langley, Aiken
y. (Collected by L. W. Stephenson and E. W. Berry.)
ileetions.-U. S. National Museum.

Family CELASTRACES.
Genus PACHYSTIMA Rafinesque.

PACHYSTIMA? CRETAC(EA Sp. nov.
Plate X, figure 6.

cption.-Leaves of small size, oblong or obovate in outline, with a rounded apex and
ed descending base, about 2 centimeters in length by 3 millimeters in greatest width.
coriaceous. Petiole short and stout. Midrib stout. Secondaries numerous, fine,
ascending.
species is totally unlike any described fossil forms and closely resembles the leaves of
ting species of Pachystima, of which only two are known, one from the East and one
the Rocky Mountain area. They are shrubs of dry situations with evergreen, more or
809-14--4







50 UPPER CRETACEOUS. AND EOCENE PLORAS OF, OUTER CAROLINA AND GEORGIA.

less revolute leaves. The writer feels much doubt about this identification, and it is possible
that this species represents the excessively narrowed leaves of AcaciaphyUites grevileoides Berry
collected from the same locality. Pachystima is unknown in the fossil state elsewhere except
for a recently described species from the Miocene of Florissant, Colo.
Occurrence.-Middendorf arkose member of Black Creek formation, near Middendorf,
Chesterfield County. (Collected by L. W. Stephenson and E. W. Berry.)
Collectione.-U. S. National Museum.

Genus CELASTROPHYLLUM Goeppert.
CELASTROPHYLLUM ELEGANT Berry.

Plate XIV, figure 11.
Ceoastrophyllum eegans Berry, Bull. New York Bot. Garden, vol. 3, 1903, p. 84, PI. XLIII, fig. 6.
Celastrophyllum degans Berry, Bull. Torrey Bot. Club, vol. 32, 1905, p. 46, P1. II, fig. 1.
Description.-Leaves obovate in outline with a broadly rounded apex and cuneate pointed
base, 6 to 8 centimeters in length by 2 centimeters or slightly more in greatest width, which is
in the middle part of the leaf. Margin entire below with undulate shallow teeth above. Sec-
ondaries numerous, branching at angles of more than 450, rather straight, parallel, camptodrome.
This species is quite distinct from any other species of Celastrophyllum, although it shows
some points of similarity with the abundant Raritan and Dakota species. It was described
from the Magothy formation of New Jersey, the occurrence in South Carolina being the first
recorded outside of the type area. It is rare in South Carolina and not especially well preserved,
the material differing from the type in the somewhat more numerous secondaries.
Occurrence.-Middendorf arkose member of Black Creek formation, Rocky Point, Sumter
County. (Collected by E. W. Berry.)
Collections.-U. S. National Museum.

CELASTROPHYLLUM CRENATUM Heer.
Celastrophyllum crenatum Heer, Flora fomeilie arctic, vol. 7, 1883, p. 41, P1. LXII, fig. 21.
Celastrophyllum crenatum Smith, On the geology of the Coastal Plain of Alabama, 1894, p. 348.
Celastrophyllum crenatum Newberry, The flora of the Amboy clays: Mon. U. S. Geol. Survey, vol. 26, 1896, p. 99,
P1. XLVIII, figs. 1-19.
Celastrophyllum crenatum Berry, Bull. Torrey Bot. Club, vol. 34, 1907, p. 197, Pl. XIII, fig. 5.
Description.-Leaves very variable in size, 2 to 8 centimeters in length by 1 to 5 centimeters
in width, ovate or eliptical in outline, broadly rounded above, narrowed and generally inequi-
lateral below. Margins entire below,'coarsely toothed above, with somewhat variable, rounded,
crenate, or crenato-dentate teeth. A few specimens are entire throughout and some have a
markedly inequilateral base. Midrib fairly stout. Secondaries numerous, nine or ten pairs,
subopposite, branching from the midrib at angles somewhat in excess of 450, slightly curved
upward and parallel, branching near the margin to form festoons from which branches enter
the marginal teeth.
This species was described by Heer from the Patoot beds of Greenland and unfortunately
only a single small leaf was figured. The Raritan leaves, which are abundant, grade into
much larger forms that are also present in the Black Creek formation of North Carolina and
the Tuscaloosa formation of Alabama.
The species is rare in South Carolina, fragmentary specimens being sparsely represented in
the Middendorf collections. The genus is characteristic of the late Lower Cretaceous and early
Upper Cretaceous of eastern North America and is not known anywhere from beds of Montana
or Senonian age.
Occurrence.-Middendorf arkose member of Black Creek formation, near Middendorf, Ches-
terfield County. (Collected by L. W. Stephenson and E. W. Berry.)
Collections.-U. S. National Museum.







. *UPPER ORETAOEOUS FLORA OF SOUTH CAROLINA. 51

CELASTROPHYLLUM CAROLINENSIS Sp. nov.
Plate XIII, figures 1-5.
im.-Leaves lanceolate in outline, with a pointed apex and a curieate base, about
in length by 2.9 centimeters in greatest width, which is about midway between
and the base, tapering equally in both directions. Midrib stout, rather flexuous.
numerous, thin, branching from the midrib at acute angles of 450 or less, curving
usually camptodrome, a few craspedodrome in the upper part of the leaf, sending
ches into the marginal teeth. Margin entire for a short distance at the base,
ach it is crenate or biconvex, the teeth large and interspersed with smaller subordinate
e same character. Leaf substance thin.
striking form is rather common at the locality near Langley, but all the leaves are
*ken, though fragments of all parts of the leaf are present, and fully warrant the
Shown in figure 5 of the plate.
* species has been compared with a very large amount of existing material in the
of the New York Botanical Garden and the United States National Museum. It
ogies with a variety of existing genera, as for example, Cunonia, Clerodendron,
-, Ternstromia, Callicarpa, Panax, and other forms, but is believed to find its nearest
Among the Celastraceme. It is not close to any described fossil species, although there
resemblance to a number of the American Cretaceous species of Celastrophyllum.
also a general resemblance to Grevilleophyllum constant I and Aralia coriacea,2 both
ian species described by Velenovsky from Bohemia. Leaves of this sort have also
r to Dryandroides (cf. quercinea Velenovsky), Myrica (cf. serrata Velenovsky),
, and Fraxinus.
.ourrence.-Middendorf arkose member of Black Creek formation, near Langley, Aiken
. (Collected by L. W. Stephenson and E. W. Berry.)
.--U. S. National Museum.

Family ANACARDIACERB.
I Genus R IS Linn6.
RHUS DARLINGTONENSIS sp. nov.
Plate IX, figures 7 and 8.

'ption.-Leaflets large, broadly ovate in outline, with an obtusely pointed apex
rounded base, about 9 or 10 centimeters in length by 4.5 centimeters in greatest width,
is the middle or lower half of the leaf. Texture subcoriaceous. Margin dentate, crenate,
loped, with subordinate crenulations. Midrib straight, fairly stout. Secondaries con-
of about nine subopposite or alternate pairs, branching from the midrib at obtuse angles
ut 600, approximately parallel, slightly curved, craspedodrome, terminating in the main
W teeth and sending short outwardly and downwardly directed branches to the subordi-
eeth. Bulk of tertiaries transverse.
s large and evidently handsome species is based upon abundant, but very poorly proe-
material, the two best specimens being those figured. In general appearance it suggests
a elongate and broader leaves of Myrica elegant Berry, but it is a much larger and broader
Ath more open secondaries.
Ive Cretaceous species of Rhus have been described, three from the Dakota sandstone and
Pm Europe, one of the latter being also recorded from Glen Cove, Long Island, by Hollick.
ruth Carolina species is very different from all these and is much more modern in appear-
uggesting a number of Tertiary species of Rhus and closely resembling the larger, less
1 Velenovsky, J., Die Flora der bOhmlschen Kreldeformatlon, pt. 2,1883, p. 3, Pl. I, figs. 6-10.
SIdem, pt. 3, 1884, p. 11, Plh. I, figs. 1-9; II, fig. 2.







52 UPPER ORETAOEOYTS AND EOCENE FLORAS OF SOUTH GABOUNA AND GEORGIA.

elongate leaves of the existing Rhus gkab Linn6; in fact, leaves of the latter can be selected
which, except for the more acute teeth, are exactly like this Cretaceous form.
Occurrence.-Black Creek formation, near Darlington, Darlington County. (Collected by
L. W. Stephenson.)
CoUeotins.-U. S. National Museum.

Order TRYMxEAT.AS.
F&., mAU3LARALR.
afnny LALUKAS T46.

LAunus PLrroxW Heer.
Plates XI, figure 2, and XIII, figure 6.
Laurus plutonia Heer, Flora fossilis arctic, vol. 6, Abth. 2, 1882, p. 75, PIs. XIX, figs. Id, 2-4; XX, figs. Sa, 4-5;
XXIV, fig. 6b; XXVIII, figs. 10 and 11; and XLII, fig. 4b.
Lawus plutonia Heer, idem, vol. 7, 1883, p. 30, Pla. LVIII, fig. 2, and LXII, fig. la.
Laurus ptutonia Velenovsky, Die Flora der bohmischen Kreideformation, pt. 3, 1884, p. 1, PI. IV, figs. 2-4.
Law phtonis Lesquereux, The florsaof the Dakota group: Mon. U. S. Geol. Survey, vol. 17, 1892, p. 91, PIs. XIII,
figs. 5 and 6, and XXII, fig. 6.
Laurus plutonia Leequereux, Geol. and Nat. Hist. Survey Minnesota, vol. 3, pt. 1, 1895, p. 14, Ple. A, fig. 6, and B,
fig. 5.
Laurus plutonia Newberry, The flora of the Amboy clays: Mon. U. S. Geol. Survey, vol. 26, 1896, p. 85, Pl. XVI,
figs. 10 and 11.
Laurus pluonia Hollick, Annals New York Acad. Sci., vol. 11, 1898, p. 60, Pl. IV, figs. 6 and 7
Laurus plutonia ? Gould, Am. Jour. Set, 4th ser., vol. 5,1898, p. 175.
Laurus plutonia Berry, Bull. New York Bot. Garden, vol. 3, 1903, p. 79, P1. L, figs. 9-11
Laurus plutonia Berry, Bull. Torrey Bot. Club, vol. 31, 1904, p. 77, PI. III, fig. 1.
Laurus plutonia Berry, idem, vol. 33, 1906, p. 178.
Laurus plutonia Berry, Ann. Rept. State Geologist New Jersey for 1905, 1906, pp. 138 and 139.
Laurus plutonia Hollick, The Cretaceous flora of southern New York and New England: Mon. U. S. Geol. Survey,
vol. 50, 1907, p. 80, Pls. XXVII, figs. 9 and 11, and XXVIII, figs. 1 and 2.
Description.-Leaves lanceolate in outline, usually tapering almost equally in both directions
but some specimens less acute at the base. Length, 7 to 11 centimeters; greatest width, 1.5
to 2.5 centimeters. Midrib fairly stout. Petiole short and stout, 6 to 15 millimeters in length.
Secondaries slender, eight or more alternate pairs, camptodrome.
This species was described by Heer from the Atane beds of Greenland and a large number
of somewhat variable and fragmentary specimens were figured. Prof. Newberry subse-
quently recorded specimens from the Raritan formation of New Jersey without giving any
specific localities. Those figured show leaves which are relatively wider than is usual with
leaves of this species, but these are comparable with some of Heer's Greenland specimens?
Entirely typical leaves occur in the top layer of the Raritan at the Hylton pits in southwestern
New Jersey.
Subsequent to its description by Heer this species was recorded from a very large number
of Cretaceous plant beds, so that its present range, both geographic and geologic, is rather
wide. Some of these records the writer believes to be not altogether above question, however,
among those of which doubt is felt being the identifications of the forms from the Cenomanian
of Bohemia by Velenovsky.
It is evidently a rare plant in the Raritan but becomes abundant in immediately succeeding
floras, being common in that of the Dakota sandstone and in the Magothy formation at a number
of localities in New Jersey and Maryland. It is a common form in the insular Cretaceous floras
and also occurs in the South Atlantic Coastal Plain. Supposed fruits are figured by Heer.'
In South Carolina this species is represented by typical leaves that are not at all uncommon.
It has not been detected in the North Carolina Cretaceous, although it ranges from the base
to the top of the Tuscaloosa formation in Alabama.


'Op. cit., vol. 6, Abth. 2, 1882, PIs. XX, figs. 5 and 11.


'Op. cit., PI. XLH, fig. 4b.







UFPPB OBETACBOUS FLOAA OF SOUTH OAlOLWA.


--Middendorf arkose member of Black Creek formation, near Middenderf,
ty; near Langley, Aiken County. (Collected by L. W. Stephenson and E. W.

.-U. S. National Museum.
LAURUS ATANENSIS nom. nov.
Plate XIII, figure 7.
reen, Flora fossils arctic, vol. 6, Abth. 2, 1882, p. 76, Pla. XX, figs. lb and 7, and XLI fig. cl
eque).
ear, idem, vol. 7, 1883, p. 30, PIl. LVII, fig. lb.
leequereux, The flora of the Dakota group: Mon. U. S. Geol. Survey, vol. 17, 1892, p. 93, PL XVI,
lltick, Bull. New York Bot. Garden, vol. 3, 1904, p. 408, PIl. LXX, figs. 10 and 11.
Bollick, The Cretaceous flora of southern New York and New England: Mon. U. S. Geol. Survey,
'1W?, p. 81, Pl. XXVII, figs. 11 and 12.
*W .--Leaves entire, linear-lanceolate, tapering about equally to the acuminate
. Length, 12 centimeters or less; some of Heer's Greenland material ranging
under this dimension. Width, about 1.5 centimeters. Midrib relatively thin.
Sthin, somewhat widely separated, branching from the midrib at acute angles, 450
to twelve alternate pairs, finally ascending along the margin, camptodrome.
Forming the characteristic areolation of the genus.
originally compared this species with Laurus plutonia, pointing out that it was more
acuminate. He also remarks that it is somewhat smaller, but this generalization
und not to hold good, even for the Greenland material. The primaries are sparser
amendingg than in Laurnu plutonia and the more linear form gives the leaf a decidedly
aspect. There can be no doubt that the two are perfectly distinct species, naturally
g certain lauraceous characters in common.
as name for this species is preoccupied by that given by Rafinesque to an existing
o that a new name is necessary, the one proposed being given in allusion to the type

rw ataneneis was described from the Atane beds of Greenland and was subsequently
Sfrom the Patoot beds. Lesquereux identified it from the Dakota sandstone of Kansas
Bek has recorded it from the clays of Northport, Long Island, which are probably of
"age. It has not been detected from the southern Atlantic Coastal Plain, although it
I be present in the unstudied collections from the Magothy formation of Maryland and
STuscaloosa formation of Alabama.
rennce.-Middendorf arkose member of Black Creek formation, Rocky Point, Sumter
i (Collected by E. W. Berry.)
igons.-U. S. National Museum.
Genus LAUXOPHYLLUX Goeppert
LAUEOPHYLLUM ELEGANT Hollick.
Plate XII, figure 6.
Hollick, Trans. New York Acad. Sci., vol. 11, 1892, p. 99, Pl. III, figs. 3 and 4.
Hollick, idem, vol. 12, 1893, p. 236, Pl. VI, fig. 1 (non Heer).
geridae Hollick, Annals New York Acad. Sci., vol. 2, 1898, p. 420, Pl. XXXVI, fig. 2 (non Heer).
degatn Hollick, The Cretaceous flora of New York and New England: Mon. U. S. Geol. Survey, vol.
, p. 81, Pl. XXVII, figs. 1-5.
" egans Berry, Bull. Torrey Bot. Club, vol. 37, 1910, pp. 26, 198.
pion.-Leaves elongate-lanceolate, somewhat flexuous, about 12 or 13 centimeters
by about 2 centimeters in greatest width, which is about midway between the apex
; from this point they narrow gradually apically into an attenuated acuminate,
tip, and basally into a long, narrowly cuneate base. Midrib and petiole stout.
numerous, usually less close and somewhat coarser than in Larophlylum nervil-




L







54 UPPER CRETACEOUS AND EOCENE FLORAS OF SOUTH CAROLINA AND GEORGIA.

losum, branching from the midrib at an acute angle below, which becomes more open above the
base of the leaf; they .are usually more curved than in Laurophyllum neriUosum and more
distinctly camptodrome. Tertiaries transverse throughout.
These leaves were recorded originally by Hollick as Laurus plutonia Heer and were later
compared with Laurus angusta Heer, which latter species they resemble more than they do the
former. In outline they are not unlike LaurophyUum angustifolium Newberry, from the Raritan
formation of Woodbridge, N. J., but differ decidedly in venation. They are also similar but
quite distinct from LaurophyUum serviUosum Hollick, of the East, and LaurophyUum reticulatum
Lesquereux, of the Dakota sandstone.
The type was obtained from transported materials associated with the terminal moraine,
from which numerous specimens have been collected. Those from Tottenville, Staten Island,
are undoubtedly of Raritan age, but those from Glen Cove may have come originally from the
Magothy formation, although they are probably Raritan. The species is certainly known from
the upper part of the Raritan at South Amboy, N. J., and is common in the Magothy formation
of Maryland. It is sparsely represented in the Black Creek formation of North Carolina and is
not uncommon near Middendorf, the latter specimens being slightly broader and consequently
having fewer secondaries ascending than the type specimens.
Occurrence.-Middendorf arkose member of Black Creek formation, near Middendorf,
Chesterfield County. (Collected by L. W. Stephenson and E. W. Berry.)
Collections.-U. S. National Museum.

LAUROPHYLLUM NERVILLOSUM Hollick.
Plate XII, figure 7.
Proteoides daphnogenoide Hollick, Annals New York Acad. Sci., vol. 2, 1898, p. 420, Pl. XXXVI, figs. 1 and 3 (non
Heer).
Laurophyllum nervillosum Hollick, The Cretaceous flora of southern New York and New England: Mon. U. S. Geol.
Survey, vol. 50, 1907, p. 82, Pl. XXVII, figs. 6, 7.
Lawurophyllum nervillosum Berry, Bull. Torrey Bot. Club, vol. 36, 1909, p. 255.
Description.-Leaves of comparatively large size, oblong-lanceolate in outline, about 15
centimeters in length by about 2.5 centimeters in greatest width, which is about midway between
the apex and the base. Apex acuminate. Base pointed, narrowly cuneate. Midrib stout.
Secondaries thin, close, parallel, branching from the midrib at angles not exceeding and usually
somewhat less than 450, ascending, nearly straight or somewhat flexuous, connected by trans-
verse nervilles, branching and inosculating near the margin, where they merge in the tertiary
venation.
This species was described originally from the terminal moraine at Tottenville, Staten
Island, and undoubtedly represents transported Raritan materials. It is also present in the
lower part of the Raritan formation at Milltown, N. J., and is not uncommon in the South
Carolina Cretaceous. It is somewhat like Laurophyllum lanceolatum Newberry, but has a
markedly different venation and a less lanceolate outline. It is also quite close to Laurophyllum
elegans Hollick, which is, however, a more slender lanceolate leaf, having narrowly produced
apex and base and a somewhat coarser venation, with camptodrome secondaries, less close and
more curved.
Occurrence.-Middendorf arkose member of Black Creek formation, near Middendorf, Ches-
terfield County. Black Creek formation, near Darlington, Darlington County. (Collected by
L. W. Stephenson and E. W. Berry.)
CoUections.-U. S. National Museum.
Genus CINNAXOMTX Blume.
CINNAMOMUM NEWBERRYI nom. nov.
Plate IX, figures 12 and 13.
This species is described in the section of this report dealing with the Upper Cretaceous
flora of Georgia (pp. 117-118). The South Carolina remains referred to it are few and poor,







.... - 'UPPER CRETACEOUS FLORA OF' SOUTH CAROLINA. 55

fragments of rather large leaves from Rocky Point and doubtfully determined
from the banks of Congaree River.
e .-Middendorf arkose member of Black Creek formation, Rocky Point, Sumter
ht bank of Congaree River, about 25 miles below Columbia, Lexington County.
by E. W. Berry, B. L. Miller, and M. W. Twitchell.)
.-U. S. National Museum.

CINNAMOMUM MIDDENDORFEN8IS sp. nOV.
Plates VIII, figure 14, and IX, figure 1.
piion.-Leaves lanceolate in outline, somewhat widened toward the acute base.
to 10 centimeters; greatest width, 1.7 to 2.4 centimeters in the basal half of the leaf,
narrowing gradually upward to produce the very much extended, acutely pointed
ares three, from the top of the petiole, the midrib being slightly the more promi-
les of divergence acute, 200 to 250. The lateral primaries assume a course approxi-
arallel to the midrib about half way between the latter and the margins and extend
the tip of the leaf, finally joining a short, upwardly curving secondary from the mid-
riaries numerous, transverse, nearly straight, and parallel. From the extreme base
side a marginal vein extends upward from one-third to one-half the distance to the
ng the ends of the transverse tertiaries which extend outward from the lateral
s, above which upwardly curved camptodrome secondaries are given off. Texture

is an exceedingly well-marked and handsome species, entirely distinct from any
us forms hitherto described. It is a typical Cinnamomum in all its characteristics,
it may be compared with a variety of tropical genera such as Leucosyke and Zizyphus.
Some resemblance to the Bohemian Cenomamian forms described by Velenovsky I as
cinnamomeus and is strikingly like certain existing Oriental species of Cinnamomum,
Sample, Cinnamomum chinense Blume or Cinnamomum albiflorum Rees. It is also close
fs Tertiary species referred to this genus.
rrwnce.-Middendorf arkose member of Black Creek formation, near Middendorf,
d County. (Collected by L. W. Stephenson and E. W. Berry.)
on.--U. S. National Museum.
Order MYRTALES.
Family MYRTACES.
Genus EUCALYPTUS L'H6ritier.
EUCALYPTUS ANGUSTA Velenovsky.
Plate XIV, figure 2.
angusta Velenovsky, Die Flora der b6hmischen Kreideformation, pt. 4, 1885, p. 3, PI. III, figs. 2-12.
angusta Velenovsky, Kv6tena &eskdho cenomanu, 1889, p. 21, Pl. VI, fig. 1.
aungwta Saporta, Flore foesile du Portugal, 1894, p. 207, Pl. XXXVI, fig. 12.
jaguuta Berry, Bull. Torrey Bot. Club, vol. 36, 1909, p. 260, Pl. XVIII, fig. 5.
piptio.-The following is Velenovcky's description of this interesting species:
Lineal, schmal lineallanzettlich, in der Mitte oder in der unteren Halfte am breitesten, ganzandig, vorne
jag Spitze vorgezogen und mit einem harten Dorm beendet. Der Primfrnerv grade, ziemlich stark,
Sverdalnnt. Die Secundlirnervenahlrich, unter spitzen Winkeln entspringend, am Rande durch einen
riander verbunden. Der Blattateil gerade, etwa 1 cm. lang, stark.
species is exceedingly common at a number of localities in the Perucer-schichten of
(Cenomanian), where Velenovsky subsequently found fruit-bearing twigs which he
and figured in 1889 and which, it would seem, conclusively establish the botanic
p these leaves.
k-" V.ineoky, J., Die Flora der b5hmlschen Kreideformaflton, pt. 4,1885, p. 4 (65), PI. VIII (31), figs. 16-21.







66 UPPER CRETACEOUB AND EOCENE FLOBAS OF SOUTH CAROLINA AND GEORGIA.

Subsequently Saporta recorded this species from the Albian beds of Portugal; the
latter material is, however, rather incomplete and open to question. Recent collections in our
Atlantic Coastal Plain show that this species was present in considerable abundance on this
side of the Atlantic at the same time that it flourished in Europe. It has been collected from
the upper part of the Raritan formation at South Amboy, N. J., where it is common, and from
the Black Creek formation of South Carolina, where it is associated with Araucaria badetenais
Berry just as it is in Georgia. It may be somewhat more fully characterized as follows: Leaves
alternate or scattered, mostly elongated, linear-lanceolate, many falcate, 4.5 to 15 centimeters
in length by 5 to 13.5 millimeters in width, with an attenuated acute tip and a narrowly cuneate
base declining to the short and stout petiole. Midrib moderately stout below, becoming
attenuated above. Secondaries very numerous, fine, and close, about 1 millimeter apart,
parallel, rather straight; they branch from the midrib at acute angles of about 30� or slightly
less and run with slight curvature to join the well-marked but fine marginal hem, which
shows in all the American material and in most of the illustrations of the foreign material.
In all respects this is one of the most characteristically Eucalyptus-like species of the
many which have been identified as such; and its totality of characters, combined with the
presence of attached fruits in the Bohemian material, which are not unlike some of those of
modern forms, renders the identification very satisfactory.
Occurrence.-Black Creek formation, right bank of Black Creek, below Williamson's bridge,
Florence County. (Collected by L. W. Stephenson.)
Cooections.-U. S. National Museum.

EucALYPTUS GEINITZI (Heer) Heer.
Plates XIII, figures 8-12, and XIV, figure 1.
Myrtophyllum geinit Heer, Flora von Moletein, 1872, p. 22, Pl. XI, fig. 3, 4.
Myrtophyllum geinitzi Heer, Flora fossilis arctica, vol. 3, Abth. 2, 1874, p. 116, Pl. XXXII, figs. 14-17.
Eucalyptus geinitzi Heer, idem, vol. 6, Abth. 2, 1882, p. 93, PIs. XIX, fig. ic, and IV, fig. 1, 13.
Eucalyptus geinitzi Lesquereux, The flora of the Dakota group: Mon. U. S. Geol. Survey, vol. 17, 1892, p. 138, PI.
XXXVII, fig. 20.
Myrtophyllum warden Leequereux, idem, 1892, p. 136, Pl. LIII, p. 53, fig. 10.
Eucalyptus? angustifolia Newberry, The flora of the Amboy clays: Mon. U. S. Geol. Survey, vol. 26, 1896, p. 111,
PI. XXXII, figs. 1, 6, and 7 (non Deevaux 1822).
Eucalyptus geinitzi Newberry, idem, 1896, p. 110, Pl. XXXII, figs. 2 and 12 (non figs. 15, 16X.
Eucalyptus geinitzi Hollick, Annals New York Acad. Sci., vol. 11, 1898, p. 60, PI. IV, figs. 1-3.
Eucalyptus geinitzi Berry, Bull. New York Bot. Garden, vol. 3, 1903, p. 87, Pl. LIII, fig. 3.
Eucalyptus? angustifolia Hollick, idem, vol. 3, 1904, p. 408, Pl. LXX, figs. 8 and 9.
Eucalyptus geinitzi Berry, Bull. Torrey Bot. Club, vol. 31, 1904, p. 78, pl. IV, fig. 5.
Eucalyptus geinitzi Berry, idem, vol. 33, 1906, p. 180.
Eucalyptus geinitzi Berry, idem, vol. 34, 1907, p. 201, Pl. XV, fig. 4.
Eucalyptus geinitzi Berry, Johns Hopkins Univ. Circ., new Ber., No. 7, 1907, p. 81.
Eucalyptus geinitzi Hollick, The Cretaceous flora of southern New York and New England: Mon.,U. S. Geol. Survey,
vol. 50, 1907, p. 96, Pl. XXXV, figs. 1-8, 10-12.
Myrtophyllum warderi Hollick, idem, p. 97, Pl. XXXV, fig. 13.
Eucalyptus angustifolia Hollick, idem, 1907, p. 95, Pl. XXXV, figs. 9, 14, and 15.
Description.-Leaves lanceolate in outline, broadest near the middle and almost equally
tapering in both directions to the acute apex and base. There is considerable variation in size,
the South Carolina leaves averaging about 15 centimeters in length by 2.2 centimeters in
greatest width. The petiole is very stout, as is the prominent midrib, which leaves a sharp
groove in impressions of the lower surface of the leaf. Secondaries numerous, thin, branching
from the midrib at acute angles, about 450, and running with only a slight curvature to the
marginal vein, which is either almost straight when the secondaries are close set, or more or
less bowed when the secondaries are some little distance apart, as they are in many specimens.
This species has an especially wide range. It was described originally from the Cenomanian
of Moravia, and has since been recorded from a number of other European localities, from
the Atane beds of Greenland and the Dakota sandstone of the West, and from Marthas Vineyard







, UPPER CRETACEOUS FLORA OF SOUTH CAROLINA. 57

along the Atlantic Coast. The South Carolina material is abundant and charac-
is certainly identical with the type, whatever may be thought of some of the leaves
been identified with this species.
.-Middendorf arkose member of Black Creek formation, near Middendorf,
County; near Langley, Aiken County; Rocky Point, Sumter County. (Collected
.tphenson and E. W. Berry.)
bo.-U. S. National Museum.
EucALYrPTus WARDIANA Berry (I).
Plate XIV, figures 3 and 4.
T dubia Berry, Bull. New York Bot. Garden, vol. 3, 1903, p. 87, P1. LII, fig. 1 (non Ettingshausen).
bodmiow Berry, Bull. Torrey Bot. Club, vol. 32, 1905, p. 47.
weadina Berry, idem, vol. 33, 1906, p. 180.
avdiana Berry, Ann. Rept. State Geologist New Jersey for 1906, 1906, pp. 138, 139, and 141.
o.--Leaves entire, linear-lanceolate in outline, with an acute apex and cuneate
10 or 12 centimeters in length by 1 to 1.6 centimeters in greatest width. Petiole
Sconsiderable length. Midrib stout. Secondaries numerous, approximately straight
branching from the midrib at angles of 450 to 500, their ends connected by a
yein, which is straight and close to the margin, with which it is parallel. This species
y described 'by the writer from New Jersey, and is found to a characteristic species
thy formation in that State as well as in Delaware and Maryland. It is very similar
Rpr aritan species, Eucalyptugs linemarifolia Berry (Eucalyptus neiroa Newberry),
the forms from the Tuscaloosa formation of Alabama correlated by Ward with
species are referable to this species. The South Carolina remains agree with the
from New Jersey better than is indicated by the figures, for the type figures are
ate.
secies may also be cpmpared with the contemporaneous Eucalyptus angusta Velenov-
is, however, a smaller, falcate, less linear leaf with still more numerous and more
secondaries.
.-Middendorf arkose member of Black Creek formation, near Middendorf,
Cd County. (Collected by L. W. Stephenson and E. W. Berry.)
.-U. S. National Museum.
Order UNMTl.LAT.RS.
Family ARALIACERM.
Genus HEDERA Linn&.
HEDERA PRIMORDIALIS Saporta.
Saporta, Le monde des plates, 1879, p. 200, figs. 29, 1 and 2.
SVelenovsky, Die Flora der bShmischen Kreideformation, pt. 1, 1882, p. 19, Pls. VIII, fig. 7; IX,
d 5; and X, figs. 3 and 4.
'is Newberry, The flora of the Amboy clays: Mon. U. S. Geol. Survey,vol. 26, 1896, p. 113, Pls. XIX,
md 9 and XXXVII, figs. 1-7.
Berry, Bull. Torrey Bot. Club, 1907, vol. 34, p. 201, P1. XVI.

.-Leaves elliptical, reniform or cordate in outline, very variable in size and
3 to 12 centimeters, width 3.2 to 12 centimeters, generally broader than long.
ed or obtusely pointed, in some specimens slightly emarginate. Margin somewhat
Entire. Base varies from truncate to deeply cordate. Petiole long and stout,
t preserved. Venation palmate from the top of the petiole. Primaries range in
Three to seven, usually five or seven, of which the midrib is the stoutest, especially
er leaves. The lowest pair of primaries, which are approximately parallel with
of the leaf, are smaller in size than the others and should be regarded as
The primaries are then normally five in number, curved and camptodrome.







58 UPPER CRETACEOUS AND EOCENE FLORAS OF SOUTH CAROLINA AND GEORGIA.

This species was figured by Saporta in 1879 from the Cenomanian of Bohemia and described
three years later by Velenovsky from the same horizon. Heer identified rather fragmentary
remains from the Atane beds of Greenland with this species, which is also abundant in the
Raritan formation of Woodbridge, N. J., and in the Black Creek formation of North Caro-
lina. It varies greatly in size and appearance, some of the smaller specimens from abroad
suggesting the genus Cercis, whereas the smaller Raritan leaves suggest somewhat the genus
Ficus. Of these variable specimens the writer is disposed to consider as typical those shown in
Velenovsky's Plate X, figure 4, and Saporta's figure 2, as well as various Woodbridge specimens,
which are, however, mostly incomplete.
This is a remarkably widespread species, and is better characterized where it does occur
than is usual in cosmopolitan types. Although the modem representation of this genus is
reduced to two species in Europe and northern Africa and a third in Japan, it seems to have
been a more or less prominent type in the Cretaceous and Tertiary floras of the globe. In addi-
tion to the present species, which has the wide range previously mentioned, 8 or 10 additional
Cretaceous species, mostly American, are known. The Eocene, both of America and Europe,
furnishes six or eight species, the Oligocene of Europe and the Arctic regions one or two species,
and the Miocene and Pliocene two or three additional. The modern Old World Hedera helix
Linn6 is recorded from the Pleistocene (Interglacial) of England, Italy, and the Paris basin,
and one of the upper Miocene species appears also to have survived into the Italian Pleistocene.
Although so abundant in our Cretaceous floras, it is not a native plant in the existing flora of
North America.
Occurrence.-Black Creek formation, Peedee River, about 6 miles below Cheraw, Chester-
field County; near Darlington, Darlington County. (Collected by E. W. Berry and L. W.
Stephenson.)
Collectione.-U. S. National Museum.

Order ERICALTS.
Family ERICACE2B.
Genus ANDROMEDA Linn6.
ANDROMEDA NOVE OSARES Hollick.
Plate XIV, figures 5 and 6.
Andromeda novecalares Smith, On the geology of the Coastal Plain of Alabama, 1894, p. 348.
Andromeda novwcsares Hollick, in Newberry, The flora of the Amboy clays: Mon. U. S. Geol. Survey, vol. 26, 1896,
p. 121, Pl. XLII, figs. 9-12, 28-31.
Andromeda novncesares Berry, Bull. Torrey Bot. Club, vol. 33, 1906, p. 181.
Andromeda novmcesares Berry, idem, vol. 34, 1907, p. 204.
Description.-Leaves small, thick, and entire, with stout petioles and midribs and obscure
secondary venation, which is immersed in the thick lamina. Length 2.5 to 5 centimeters.
Width varies from 0.9 to 1.3 centimeters. Venation where visible shows numerous parallel,
camptodrome, relatively long and thin secondaries which branch from the midrib at an acute
angle. Though the majority of these leaves are equally acuminate at both ends, there is
a great deal of variation in this respect, and a considerable number of specimens which are
relatively broader, especially in the upper half, exhibit a well-marked tendency toward an
obtusely rounded apex, in which the termination of the midrib shows as a small mucronate
point. The base in these forms gradually narrows to the stout petiole. The variations in
outline of this species are well shown in the figures reproduced in Newberry's monograph, the
specimens from the southern Coastal Plain seeming to have more commonly than those from
New Jersey an obtusely rounded apex.
In the Raritan formation this species is only known with certainty from the uppermost
beds at South Amboy, N. J. It becomes more abundant in the overlying Magothy formation,
occurring from New Jersey to Maryland in beds of this age. Farther south it is found as one







U UPPER CRETACEOUS FLORA OF SOUTH CAR OLINA. 69

ical fossils of the Black Creek formation in North Carolina, being a prominent but
dant element in the dark lignitic laminated clays of the upper beds, associated with
a, Cunninghamites, Pistia, and other genera, and with a marine fauna.
) also occurs in clays of the Middendorf member of the Black Creek formation of South
ha and in the Tuscaloosa formation of Alabama, and was one of the few leaves which
*fuly resisted maceration in the shallow shoreward deposits known as the Cusseta sand
of the Ripley formation of Georgia, occurring at Buena Vista in association with
*ia bladenensis just as it does along Black River in North Carolina.
currence.--Middendorf arkose member of Black Creek formation, near Middendorf,
rfeld County. (Collected by L. W. Stephenson and E. W. Berry.)
eon.-U. S. National Museum.
ANDROMEDA GRANDIFOLIA Berry.
Plate XIV, figure 10.
I latfolia Smith, On the geology of the Coastal Plain in Alabama, 1894, p. 348.
aatifolia Newberry, The flora of the Amboy clays: Mon. U. S. Geol. Survey, vol. 26, 1896, p. 120, Pls.
XXIII, figs. 6-8, 10 (non fig. 9); XXXIV, figs. 6-11; and XXXVI, fig. 10 (non Wright).
s lcattfolia Hollick, Bull. New York Bot. Garden, vol. 3, 1904, p. 416, Pl. LXXIX, fig. 3.
fa latifolia Hollick, The Cretaceous flora of southern New York and New England: Mon. U. S. Geol. Survey,
5. 50, 1907, p. 100, Pl. XXXIX, fig. 1.
f fgrandifolia Berry, Bull. Torrey Bot. Club, vol. 34, 1907, p. 204, Pl. XV, fig. 3.
eacription.-Leaves thick and coriaceous, varying considerably in size and shape. From
0 centimeters in length by 1.5 to 7 centimeters in width. Ovate-lanceolate in outline,
a entire, usually somewhat undulate or unsymmetrical margin. Apex obtusely pointed
me specimens rounded. Base somewhat wedge-shaped. Midrib and petiole very stout.
rines relatively few, six to eight pairs, stout and flexuous, branching from the midrib
acute angle and sweeping upward in long curves, eventually inosculating to complete
otly camptodrome venation.
i species occurs from the lower part of the Raritan formation of New Jersey to the top
eastern leaf-bearing Cretaceous. It is a not uncommon fossil in the Magothy formation,
Gak Creek formation of North Carolina, and the Tuscaloosa formation of Alabama. It
r, relatively broader, and less regular than Andromeda parlatorii Heer, the two leaves
[ on Plate XIV showing the average shape, with a length of about 10 centimeters and a
of about 5 centimeters.
c .rence.-Middendorf arkose member of the Black Creek formation, Rocky Point,
r County. (Collected by L. F. Ward and L. C. Glenn.)
bections.-U. S. National Museum.

ANDROMEDA EUPHORBIOPHYLLOIDES sp. nov.
Plate XIV, figure 7.
sription.-Leaves small, entire, obovate-lanceolate, with a broadly rounded apex and
wed cuneate base, about 5.3 centimeters in length by 1.25 centimeters in greatest width,
in the apical half of the leaf. Petiole short and stout. Midrib stout. Secondaries
ms, approximately parallel, thin, branching-from the midrib at acute angles, long,
og, camptodrome. Texture coriaceous.
species in its size, outline, and venation is referable to the genus Andromeda, greatly
| some of the obovate leaves of the contemporaneous Andromeda novrcesares Hollick,
in its more elongate form and straighter, more produced base, It also resembles some
ves of the Dakota sandstone referred by Lesquereux1 to his species Eugenia primsw
re oblanceolate, with less full margins and straighter secondaries. It also resembles,
in its general form, certain curious leaves described by Saporta from the Cenomanian
L The flora of the Dakota group: Mon. U. S. GeoL Survey, Vol. 17, 1899, p. 137, PL LII, fig. 7.







60 UPPER CRETACEOUS AND BOCENE FLORA& OF SOUTH CAROLINA AND GEORGIA.

of Portugal and from the Turonian of France,3 which are made the basis of a new genus,
Euphorbiophyllum, and referred to the family Euphorbiacem.
Occurrence.-Middendorf arkose member of Black Creek formation, near Middendorf,
Chesterfield County. (Collected by L. W. Stephenson and E. W. Berry.)
CoUecione.-U. S. National Museum.

ANDROMEDA PARLATORiU Heer.
Andromeda parlatorii Heer, Phyllites cr6tac6es du Nebraska, 1860, p. 18, Pl. I, fig. 5.
Prunus ? parlatorii Leequereux, Am. Jour. Sci., 2d ser., vol. 46, 1868, p. 102.
Leucothoe parkatorii Schimper, Pal6ontologie vWgtale, vol. 3, 1874, p. 11.
Andromeda parlatorii Heer, Flora foesilis arctic, vol. 3, Abth. 2, 1874, p. 112, PI. XXXII, figs. 1 and 2.
Andromeda parlatorii Lesquereux, The Cretaceous flora, 1874, p. 88, P1. XXIII, figs. 6 and 7, and XXVIII, fig. 15.
Andromeda parlatorii Heer, idem, vol. 6, Abth. 2, 1882, p.79, Ple. XXI, figs. lb and 11, and XLII, fig. 4c.
Andromeda parlstorii Lesquereux, The flora of the Dakota group: Mon. U. S. Geol. Survey, vol. 17, 1892, p. 115, Pls.
XIX, fig. 1, and LII, fig. 6.
Andromeda parlatorii Smith, On the geology of the Coastal Plain of Alabama, 1894, p. 348.
Andromeda parlatorii Newberry, The flora of the Amboy clays: Mon. U. S. Geol. Survey, vol. 26, 1896, p. 120, PIt.
XXXI, figs. 1-7, and XXXIII, figs. 1, 2, 4, and 5.
Andromeda parlatorii Hollick, Annals New York Aced. Sci., vol. 11, 1896, p. 420, PI. XXXVII, figs. 1-4.
Andromeda parlatorii Berry, Bull. New York Bot. Garden, vol. 8, 1903, p. 97, Pl. L, figs. 1-4.
Andromhesp erpatorii Berry, Bull. Torrey Bot. Club, vol. 31, 1904, p. 79, PI. I, figs. 1 and 2.
Andromeda parlatorii Berry, idem, vol. 33, 1906, p. 181.
Andromeda parlatorii Hollick, The Cretaceous flora of southern New York and New England: Mon. U. S. Geol. Survey,
vol. 50, 1907, p. 101, PI. XXXIX, figs. 2-5.
Andromeda parltaorii Berry, Bull. Torrey Bot. Club, vol. 34, 1907, p. 203, PI. XV, fig. 2.
Andromeda parlatorii Berry, Johns Hopkins Univ. Circ., new ser., No. 7, 1907, p. 81.
Description.-Leaves ovate-lanceolate in outline, with a long and gradually narrowed apex
and a broad, somewhat rounded, but finally cuneate or slightly decurrent base. Petiole and
midrib stout. Length about 10 to 12 centimeters. Width about 3 centimeters in the lower
half of the leaf. Secondaries numerous, rather thin, subparallel, branching from the midrib
at acute angles, long and ascending, at length camptodrome. Tertiaries mostly straight
transverse. There is considerable variation in the size of these leaves and in the angle which
the secondaries form with the midrib and consequently in their length and degree of curvature.
Some of the specimens are much like the small leaves of Andromeda g9rndifoia Berry, but are
not so slender nor so attenuated apically as the normal leaves of that species.
This species was first described by Heer in one of the earliest published accounts of the
flora of the Dakota sandstone, and it has since been found to have a wide geographic range.
It is one of the commonest fossils in the Dakota sandstone, having been recorded from Minnesota,
Kansas, and Nebraska. In eastern North America it is recorded from the Atane beds of
Greenland, the Raritan formation of New Jersey, the Magothy formation of Marthas Vineyard,
New Jersey, Delaware, and Maryland, the Black Creek formation of North Carolina, and the
Tuscaloosa formation of western Alabama.
In South Carolina, this species has not been found in any abundance, a fact explained
entirely by accidents of preservation.
The genus Andromeda of Linn6 has been much segregated by subsequent taxonomists,
and this tendency to segregation is reflected in Schimper's proposal to refer this species to the
genus Leucothoe. However, the more comprehensive name has obvious advantages for the
paleobotanist where it is impossible to discriminate between the various ericaceous genera
with any degree of accuracy.
Occurrence.-Middendorf arkose member of Black Creek formation, near Langley, Aiken
County; Rocky Point, Sumter County (1). (Collected by E. W. Berry.)
CoUections.-U. S. National Museum.
I Saporta, G. do, Florefosile du Portugal, 1894, p. 218, PI. XXXIX, fig. 23.
SSaporta, 0. f., solution des Pbandogames, vol. 2, 18M, p. 117, fig. 125 C.







SUPPER CRO TACBOUS FLORA OF SOUTH CAROLINA. 61

Order PA39MUiLAIS.
Family MYRSINACESB.
Genus MYtRSIN Linn6.
MYRSINE GAUDINI (Lesquereux) Berry.
Plate XIV, figure 9.
SLeequereux, The flora of the Dakota group: Mon. U. S. Geol. Survey, vol. 17, 1892, p. 115, PI.

Hollick, Bull. Torrey Bot. Club, vol. 21, 1894, p. 54, Pl. CLXXVII, fig. 2.
Newberry, The flora of the Amboy clays: Mon. U. S. Geol. Survey, vol. 26, 1896, p. 122, Pl. XXII,

Bollick, Annals New York Acad. Sci., vol. 11, 1898, p. 420, Pl. XXXVIII, figs. 3, 4b, and 4c.
Hollick, The Cretaceous flora of southern New York and New England: Mon. U. S. Geol. Survey,
, 1907, p. 102, Pls. VIII, fig. lb, and XXXIX, figs. 13 and 14.
iBerry, Bull. Torrey Bot. Club, vol. 36, 1909, p. 262.
.n.-Leaves oblanceolate or elongate-obovate in outline, 5.5 to 7 centimeters in
1.9 to 2.5 centimeters in greatest width. Margins entire. Apex obtusely rounded.
Mswhnt elongated, narrowly cuneate. Petiole present, stout. Midrib stout below,
ia~ninishing in caliber. Secondaries numerous, eight to ten pairs, alternate, branching
HAidrib at angles of 40� to 450, camptodrome. When the tertiary venation is distinctly
1, the venation is more typical of the genus than when only the secondaries are partly

kopecies is well distributed in the Raritan formation and has been recorded also from
wd and Staten Island. The identification of Myrsinitest gaudini Lesquereux with
forms, with which it is obviously identical, extends the range eastward from Kansas
'Island. It may be readily distinguished from the other species of Myrsine by its
narrow, elongated form. It is present in the Black Creek formation of North Caro-
4the Tuscaloosa formation of Alabama. It is not abundant in the South Carolina Creta-
sing present only in Aiken County. The figured specimen is typical and extremely
bhe type, being a trifle more elongated, as are also the leaves of this species from the
A Atlantic Coastal Plain.
hrmnce.-Middendorf arkose member of Black Creek formation, near Langley, Aiken
(Collected by L. W. Stephenson and E. W. Berry.)
betione.-U. S. National Museum.

Order EBENALES.

Family EBENACE.B.
Genus DIOSPYROS IAnn6.
DIOSPYROS PRIMAEVA Heer.
Plates XI, figure 3, and XIV, figures 12 and 13.
Heer, Phyllites cretac6es du Nebraska, 1866, p. 19, P1. I, figs. 6 and 7.
Beer, Flora foseilis arctica, vol. 6, Abth. 2,1882, p. 80, PI. XVIII, fig. 11.
Heer, idem, vol. 7, 1883, p. 31, Pl. LXI, figs. 5a, 5b, and 5c.
Leaquereux, The flora of the Dakota group: Mon. U. S. Geol. Survey, vol. 17, 1892, p. 109, Pl.
Ass. 1-3.
Smith, On the geology of the Coastal Plain of Alabama, 1894, p. 348.
Newberry, The flora of the Amboy clays: Mon. U. S. Geol. Survey, vol. 26, 1896, p. 124, PI. XXX,

am Knowlton, Twenty-first Ann. Rept. U. S. Geol. Survey, pt. 7, 1901, p. 317, Pl. XXXIX, fig. S.
Berry, Bull. Torrey Bot. Club, vol. 32, 1905, p. 46, PI. II.
a Berry, idem, vol. 34, 1907, p. 204.
Isnava Hollick, The Cretaceous flora of southern New York and New England: Mon. U. S. Geol. Survey
50, 1907, p. 103, Pl. XL, figs. 2 and 11.






62 UPPER CRETACEOUS AND EOCENE FLORAS OF BOUTH CAROLINA AND GEORGIA.

Description.--Leaves oblong-ovate in outline, variable according to age, ranging from 3 to
15 centimeters in length by 1.3 to 5 centimeters in greatest width, which is in the middle part
of the leaf. Apex acute or obtuse. Base cuneate. Margins entire. Petioles rather long and
very stout. Midrib also stout. Secondaries branching from the midrib usually at acute angles,
subopposite or alternate, parallel, camptodrome. Tertiaries forming polygonal areoles, whose
relative prominence is one of the features of this species.
This species, which is quite suggestive of the modern Diospyros virginiana Linn6, was
described by Heer from the Dakota sandstone of Nebraska nearly half a century ago. It has
proved to be a most wide-ranging form, having been identified at both the Atane and Patoot
horizons in Greenland; from various localities within the Dakota sandstone, and its probable
southern equivalent the Woodbine sand of Texas; and, besides the fragments, from Marthas
Vineyard and Long Island, which are of questionable identity, it is common in either the Rari-
tan or Magothy or homotaxial formations from New Jersey to Alabama. It has been recorded
from the Cenomanian of Saxony and the Turonian of Bohemia.
Its most marked character is the prominence of its tertiary areolation. This species is
common at all leaf-bearing horizons in the South Carolina Cretaceous deposits, and the material
gathered near Langley is notable for the large number of small leaves associated with those of
normal size. These conform to the type in all respects except that they are more slender and
acuminate, a feature to be expected in the small leaves of Diospyros, and it is believed that
they are not distinct from the type. Two of these small leaves are figured, the normal-sized
leaves having been amply illustrated by Newberry.
Occurrence.-Middendorf arkose member of Black Creek formation, near Middendorf, Ches-
terfield County; near Langley, Aiken County; Rocky Point, Sumter County; right bank of
Congaree River, about 25 miles below Columbia, Lexington County (1). (Collected by L. W.
Stephenson, E. W. Berry, Earle Sloan, B. L. Miller, and M. W. Twitchell.)
Collections.-U. S. National Museum.

DIospyos ROTUNDrIOLIA Lesquereux.
Plate XIV, figure 14.
Diospyros rotundifolia Lesquereux, The Cretaceous flora, 1874, p. 89, P1. XXX, fig. 1.
Diospyrio rotundifolia Lesquereux, The flora of the Dakota group: Mon. U. S. Geol. Survey, vol. 17, 1892, p. 112, PI.
XVII, figs. 8-11.
Diospyros rotundfolia Berry, Ann. Rept. State Geologist, New Jersey, for 1905, 1906, p. 139.
Diospyros rotundifolia Berry, Bull. Torrey Bot. Club, vol. 33, 1906, p. 181.
Description.-Leaves entire, of various sizes, 4 to 10 centimeters in length by 2 to 7 centi-
meters in greatest width, which is in the middle part of the leaf. Outline broadly oval or
elliptical. Apex broadly rounded. Base similar or somewhat narrowed and pointed. Petiole
and midrib stout. Secondaries, six or seven pairs, branching from the midrib at angles of 50�
to 600, arched, camptodrome. Texture subcoriaceous. Venation less prominent than in
Diospyros primeva Heer.
This species is a characteristic element in the post-Raritan flora of the Atlantic Coastal
Plain, although at times liable to be confused with Myrsine borealis Heer or with some of the
smaller, more orbicular, entire leaves of Populus. The venation is markedly different, however.
It was originally described from the Dakota sandstone of Kansas and is common in the
Magothy formation in New Jersey, Delaware, and Maryland. It has also been recorded from
the Tuscaloosa formation in western Alabama. In South Carolina it has been found only at
the Rocky Point locality. The material, though complete, has the venation mostly opposite,
and it is possible that it may represent some orbicular leguminous leaflet instead of this species,
with which, however, it agrees admirably.
Occurrence.-Middendorf arkose member of Black Creek formation, Rocky Point, Sumter
County. (Collected by E. W. Berry.)
Collections.-U. S. National Museum.







UPPER CRETACEOUS FLORA OF SOUTH CAROINA.


DICOTTLEDONm INCERTe SEDIS.
Genus CALYOITRS Auct.
CALYCorTE MIDDENDORFENSIS sp. nOV.
Plate X, figure 4.
f o.--Calyx-like organism, with a small central disk or receptacle from which
linear, apically rounded, equidistant sepals, measuring 1.1 centimeters in diameter
tip. Sepals slightly narrowed proximal, 1 millimeter or a fraction more in width.
of this sort are commonly referred to the form genus Calycites, as the botanic
but few specimens can be determined. A considerable number of species of Calycites
described, coming from the Raritan, Magothy, and Dakota formations. It may be
whether forms like those described from the Magothy of Marthas Vineyard and
d by Hollick should not receive some other names, for they hardly represent calices,
iously fruits comparable with certain modern Dipterocarpacese or Hippocrateacees.
t species is based on the single complete specimen figured.
ence.-Middendorf arkose member of Black Creek formation, near Middendorf,
d County. (Collected by E. W. Berry.)
.-U. S. National Museum.
-r BOTANIC CHARACTER OF THE FLORA.
* Cretaceous flora of South Carolina as made known in the present contribution consists
ci, distributed among 49 genera in 36 families and 26 orders. There are represented
erof the phylum Thallophyta, 2 of the Pteridophyta, and 73 Spermatophyta, includ-
ospermwe (1 Cycadales and 13 Coniferales) and 59 Angiosperm. (5 Monocotyledon.
Dicotyledons). The largest orders are the Urticales, Ranales, Thymeleales, and Sapin-
ach of which has 6 species.
largest single genus is Ficus, which is represented by 5 species, and this genus is also
5 abundant individually. The genera Salix, Magnolia, and Andromeda have 4 species
ucaria, Celastrophyllum, and Eucalyptus have 3 each; and the following genera are
nted by 2 species each: Myrica, Quercus, Proteoides, Leguminosites, Laurus, Lauro-
, Cinnamomum, and Diospyros.
plants are very unequally represented at the 11 localities enumerated, the locality
iddendorf furnishing 42 species, whereas the locality on Black Creek below Williamsons
(3.9) has furnished no positively identified forms. The bulk of the described species
ome from the Middendorf arkose member of the Black Creek formation, identifiable
a being uncommon in other deposits of the Black Creek formation, although lignites,
etary leaves, and comminuted vegetable matter are universally distributed throughout
pical Black Creek deposits and are relatively rare in the Middendorf member. This
*on is due to a certain extent to the distribution of vegetation on the near-by land and
largely to what may be termed the accidents of preservation and is paralleled by the
ution of the plants in the Black Creek formation of North Carolina, where most of the
come from the single favorably situated locality at Courthouse Bluff on Cape Fear River.
more distinctly marine sedimentation, more remote from the shore or from Cretaceous
the plant remains were much macerated and triturated before entombment. For
ason a better idea of the flora as a whole and of the accompanying physical conditions
obtained from the Middendorf flora, of which the general botanic bearing is equally
able to the typical Black Creek flora, although the latter indicates somewhat different
ip grouping and considerably different conditions of deposition.
r a brief statement of the character of the South Carolina Cretaceous flora as a whole,
rt will be made to picture the environmental factors-ecologic, topographic, climatic, and
o---which may legitimately be deduced from the plant assemblages found fossil in the
ndorf member and the typical Black Creek deposits.







64 UPPER CRETACEOUS AND EOCENE FLORAS OF SOUTH CAROLINA AND GEORGIA.
The.flora as a whole furnishes a single thallophyte, which is represented by poorly preserved
remains of a dichotomously branched thallus of undetermined botanic affinity. It is confined
to the Black Creek formation in South Carolina, but occurs rather more abundantly at the
same horizon in North Carolina and also in the Magothy flora of Maryland; it indicates marine
sedimentation.
The Pteridophyta are represented by one species each of Onoclea and Lycopodium, and
strangely enough both are fruit-bearing specimens, the Lycopodium closely resembling the exist-
ing members of the genus and heretofore not being certainly known from the Mesozoic. The
botanic affinity of the Onoclea is not conclusively determinable, although the remains represent
a type found also in the Magothy formation and in the Upper Cretaceous of Greenland.
The Cycadales are represented by a single species of Podozamites, which is confined to
the Middendorf member in South Carolina, but which occurs in the Black Creek formation of
North Carolina, northward to New Jersey, in the Dakota sandstone of the West, and in Europe.
The remains are detached leaflets of the type usually referred to Podozamites, although the
botanic affinity of many of these leaflets is not fully established.
The Coniferales are well represented. The Taxaces are recognized by the Cephalotaxus-
like fruits which are so common in the Black Creek formation of North Carolina and which occur
also in the upper part of the Tuscaloosa formation of Alabama.. The other member of this
family is referable to the curious fernlike genus Protophyllocladus, a characteristic Upper
Cretaceous type widely distributed in North America. The species is new but occurs also in the
Magothy formation of Maryland.
The Araucariacese, abundant in the Mesozoic though antipodean in existing flora, have
three species of typical araucarias-one based on foliage, another on cone scales, and the third
on seeds-but it is quite probable that all may belong to a single botanic species. The other
family in this order, the Brachyphyllacem, is represented by the characteristic and widely
distributed BrachyphyUum macrocarpum Newberry, which ranges from the Raritan to the
Montana and which is the last survivor of a common older Mesozoic type of plant.
The Pinacem are represented by a species of Pinus of modern aspect and by the charac-
teristic remains of the wide-ranging and widely distributed Sequoia reicdenbachi (Geinitz)
Heer, a species whose Raritan representative Hollick and Jeffrey propose to refer to the Arau-
cariaces, although it may be noted that their argument is inconclusive. Furthermore, it seems
probable that all of the forms referred to this species may not be identical. The modern oriental
genus Cunninghamia is doubtfully represented by a species of Cunninghamites, which is quite
widely distributed and which ranges from the Cenomanian to the Senonian abroad and probably
as widely in this country. The genus Moriconia is abundantly represented in the Middendorf
member, its most southerly known occurrence. From this region it ranges northward, a closely
allied species occurring both in Greenland and in Europe.
The Cupresseaces are certainly represented by Widdringtonites ubtilis Heer, a form which
ranges from the Atane beds of Greenland southward along the Atlantic coast to Alabama.
Rather characteristic, apparently four-valved cones of the Widdringtonia (Callitris) type are
associated with the foliage in clays of the Middendorf member. In the Tuscaloosa formation of
western Alabama this species is very common and has furnished a number of specimens with the
cones attached to the leafy twigs of this type, so that the botanic affinity of this species
seems to be established beyond dispute. In addition to the foregoing forms of more or less
certain botanic relationship a cone is described from the Black Creek formation and a cone
scale from the Middendorf member, both of unknown affinity.
The Monocotyledonse have furnished five species, a Potamogeton, an Arundo, a Phragmites,
and a Carex, waterside types whose occurrence as fossils are easily explained, as well as frag-
mentary remains of a large palmetto-like fan palm, Sabalites, oneof the earliestknown occurrences
of a plant of this type. Palms appear simultaneously in the early part of the Upper Cretaceous
in both Europe and America, and before its close they appear to have become numerous and
diversified as well as widely distributed.
Of the Dicotyledone, the amentiferous families are represented by nine species. The
Juglandales have a species of Juglans which ranges from Greenland to Georgia.







UPPER CRETACEOUS FLORA OF SOUTH CAROLINA. 65
yricales have two characteristic species of Myrica, a genus of considerable importance
taceous floras everywhere and in more recent floras as well. The willows have
one peculiar to the Middendorf member and three widely distributed in beds of
y the same age. The oaks have furnished two species, both clearly defined. It
that the oaks appear in such abundance both in this country and Europe soon after
'of the Upper Cretaceous. They afford one of the marks of post-Raritan floras in
tic Coastal Plain.
rticales form one of the most abundant orders in the South Carolina Cretaceous-in
numbers of individual specimens easily the most abundant. A single doubtful species
j to the modern warm-temperate genus Momisia, which, except for a species recently
i by the writer from the Eocene of Georgia, is unknown in the fossil state. The figs
yve species, including four species with lanceolate leaves and pinnate venation and one
rith shorter and broader leaves and palmate venation. One of the lanceolate figs,
phenoni, is new, although it occurs in North Carolina; the others are all well known
p forms of wide distribution. They are exceedingly abundant in South Carolina; both
ipes Heer and Ficus krausiana Heer occur at four localities at least and are very
at certain of these, particularly in the clay ironstones at Rocky Point. One of the
rasts to be noted in tracing this Upper Cretaceous flora southward from Greenland is
ency shown by a number of forms toward the development of prolonged attenuated
is prominently shown in these species of Ficus, both here and in the Tuscaloosa
Alabama, and it has been interpreted as indicating a heavier rainfall and more humid
than in higher latitudes. This peculiarity is not confined to the genus Ficus, but is
a number of other genera belonging to these floras.
)he Proteales, the Proteaces, at present largely and almost exclusively developed in the
*;Hemisphere, are represented by two species of Proteoides, so called from their close
with the.modern species of Protea. Many botanists, notably in England, have ques-
ch identifications, especially as made by Ettingshausen and others in. studies of the
floras of Europe. Most of these identifications have a large element of certainty,
and are paralleled and confirmed by the similar Cretaceous range and Modern restriction
e number of unrelated genera. The present place is unsuitable for controversial matter,
'iter will say that rather extensive distributional studies have served in a large measure
an the presence in the Cretaceous floras of North America of species of Proteacese,
M, and other types.
\order Ranales, which at the present time has received such undue prominence through
netic speculations of Wieland and Arber, is represented by six species-a Dewalquea
able and striking appearance, common to the Middendorf member and to the Tuscaloosa
Sof Alabama; a new species of Illicium; and four well-known species of Magnolia,
ichrange northward as far as Greenland and three southward into Alabama. Magnolias
pon everywhere in Upper Cretaceous floras from Greenland to Alabama.
order Rosales has no very definite or remarkable representatives in the South Carolina
s. A form doubtfully identified as Hamamelites occurs at Rocky Point, and two
referred to Leguminosites, one as a Cmesalpinia, and one as an Acacia-like form.
prder Geraniales, though poorly represented, contains two remarkable forms, a Citro-
ery close to the modern genus Citrus, which ranges northward as far as New Jersey,
mnophyllum, a genus allied to the modern genus Croton of the Euphorbiaces. Croto-
is rather common in the Middendorf member, the only other species of the genus
ftnomanian form from Bohemia.
order Sapindales is a large one, represented by six species-one a Sapindus, one a
fully referred to the genus Pachystima, one a large Rhus, and three species of
gyllum. The last genus becomes abundant at the close of the Lower Cretaceous,
Species in the Patapsco formation. There are no less than nine species in the Raritan
Sbut this number is reduced to two in the Magothy, two in the Black Creek, and three
endorf member. Celastrophyllum is unrepresented in the Montana flora of the West.
]--14----5







66 UPPER CRETACEOUS AND EOCENE FLORAS OF SOUTH CAROLINA AND GEORGIA.

The order Thymeleales also includes six species, of which two are well-known species of
Laurus, two are species of Laurophyllum confined to this horizon in the Atlantic Coastal Plain,
and two are species of Cinnamomum, one being the widespread Cinnamomum newberryi Berry
and the other new.
The order Myrtales has three species of Eucalyptus, all close to modern species. This is
particularly true of Eucalyptus angusta Velenovsky, which is also associated with typical
Eucalyptus fruits in the Cenomanian of. Europe.
The order Umbellales is represented by a single species of Hedera, a form of wide distribution.
The Araliaceae, so common in homotaxial American floras, have not beendetected in either the
Middendorf member or other deposits of the Black Creek formation in South Carolina, though
a single Aralia is found in the Black Creek formation of North Carolina and several species
appear in the Tuscaloosa and Magothy formations.
The order Ericales has four species of Andromeda, one new and the other three widespread
in deposits of this or nearly the same age in the Atlantic Coastal Plain.
The Primulales are represented by a single rather widespread species of Myrsine and the
Ebenales by two species of Diospyros, one the relatively large-leaved Diospyros primreva of
Heer, which ranges from Greenland to Alabama in eastern North America and from Kansas and
Nebraska to Texas in the interior, and which is exceedingly common in the South Carolina
Cretaceous; and the other the less common Diospyro8 rotundifolia Lesquereux of the Dakota
sandstone and the Magothy formation. These leaves are certainly allied to the modern
Diospyros and are associated in the Raritan formation of Maryland with characteristic fruit
calices scarcely distinguishable from those of certain modern species.
PHYSICAL CONDITIONS INDICATED BY THE FLORA.
An effort to picture accurately the environment of a fossil flora is beset with unusual
difficulties, as may be readily imagined, and these difficulties increase in geometric ratio to the
time that has elapsed since it existed in life. Furthermore, the science of plant ecology is so
recent in development that the data with which to compare fossil floras are very inadequate.
Particularly is this true of the existing floras of the tropical and subtropical zones with which
the Cretaceous flora is most naturally compared. The strand flora of the tropics is fairly
well known, but of inland and upland floras over wide areas practically no information is
available, although the Philippine Forest Service has made a most laudable beginning in this
direction.
For this reason the discussion which follows probably merits more or less criticism,
especially of the statements where the writer has endeavored to particularize. An effort has
been made, however, to avoid purely speculative points, and it is believed that the more general
statements regarding climate and other conditions of growth will stand, whatever may be
the fate of the details.
As has been shown, the sediments and their contained floras indicate shallow seas and a
considerable elevation and relief of the Piedmont area. River gradients were high and the
streams numerous and more or less torrential in character. In the early part of the period
coastal sounds or bays were present, but these were subsequently submerged and the coast
line appears not to have been subsequently broken by any large reentrants, although part of
the strand flora was probably a swamp flora and swamps were also present in the lower courses
of the streams, especially in the later half of the period. With regard to the climatic conditions
the Cretaceous floras are to a certain extent unlike those of later periods and are so far removed
from the present that no very precise conclusions are possible. It is safe to assume that the
climate was mild, however, for the plant grouping clearly shows this. Seasonal changes were
not strongly marked, as is shown by the lack of periodic alterations (growth rings) in the petri-
fled and lignitized woods. These have not been critically studied, nor are they included iT
the systematic account of the flora, but they have been examined sufficiently to corroborate th(
foregoing statement. That the climate was not tropical in character may be assumed from
the manner in which this flora preserves its integrity when traced northward over man
degrees of latitude. It is essentially a unit from Alabama to New Jersey, and preserves eve:







B UPPER CRETACEOUS FLORA OF SOUTH CAROLINA. 67
hern occurrence on the west coast of Greenland-though differences are percepti-
remarkably similar. Even in Greenland this flora has nearly as many species
modern floras are customarily associated with the warmer temperate and subtropical
its most southerly occurrence. We may confidently assert that frosts were unknown.
were not all evergreen; although some were, and the leaves, from the manner in which
found fossil, were apparently shed at maturity and not seasonally. It is believed
nation, or the pseudoxerophytism of swamp habitats, rather than any approach to
lains the presence of numerous coriaceous-leafed forms, such as the four species
eda, and the abundance of LeguminosEe and of gymnosperms, many of them having
ves like Brachyphyllum, Moriconia and Widdringtonites, or developing phylloclads
Protophyllocladus of the Middendorf member or Androvettia of the homotaxial beds
a Creek formation in North Carolina and the Eutaw formation in Georgia.
oia in the existing flora thrives only in a belt fed by the moisture-laden winds from the
As a fossil it is excessively abundant in the Middendorf member and, in fact, in all
tan deposits of the Atlantic Coastal Plain Cretaceous which furnish any flora. That
us rainfall was plentiful may be inferred not only from the species of plants pre-
t also from the formation of dripping points on various leaves, this feature being espe-
hasized in the Tuscaloosa flora of Alabama. Judged by the facts of the present-day
* distribution of plants, the flora as a whole presents an antipodean facies with its
f Eucalyptus and Proteoides and its abundant Araucariacese, but this is only another
phasizing its Mesozoic character, for the abundant evidence at our command shows
these types were practically cosmopolitan in the Mesozoic. Another feature which
e to modern plant geographers is the curious mingling of forms which in the exist-
' are to a greater or less extent climatically segregated. Willows and walnuts growing
p, eucalyptus, laurels, and araucarias would indeed be anomalous in the present flora,
e and similar associations are familiar enough in fossil floras, not only during the
in but well into the Cenozoic.
though no close comparisons with modern ecologic groups are possible, it would seem
the Upper Cretaceous flora were existing at the present time it would be included by
botanists under that somewhat elastic head which Schimper calls "temperate rain
" In no other modern plant associations do we find that commingling of temperate
ical types that we find in certain present-day temperate rain forests, as, for example,
,southern Chile, southern Japan, Australia, and New Zealand. In the last-named
SAralia, Laurus, Cinnamomum, Magnolia, and Sterculia are associated with Quercus,
Gleichenia, Dryopteris, and Dicksonia. In some respects the flora of New Zealand is
lopical in its facies and more like our eastern Upper Cretaceous floras than any other
sting. In New Zealand conifers are abundant and include forms with reduced leaves
and Dacrydium, as well as forms with broad leaves like Dammara, Podocarpus,
llocladus. Dicotyledonie are numerous and varied, including between 100 and 150
among which forms of Myrtacese, Lauraces, and Proteaces, with coriaceous leaves, are
Mit. The undergrowth is rich in tree ferns and various genera of Araliacese.
this modern flora is compared element for element with the Coastal Plain Cretaceous
y differences naturally become apparent; nevertheless, the resemblance between the
remarkable. In the Upper Cretaceous flora of the Coastal Plain the narrow or scale-
nifers are represented by Sequoia, Moriconia, Brachyphyllum, and Widdringtonites.
Ia represents the broad-leafed araucarias; Androvettia and Protophyllocladus represent
Phyllocladus. The Dicotyledonse are numerous and varied; temperate and trop-
e are mixed, and there are numerous coriaceous forms belonging to a number of the
lilies as in the New Zealand flora. Aralias are common in both floras.
t the physical conditions approximated the foregoing outline is further indicated by
nee of many plants which normally grow .in streams with considerable flow or along
margins. There is an aquatic species, Potamogeton middendorfensis, which it is hard to
Growing at sea level in the latitude of South Carolina during the Upper Cretaceous.
two large-leafed grasses (Arundo and Phragmites) and a sedge, all strongly mesophytic








(6' UPPER CRETACEOUS AND EOCENE FLORAS OF SOUTH CAROLINA AND GEORGIA.

types; four species of willow; and four thick-leafed laurels and two Cinnamomums, whose
existing descendants flourish in humid localities, as do those of the numerous figs and magnolias.
The figs are very abundant individually in the Middendorf member, especially at localities like
that at Rocky Point. They are for the most part lanceolate-leafed forms and have developed
to a greater or less extent long attenuated tips that are absent in the same species toward the
northern limit of their range. These tips are commonly known as dripping points and are
understood to indicate a considerable rainfall. They are especially noticeable in the South
Carolina representatives of Ficus crassipes Heer. It is believed that these figs found their
optimum conditions in the coastal swamps and extended from them up the river valleys to the
amphitheaters toward the heads of streams, though they were not confined to such situations,
as is proved by their wide geographic distribution at this time. In similar situations grew the
broad-leafed gymnosperms, the palms, the entire-leafed Quercus sumterensis, and the various
species of Proteoides, Illicium, Eucalyptus, Diospyros, Cinnamomum, Citrophyllum, Laurus,
and Magnolia. The flora of the typical Black Creek deposits of South Carolina is so scanty,
embracing only 17 species, that it does not in itself furnish adequate data for an attempt to
picture its environment. When supplemented by the Black Creek flora of North Carolina
somewhat more data are provided, but still the material is insufficient for the purposes men-
tioned. The occurrence of obscure remains of a marine alga, present also in North Carolina and
in the Magothy of Maryland, may be noted as indicative of the presence of the sea or at least
of brackish water. Lignite and amber are common, also the remains of Araucaria, Cephalotaxus-
like fruits, Eucalyptus, Ficus, Myrica, and similar forms. The flora is mixed, including upland
types which must have made perhaps very considerable river journeys before fossilization.
Strand and swamp plants are also present, and coriaceous forms predominate, owing to their
survival in agitated waters which destroyed the more delicate plant remains. The character
of the fossils is a clear indication that the bays or sounds, which had been present in at least a
part of the area during Middendorf time, had disappeared, and the Coastal Plain lacked large
estuaries which usually afford such admirable means for fossilization. We may therefore infer
that the coast line was unbroken, or that if there were estuary plant beds they have been
destroyed by the erosion of the landward margin of the deposits or are not exposed at the
present time. The differences in the flora between the early Middendorf and late Black Creek
were probably not considerable, and physical conditions were not very dissimilar. The land
had approached sea level and the ground water would be nearer the surface. The climate
probably underwent no appreciable change, and the rainfall and humidity were still ample,
although the writer is inclined to think that the rainfall was somewhat diminished. The natural
radiation of the individual species in the floras doubtless caused great changes in distribution,
and other changes were doubtless due to the fact that the area was in the direct line of migra-
tion between the north and the south, but sufficient data to illustrate these various changes
have not yet been accumulated.
CORRELATION OF THE BLACK CREEK FORMATION.
In considering the correlation of the Middendorf member and the other Black Creek deposits
the first question to be decided is the relation which they bear to each other. Though the writer's
position on this point may be inferred from what has gone before, a few comments are necessary,
as at first sight the floras apparently show considerable differences.
Of the 76 species described in the foregoing pages, 62 come from the Middendorf member
and 17 from the other deposits of the Black Creek formation. Of these 76 species the following
are described as new:
Acaciaphyllites grevilleoides. Ficus celtifolius.
Algites americana. Heterolepis cretaceus.
Andromeda euphorbiophylloides. Illicium watereensia.
Araucaria darlingtonenais. Leguminosites middendorfensis.
Calycites middendorfensis. Lycopodium cretaceum.
Celastrophyllum carolinensis. Momisia carolinensis.
Cinnamomum middendorfensis. Pachystima? cretacea.
Crotonophyllum pandurseformis. Potamogeton middendorfensis.








*. UPPER CRETACEOUS FLORA OF SOUTH CAROLINA. 69

la. Rhus darlingtonensis.
us lobatus. Sabalites carolinensis.
terensis. Salix sloani.
doweetfalica. Strobilites anceps.
ugh these new species are of slight value in exact correlation, their evidence is entirely
with that deduced from the remainder of the flora. The bulk are from the Middendorf
only Algites americana, Araucaria darlingtonensis, Rhus darlingtonensis, and Strobilites
Sming from the other Black Creek deposits. Extensive collections from all parts
tal Plain are undergoing elaboration by the writer, who, since the original descrip-
anuscript, has identified Algites americana from the Black Creek formation of North
and from the Magothy formation of Maryland, and Protophyllocladus lobatus from
hy of, Maryland. Araucaria darlingtonensis has been shown to be closely allied to
Black Creek araucarias, and Andromeda euphorbiophyUoides, in all probability, occurs
order between the Tuscaloosa and the Eutaw in Hale County, Ala. Quercus pseudo-
is now known to occur in the Black Creek formation of North Carolina, and the
is seen to be close to Sabalites magothiensis Berry from the Magothy formation of
rn Coastal Plain from New Jersey to Maryland. The genera Crotonophyllum and
are most closely related to species described from the Cenomanian of Bohemia. None
era occur in the Montana flora of the west except Ficus and Salix, genera present
Raritan to the Recent.
y the following species are common to the Middendorf member and other deposits
ack Creek formation in the South Carolina area:
Irliana Heer. Proteoides lancifolius Heer.
yllum nervillosum Hollick. Salix lesquereuxii Berry.
Though this seems a very small common element, all the forms are typical of this stage
Cretaceous, and all except Proteoides are especially characteristic of the Magothy, Black
and Tuscaloosa floras. The identity of the Middendorf and other Black Creek floras is
strengthened when comparisons afe made with the latter flora as developed in the
Carolina area, the result showing 27 species common to the two formations.
he following Middendorf species which have not been found in the other Black Creek
0 of South Carolina occur in beds of Black Creek age in North Carolina:


ja grandifolia Berry.
novwecwesarete IHollick.
parlatorii Heer.
Sjeffreyi Berry.
hyllum macrocarpum Newberry.
hyllum crenatum Heer.
mum newberryi Berry.
rnum aligerum (Lesquereux) Berry.
*amites elegans (Corda) Endlicher.
primeva Heer.
geinitzi (HIeer) Heer.
'pee Heer.
na Heer.
[phensoni Berry.


Juglans arctica Heer.
Laurophyllum elegant Hollick.
Leguminosites robinifolia Berry.
Magnolia capellinii Heer.
Moriconia americana Berry.
Myrsine gaudini (Lesquereux) Berry.
Phragmites pratti Berry.
Pinus raritanensis Berry.
Podozamites knowltoni Berry.
Quercus pseudowestfalica Berry.
Salix flexuosa Newberry.
Salix lesquereuxii Berry.
Sequoia reichenbachii (Geinitz) Ieer.


Ie following species occur in the Middendorf member but have not been detected in the
lack Creek deposits, either in North or South Carolina:


yl8ites grevilleoides Berry.
&da euphorbiophylloides Berry.
enlandica Heer.
ia middendorfensis Berry.
middendorfensis Berry.
k Berry.
yllum carolinensis Berry.
hyllum elegans Berry.
mum middendorfensis Berry.
yUllum pandurieformis Berry.


Dewalquea smith Berry.
Diospyros rotundifolia Lesquereux.
Eucalyptus wardiana Berry.
Ficus atavina Heer.
Ficus celtifolius Berry.
Hamamelites? cordatus Lesquereux.
Heterolepis cretaceus Berry.
Illicium watereensis Berry.
Laurophyllum nervillosum Hollick.
Laurus atanensis Berry.








70 UPPER CRETACEOUS AND EOCENE FLORAS OF SOUTH CAROLINA AND GEORGIA.

Laurus plutonia Heer. Protophyllocladus lobatus Berry.
Leguminosites middendorfenais Berry. Potamogeton middendorfensis Berry.
Lycopodium cretaceum Berry. Quercus sumterensis Berry.
Magnolia tenuifolia Lesquereux. Sabalitee carolinensis Berry.
Magnolia obtusata Heer. Salix pseudohayei Berry.
Momisia carolinensis Berry. Salix Bloani Berry.
Onoclea inquirenda (Hollick) Hollick. Sapindus morrisoni Heer.
Pachystima? cretaces Berry. Widdringtonites subtilis Heer.
Proteoides parvula Berry.
These include 17 new species thus far confined to the Middendorf member and without
value in close correlation. Of the remaining 21 species with a wider distribution 4 occur in the
Raritan of New Jersey, 6 in the Tuscaloosa formation, the large number of 15 in the Magothy
formation, 8 in the Dakota sandstone, and 9 in the Greenland Cretaceous. This illustrates very
well the remarkable unity of this flora from Greenland to Alabama, and also indicates conclusively
the practical synchroneity between the Middendorf member and the other deposits of the Black
Creek formation of South Carolina.
A similar relationship is shown by the following list of North Carolina Black Creek species
which have not been detected in either the Middendorf member or the other Black Creek deposits
in the South Carolina area:
Acerates amboyense Berry. Malapoenna horrellensis Berry.
Androvettia carolinensis Berry. Myrica cliffwoodensis Berry.
Araucaria clarki Berry. Myrsine borealis Heer.
Celastrophyllum undulatum Newberry. Phaseolites formus Lesquereux.
Cinnamomum heerii Lesquereux. Pisonia cretacea Berry.
Cornophyllum sp. Pistia nordenskioldi (Heer) Berry.
Cycadinocarpus circularis Newberry. Planera cretacea Berry.
Dammara borealis Heer. Pterospermites carolinensis Berry.
Dewalquea groenlandica Heer. Pterospermites crednerafolia Berry.
Eucalyptus attenuata Newberry. Quercus pratti Berry.
Eucalyptus linearifolia Berry. Salix newberryana Hollick.
Ficus ovatifolia Berry. Salix eutawensis Berry.
Ficus daphnogenoides (Heer) Berry. Sassafras acutilobum Lesquereux.
Gleditsiophyllum triacanthoides Berry. Sequoia heterophylla Velenovsky.
Kalmia brittoniana Hollick? Sequoia minor Velenovsky.
Liriodendron dubium Berry. Tumion carolinianum Berry.
Liriodendron cf. primsevum Newberry.
Here again the species which are not confined to North Carolina are mingled in other
homotaxial deposits, such as those of the Magothy formation of the North Atlantic Coastal
Plain or the Tuscaloosa formation of Alabama, with the previously enumerated forms not com-
mon to the described floras of the two States.
The conclusion seems incontrovertible that in correlation the Middendorf member and the
typical Black Creek may be considered as a unit, and that within the State they confirm the
field observations that the stratigraphic sequence comprises initial Middendorf sedimentation
of short duration, contemporaneous Middendorf and typical Black Creek sedimentation, espe-
cially when the Peedee and Aiken areas are compared, and finally typical Black Creek sedimen-
tation only for a considerable period. When the flora as a whole is compared with outside
areas it is brought out that 32 of the species in the Black Creek formation of North Carolina are
of admitted identity in South Carolina and that 26 species occur either in the Tuscaloosa for-
mation of Alabama or the Eutaw formation of Georgia. It seems to the writer that the syn-
chroneity between these beds and the upper part of the Alabama Tuscaloosa and the lower
portion of the Eutaw must be admitted, for the close similarity in their floral characteristics
is corroborated by similar lithologic characters. It is believed, however, that the lower part
of the Tuscaloosa of western Alabama is older than any Upper Cretaceous of the eastern
Gulf or Atlantic Coastal Plain as far north as the New Jersey-Maryland area, where the upper-
most Raritan is to be considered contemporaneous with it, the rest of the Raritan formation
being still older. With the Magothy flora of the northern Coastal Plain, the Middendorf flora
has 35 species in common, or about 50 per cent of its total flora. That the two floras are
essentially a unit seems certain. In the New Jersey area the Magothy flora is confined to







UPPER CRETACEOUS FLORA OF SOUTH CABOLIN.'.. 71

*tional beds lying above the Raritan formation and at the base of the marine series of
whereas in the southern and Gulf coastal plains similar fossil plants occur in lenses
interstratified with marine fossiliferous beds. This shows that the southern repre-
of the Magothy flora was contemporaneous with the southern representative of the
fauna and makes it probable that the Magothy flora in the New Jersey-Maryland area
through Matawan time. The Matawan formation has yielded only one or two fossil
Ficus in the clays of Woodbury, N. J., and a Dammara in Maryland-and in both
species occur in the Magothy formation also.
Black Creek flora contains 17 species found in the Atane beds of Greenland and 9
the Patoot beds. In the Cenomanian of Europe 9 of its species are found, in the florally
erous Turonian 1 species, and in the Senonian 4 species.
e. eastern Cretaceous floras above the Raritan, possibly including those of the upper-
.aritan, correspond with the flora usually known as that of the Dakota sandstone. The
containing them are conformably overlain by deposits carrying a marine fauna but very
plants. There are 23 species common to the South Carolina Cretaceous and the
)& sandstone. The Black Creek formation of North Carolina, out of a total of 66 species,
O species in common with the Dakota. The Montana group flora is entirely unlike the
, Cretaceous floras, having scarcely a single element in common. Certain stages in the
ton of the eastern Cretaceous flora can be made out, though in general the forms have a
stratigraphic range. This is paralleled, however, by an almost equally wide stratigraphic
of the faunas.
'o render intelligible to a wider circle of readers the results of the present discussion the
ble European equivalents of these floras should be indicated, although it is admitted that
apted exact correlations between geologic formations on opposite sides of the Atlantic
always be more or less untrustworthy.
In Europe there are available for comparison abundant Cenomanian floras in Portugal,
ee, Germany, and especially in eastern Europe (Bohemia, Dalmatia, and other regions).
Turonian of Europe, on the other hand, is for the most part lacking in fossil plants, which
me abundant again in the Senonian of Prussia, Saxony, and Bohemia. Our Dakota flora
ways been considered Cenomanian; most paleozoologists have considered the Benton as
(a view that is widely accepted in this country at the present time); and the Montana
been uniformly considered as representing part of the Senonian. The Atlantic
s floras have been considered Cenomanian, and the associated and overlying faunas
(exclusive of the Rancocas and Manasquan faunas of New Jersey, which have been
considered Danian by Clark and others).
the writer's opinion no Cenomanian floras are known in America, unless the Racitan
and that of the Washita group of the Texas and southern Arkansas area represent that
of European geology, and the post-Raritan floras of the East1 are for the most part of
age, as is also the major part, at least, of the Dakota flora of the West.2
e occurrence and range of the species of the South Carolina Cretaceous which form the
basis of the foregoing discussion are fully set forth in the accompanying tables:
Approximate equivalents of the plant-bearing Cretaceous deposits of South Carolina.

South Carolina. North Carolina. Western Alabama. New Jersey-Maryland. Europe.

Marine Cretaceous Marine Cretaceous Marine Cretaceous Marine Cretaceous Emseherlan.
(no known flora). (no known flora). (no known flora). (no known flora).
Black Creek. Eutaw. Matawan
(Marine, with no known Turonlan.
flora).
= r� Black Creek. Magothy.
ber of Magohy.
ok Creek Tuscaloosa.
Hiatus. Hiatus. iatus Raritan. Cenomanian.

Lower Cretaceous Lower Cretaceous Carboniferous. Lower Cretaceous, Triassic,
(no known flora). (no known flora). or crystalline rocks.
I With the exception of the meager floras in the Ripley formation and its equivalents.
I Some paleontologists consider the Coastal Plain plant-bearing formations to be lower Senonlan.












72 UPPER CRETACEOUS AND EOCENE FLORAS OF SOUTH CAROLINA AND GEORGIA.


Distribution of Upper Cretaceous plants in South Carolina and their range elsewhere.


Black Creek formation of South Carolina.


Middendorf arkose � S
member.


v ---------- - (Z








Algites americana ................. ......... .......................
Onole inquirenda ..........................
Lycopodium cretaceum ...... ....
odo sites knowitoni ........ .......... ....
Protophyllocladus lobatus ..................... .. ....
Cephalotoxospermum carolinanum.... ........ .... .... .... .... ....
Ara caria bladenensis ........................ . ....... ....
Araucaria darlingtonensis ............ .... .... A .
Araucaria jeffrey ........................... .... . ...
Brachyphyllum macrocarpum .. X .
Pinus raritanensis ....................... X .....-- - - ...........
Sequoia reichenbachi ...................X................... X
Cunninghamites elegans .......................... . . .. . .... .... .... ....
Widdringtonites subttilis ................. X X
Moriconia americana......... .......... .
Strobilites anceps .. ............... .... .... ....
Heterolep cetacus. ...... ...........
Potamogeton middendorfensis............ - - - - - - - -X- --* * . .
Phagmites pratti ...
A r gmde g rnl a d ... ...................... ....
Carex clarki. ... .. ...................... .. . .... X . . . . . . . .... .... . ...
Sabalites carolinensis ...... ... ...
Juglans artica . ........................... X . . . . ... ... .
Myrica brittoniana ........... .. - . . . - . . .
M yrica elegans ................. ...... ....... . .... .....
Sa .x flexu osa ye. ............... .... .. .. -- .------- X .... X ........
Salix flexuosa que ... ............ .....----- X X X..

Salix pseudohayei ............. - . X - . . ... .. ....... ..
Salix sloani ...... ................... . ....... . ...
Quercus sumterensis ....................... . X .. ............
Quercus pseudowestfalhca ......... ........- X X I. . . . ..
Momisiacarolinensis........................ X .... ... . -
Ficus atavina .............. . X X ........
Ficus celtifolius .................. .. . . X . . ..
Ficuscrassipes ............................ x X X X .. . . .
Ficus kraus ana.. ................. X ... - X X ..
Ficus stephensoni ................ X ..
Proteoideslancifolius .......... . . .. ..... X X .... .. .......
Proteoides parvula ............ ... X .............
, .... . . .. . . .... ..
M .., ,, . .. ... x x ........ ..... .
yA 1 , , ... X ..... ........
- , i- -i i . .. x I ..... .... . .



C salpina.... iddendorfe ..... .. .. .. . . . . . ........
: i. . .. .... . - . ........ . . . . . . .



( i j. X . . . . . .. .
C.Sapindasmorrisoddeni s ........... X ..........

Sapindus m orrlsoni ..................... . ... ... . ... X - - ---- -- ---- ---- --
Pachystima (?) cretace .............. . . . X ....... ....
Celastrophyllum elgis ........... ..... .. ... x .-. .--. . ...
Colastrophyllunim crentum ............... ...
Celistrophilumncarolinensis ............... ... ........ X
Rhis darlingtonensis X -- - -
Laurusplutonia ................ .. .... . . . . . . . ..
Laurus atanensis ................... X
Laurophyllum elegans .................. ..
Laurophyllum nervillosi .. . ........ -. .. ...
Cinnamomum newhenorfei.. ......- . ......
Cinnamomum iiddendorfensis ............ .. . i - --
] , . . ....... x

] .. .. . ... . ... . x .... ....
] I i .i.i I .. .... .. . . . . - - -

Andromeda novesare ............... X .. ...
Andromedagrandifolia ...... ............ X .. .
Andromeda euphorbiophylloides........... ..
Andromeda parlatorii ....... ... X
M rsinega dini .. -. ... .
Diospvros prim i........ .. X X X
Diospvros rotundoi ifi ... ..- . X -... ... .
tCalycites middendorfensis .. . ...-- ---.........


X X.... .. ....

x .... -.. ....
x x .... . ....
. x .... .... ....
.... .... .... ....
..X X X ....


x x........
S.... x ....
X X X . ...
.x x x .... ..
.x x xx x
, .-X X X . -
. .x x x x . - - - -
. x ..... ....

X ..."'' X " X . "

x x ..
----. x ---- .----

. .. . . . . . . . . . . . .. .











.A. .... . . . x
.A.






. .... ..... . x
. .... .... .... ....
.xlx x x ..x.


x. ... ....
-.. x x x x








.. x x.. ...x
.X .. . .. .. ....




. ... . . . . . . . . . . . .. . . . ..
.X X X X
---------"x -----------











. ... ... x ...... ....
X X X ... X


. .... .x.. . ... . ... ... .
SX X X .. X -.
. x - -- - - - - - -. - . . -













.x x x .... x

. x . xx..x.
. . X X . .... X












....- x .- X - --
I ... .... . .x -. -. ...
-. X X X .. X








XX ..X..

.X X X . . X
. .... ... .. .. .... ....
. X .. .. .X.. . .. .
X . . .. .. . .. . . .
X X X - - - -
x xx..
. ... ... .. X. . . .
x \ -- --
X X .. .. . .. . .. .
X -- - - - - -




xX x
X < X


... x ..








.... . ..









.... x. ...
.... x ....




.x.. .x ....
.... x ;- ---
.... X I

















x... x ...
xX




























. . .. . .
... ... . x'.-..
.... .. ....
















.... ... ....
I















.... ....-...






'X X --
. X. . ... . ..


.... X . . ..

x --















.... ..X ...
... .. . . .

... .. . . .

.. . . .. . .

.. .. . . .
X ...


X X
.... x ....


.... ......
















. x.. .......




















X.. X......
x. . .. .. ....























X X X -..




























PLATE II.




I 73

































PLATE II.
Page.
FIGURES 1-6. Lycopodium cretaceum Berry, Middendorf..................................................... 15
1. Large fruit spike.
2. Smaller fruit spike, X 5.
3. A single scale, X 10.
4. Diagrammatic ventral view of sporophyll, X 10.
5, 6. Diagrammatic view of sporophyll in median longitudinal section, X 10.
FIGURES 7, 8. Onoclea inquirenda (Hollick) Hollick, Middendorf............................................. 14
FIounRES 9-13. Protophyllocladu lobatus Berry, Rocky Point................................................ 17
FIGUREs 14-17. Widdringtonites subtilis Heer, Middendorf.................................................. 25
14, 15. Cones.
16. Twig.
17. Twig enlarged.
FIGURES 18, 19. Widdringtonites reichii (Ettingshausen) Heer. Cone-bearing twigs from the Cenomanian of
Dalmatia, introduced for comparison.
All the specimens except those shown in figures 18 and 19 are from the Middendorf arkose
member of the Black Creek formation.







PROFESSIONAL PAPER 84 PLATE II


5


6


4


10


UPPER CRETACEOUS PLANTS FROM SOUTH CAROLINA.


2






3


17





























PLATE III.




75



































PLATE III.
Page.
FIGURE 1. Araucaria darlingtonensis Berry, near Darlington................................................. 20
FIGURE 2. Brachyphyllum macrocarpum Newberry, Middendorf .---.......................................---. 21
FIGURE 3. Heterolepis cretaceus Berry, Rocky Point ..............................-.........--- ............. 27
FIGURE 4. Cepholotaxospermum carolinianum Berry, near Florence......................................----------. - 18
FIGURE 5. Strobilites anceps Berry, near Darlington......................................................... 27
FIGURES 6, 7. Detached leaves of Araucaria bladenensis Berry, near Florence ............................... 19
The specimens shown in figures 2 and 3 are from the Middendorf arkose member of the Black
Creek formation. Those shown in figures 1, 4, 5, 6, and 7 are from other beds in the Black
Creek formation.
76




PROFESSIONAL PAPER 84 PLATE III


4


w4b
, 440


i


A


A


' A

N4


6


UPPER CRETACEOUS PLANTS FROM SOUTH CAROLINA.


U. & GEOLOGICAL SURVEY




























PLATE IV.




77






































PLATE IV.
Page.
FIGURES 1-4. Sequoia reichenbachi (Geinitz) Heer, Middendorf.............................................. 23
1. Cone-bearing twigs.
2-4. Twigs showing variations in size and appearance.
FIGURE 5. Podozamites knowitoni Berry, Rocky Point................................ .. ...................... 16
FIGURE 6. Potamogeton middendorfensis Berry, Middendorf.......................--........................ 27
FIGURE 7. Arundo granlandica Heer (?), Rocky Point.................................................... 28
All the specimens are from the Middendorf arkoee member of the Black Creek formation.
78





PROFESSIONAL PAPER 84 PLATE IV


,'u
I''



6





I


/




4


UPPER CRETACEOUS PLANTS FROM SOUTH CAROLINA.


U.& GEOLOGICAL SURVEY





















PLATE V.









































PLATE V.
Page.
Sabalites carolinensis Berry, showing broad rays. From Middendorf arkose member of the Black Creek forma-
tion, near Langley ..................---.........................--.....----..... --...............--............ 29
80

























































































UPPER CRETACEOUS PLANT FROM SOUTH CAROLINA.


AL SURVEY PROFESSIONAL PAPER 84 PLATE V
































PLATE VI.




8069-1 -81
80690-14--





































PLATE VI.
Page.
Sabalite carolinensis Berry, showing keeled and narrowed rays near point of insertion on the rachis. From
Middendorf arkose member of the Black Creek formation; near Langley................................ 29
82









PROFESSIONAL PAPER 84 PLATE VI


4


UPPER CRETACEOUS PLANT FROM SOUTH CAROLINA.


U.S. GEOLOGICAL SURVEY



























PLATE VII.




83
































PLATE VII.
Pat
FIGURES 1-4. Moriconia americana Berry, Middendorf ................-.........-------... ..------------- 26
FIGURES 5-10. Crotonophyllum parndurformis Berry .......................................--...----------..----...--------... 48
5. Rocky Point.
6-8. Langley.
9-10. Middendorf.
FIGURns 11-13. ali lesquereuxii Berry .--------...-...---.----.........-----..---------........-----........-------------------
11. Middendorf.
12. Rocky Point.
13. Langley.
FIoURES 14-16. Sali flexuosa Newberry, Middendorf -.........---------------.. ------............----.. 32
FIGURES 17, 18. Myrica brittoniana Berry, Black Creek ..................................................... 31
All the specimens except those shown in figures 17 and 18 are from the Middendorf arkose
member of the Black Creek formation. The specimens shown in figures 17 and 18 are from other
beds in the Black Creek formation.
84







PROFESSIONAL PAPER 84 PLATE VII


18








10
11/




14















8 13 12 15 16


UPPER CRETACEOUS PLANTS FROM SOUTH CAROLINA.


U. S. GEOLOGICAL SURVEY



































PLATE VIII.

















I-


86



































PLATE VIII.
Page.

FIGURnE 1,2. Juglans ardcica Heer, Rocky Point......................................................... 30
FIGURES 3-9. Dewalqufa smith Berry, near Langley. ........................................... 41
Fiuorse 10-12. Salizx loai Berry, near Langley ..................................... 34
FIGURE 13. Leguminosites middendorfenads Berry, Middendorf................. ........ ................ 46
FIGURE 14. Cinnamomum middendorfensis Berry, Middendorf............................................... 55
All the specimens are from the Middendorf arkose member of the Black Creek formation.
86








PROFESSIONAL PAPER 84 PLATE VIII


11
11


14 8
UPPER CRETACEOUS PLANTS FROM SOUTH CAROLINA.


1









13


U. 8 GEOLOGICAL SURVEY




















PLATE IX.

































PLATE IX.
Pag.
FIGURE 1. Oinnamomum middendorfendi Berry, Middendorf ---------------.........................--------------------...................... 55
FIGURES 2, 3. Magnolia tenuifolia Lesquereux (?), Middendorf ..................---------------------------......---------...... 44
FIGURE 4. Myrica dlegans Berry, near Darlington..................... .................... 31
FIGURE 5. Quercus pseudouwetfalica Berry, Rocky Point.................................................... 35
FIGURE 6. Sapindus morrisoni Heer, near Langley........................................................ 49
FIGURES 7, 8. Rhus darlingtonensis Berry, near Darlington.......................-----........................... 51
FIGURES 9, 10. Acaciaphyllitee grevilleoides Berry, Middendorf ................................................ 45
FIGURE 11. Leguminosites robin folia Berry, Langley....................................................... 46
FIGUREs 12, 13. Cinnamomum newberryi Berry, Rocky Point.............................................. 54
All the specimens except those shown in figures 4, 7, and 8 are from the Middendorf arkose
member of the Black Creek formation. The specimen@ shown in figures 4, 7, and 8 are from
other beds of the Black Creek formation.
88







PROFESSIONAL PAPER 84 PLATE IX


2
1








7






(i) .i


6 1






10 1




11


8 4


12 18


UPPER CRETACEOUS PLANTS FROM SOUTH CAROLINA.


U. S. GEOLOGICAL SURVEY



























PLATE X.




89
































PLATE X.
FIGURE 1. Proteoides lancifoliua Heer, Middendorf........................................ ............ 40
FIGURE 2. Hamamelitesf cordatus Lesquereux, Rocky Point......................---------------------............... 45
FmIGURE 3. Magnolia capellinii Heer (?), Rocky Point............................. ................ 43
FIGuaE 4. Calycites middendorfensis Berry, Middendorf.....................................--------------------------- ---------.. 63
FIGURE 5. Proteoides parvula Berry, Middendorf ..................................................... 40
FIoun 6. Pachystimaf cretacea Berry, Middendorf .................. .................................... 49
FIGURE 7. Cnsalpinia middendorfends Berry, Middendorf ......................................... 46
FIGURE 8. Saliz pseudohayei Berry, Middendorf....................................................... 34
FIGURES 9, 10. Quereus sumterensis Berry, Rocky Point..................................................... 35
FIGURE 11. icus atavina Heer, Rocky Point............................................................. 36
FIGURE 32. Ficus crassipes Heer, Rocky Point................... ...................................... 37
All the specimens are from the Middendorf arkose member of the Black Creek formation.
90








PROFESSIONAL PAPER 84 PLATE X


*
4
7
I' . I



Ik~ 6







1 I~


UPPER CRETACEOUS PLANTS FROM SOUTH CAROLINA.


U. S. GEOLOGICAL SURVEY