'flk ^v5i''^^

dp 4

.1^416
1903

THE FLORA OF
THE MATAWAN FORMATION

(CROSSWICK'S CLAYS)

By EDWARD W. BERRY

LICRARV
NEW YORK
BOTANICAL

GARDEN

NEW YORK

September, 1903

From the Bulletin op the New York Botanical Garden.
Issued September 12, 1903

[From the Bulletin of the New York Botanical Garden, Vol. 3, No. 9. 1903.]

The Flora of the Matawan Formation (Crosswicks Clays).

By Edward W. Berry.

Introduction.
Some of the earliest of American geological writings refer
to the New Jersey Cretaceous, and the accessibility of this
area has ever since made it a favorite field for investigation.
Interest for a long time centered about the marl deposits and
those of the plastic clays because of their economic impor-
tance ; the present Matawan formation was included in the
*' plastic clay and sand formation" of the geologists of the
first half of the nineteenth century, and their stratigraphic posi-
tion was considered to be Lower Cretaceous by Rogers in his
first report published in 1840, although they were not clearly

(46)

defined by him. Professor Cook from 1863 until his death
in 18S9 published annual reports as state geologist of New
Jersey and early subdivided the Cretaceous into the three
marl beds, the clay-marls and the plastic clays. The Cre-
taceous was extensively summarized and described by him in
the Geology of New Jersey, published in 1868, the clay-marls
being divided into a lower member of clay containing green-
sand and an upper member consisting of laminated sands
The thickness of the formation was placed at 277 feet, 170
feet for the upper, and 107 feet for the lower member, and
over a dozen localities were enumerated where the clay-marls
were dug as fertilizer. In 1891 Professor William Bullock
Clark commenced a study of the coastal series of New Jersey
which has been in progress ever since. Three official reports
have been published : a Preliminary Report,* a Report of
Progress,! and a Final Report ; | besides numerous other
papers from which many of the following facts in regard to
the areal distribution and thickness of the Matawan forma-
tion have been quoted.

The name Clay-marls was proposed by Cook ; his char-
acterization was incomplete, however, and was confined almost
entirely to their development in northern New Jersey. This
name does not adequately designate the formation lithologi-
cally and has been superseded by the name Matawan
formation of Clark. § The Matawan is separated from
the Piedmont plateau by a tract of Raritan, or Lower Cre-
taceous, which is some ten to fifteen miles wide. The Mata-
wan is nine to twelve miles wide in Monmouth county,
becoming narrower to the southward, being reduced to about
six miles in width in southern New Jersey ; on the western
shore of the Delaware river in Delaware it is further reduced
to from two to three miles ; further south on the eastern
shore of Maryland it broadens, being about five miles wide

*Ann. Rep. State Geol. N.J. 1892: 167-245. 1893. (Clay-marls, pp.
186-190.)

tAnn. Rep. State Geol. N. J. 1893: 329-355. 1894.
t Ann. Rep. State Geol. N. J. 1897 : 161-210. 1898.
2 Jour. Geol. 2 : 163. 1894.

(47)

below the Sassafras river ; on the western shore in Anne
Arundel county the areal distribution is variable on account
of the broken character of the country, but is on the whole
narrower; further south, in Prince George county, it forms
but a narrow strip less than a mile in width.

The materials are variable ; sands and clays predominate.
The sands are sometimes white and coarse, but more com-
monly fine-grained and colored by iron, even causing local
induration, or they may be mixed with argillaceous materials
forming silvery micaceous sand, or chocolate-colored marl,
glauconite grains being present in greater or less amounts.
The clays are generally black or drab, locally carrying seams
and pockets of glauconite ; occasionally they are calcareous
as a result of their molluscan contents.

The thickness is variable, but becomes reduced to the
southward. It increases considerably to the southeast, judg-
ing from the well records.* In northern Burlington county
the Matawan is less than 200 feet thick ; east of Philadelphia
and Camden it is 125 feet ; in Gloucester county it is 175 feet
in places ; in Salem county it is 80 feet ; in Delaware not
over 60 feet ; near the mouth of the Sassafras river in Mary-
land it is 100 feet ; in eastern Anne Arundel county it is 60
feet ; in western Anne Arundel county and Prince George
county it is thinner, until at Fort Washington bluffs it is a
little more than 15 feet. Its farthest known southern appear-
ance is in the valley of Piscataway creek ; on the opposite
shore of the Potomac the Eocene rests directly on the Poto-
mac formation.

Long thought to conformably overlie the Raritan, an un-
conformity is now known to exist, although the time interval
was not very great. Along Raritan Bay in the vicinity of
Cheesequake creek where the upper Raritan contains dark-
colored clays, the interbedded sands and clays gradually
grade from Raritan into the Matawan. Further inland and
to the southward the interval was greater since the Matawan
gradually transgresses the Raritan and comes to rest, in cen-

*Woolmau, Anu. Rep. State Gcol. N. J. 1895 : 63-95. 1896.

(48)

tral and southern Maryland, upon the lower members of the
Potomac group which are not represented in New Jersey.
Elsewhere in New Jersey the upper Raritan consists of white
sands or fine gravel and the line of contact is sharp, except
where it is obscured by Tertiary or later deposits. The Mat-
awan is conformably overlain by the lower Monmouth forma-
tion ; the lithological differences are clearly marked, however.

In its northern portion the Matawan is readily separable
upon lithological grounds into Crosswicks Clays and Hazlet
Sands ; outside of New Jersey in Delaware and Maryland
these divisions cannot be recognized with any certainty.
The Crosswicks Clays consist of slate or drab-colored clays
with thin seams and pockets of glauconite, becoming dark,
almost black, locally interstratified with white sand, contain-
ing much lignite and beds of leaves on Raritan Bay. The
lignified trunk of a large tree was found in the clays in this
vicinity, as well as many fragments. Further southward the
clays become brittle, more arenaceous and micaceous and
contain less iron sulphide. The Hazlet Sands are highly
ferruginous, brown in color, with indurated crusts in their
lower layers ; above these there is frequently a well-developed
layer of dark-colored clay, overlain with very micaceous
sands, which are sometimes dark-colored, especially toward
the south where they are also argillaceous.

In his report on surface geology Professor R. D. Salisbury
states * that his assistant, Mr. G. N. Knapp, distinguished five
layers in the clay-marls and traced them across the state.
These he designates Merchantville bed (marly clay), Wood-
bury bed (dove-colored clay), Columbus bed (sand), Mar-
shalltown bed (marly-clay sand), and Wenonah bed (sand).
These features, although more or less marked, are not sharply
defined throughout the entire area of the Matawan, and Pro-
fessor Clark has never attempted to name or map any subdi-
visions other than the lower clay member and the upper sandy
member.

The Matawan is abundantly fossiliferous, especially along

*Ann. Rep. State Gcol. N.J. 1898: 35. 1899.

(49)

Crosswicks and Pensauken creeks. Clark enumerates 86
species of invertebrates, mostly molluscs, and Lewis Wool-
man in his artesian well reports has added several others, as
has also Mr. C. W. Johnson,* who points out their identity
with the Ripley fauna of the Gulf region. Other remains
include sharks' teeth ; Foraminifera, of which 20 species are
recorded by Bagg ; f echinus and other spines ; Ostracoda ;
gavial (?) teeth : dinosaurian bones ; | etc.

The exposure fronting on Raritan Bay near Cliffwood, N.
J., and forming a bluff some thirty feet high northwest of
Matawan creek, has been admirably described by Hollick,§
who records obscure crustacean and molluscan remains,
from which Professor Whitfield identified eight species of
molluscs, and enumerates twenty-six species of plants, of
which ten were new. I have found some few molluscan re-
mains here, occurring in the ferruginous concretions picked
up on the beach, from which Professor Clark has identified
the following : Idonearca vulgaris Morton, Vcleda lintca
Conrad, Cardium sp., Turritclla vertebroidcs Morton and
one or two other species, new to the formation, not yet thor-
oughly studied.

I have nothing to add to the details of the exposure. It is
capped with gravel and in places consists of regularl}'- alter-
nating beds of fine sand several inches thick and seams of
comminuted vegetable matter an inch or two in thickness
(^/. jd). These are replaced by alternating beds of clay
and sand with lignite, and sparingly with greensand. The
face of the bluff is almost entirely hidden as shown in ^l.
55, and while the majority of my plant remains have been
collected from dropped boulders of clay, all have come
from near the base of the exposure except the large cone
{^Sequoia sp.). These plant beds are some distance above
the base of the formation, however, and their preserva-

*New Cretaceous Fossils from an Artesian Well-boring at Mount Laurel,
N.J. Proc. Phila. Acad. 1898: 461-464. 1898.
tU. S. Geol. Surv. Bull. 88, 1898.
t Ann. Rep. State Geol. N. J. 1896 : 248. 1S97.
§ Trans. N. Y. Acad. Sci. 16: 124-136.//. 11-14. 1897.

(so)

tion is due in a large measure to the character of the ma-
terials, as vegetable remains are abundant in the form of lig-
nite, forming thin seams intercalated in the sands nearly to
the top of the exposure. The various layers are not con-
tinuous for any distance along the bluff and evidently indi-
cate an inshore shallow fresh-water deposit which as time
progressed gradually became marine through encroachments
of the sea ; the upper layers of sand with thin seams of com-
minuted vegetable matter indicating changed conditions and
deposits in less quiet waters.

It is quite evident that sufficient material has not yet been
accumulated to warrant an exhaustive discussion of the flora.
I am enabled to enumerate sixty-seven different species of
plants of which fourteen are new ; of these sixty-seven species
some nineteen are of doubtful affinities, such as the various
species of Carfolithus, Arisacma, Podozann'tcs, Phragmiies,
and the various fragments provisionally determined. There
are present, however, in great abundance, such characteristic
mid-Cretaceous forms as Dammara, Cnnninghamites ^ Detval-
guca, Aloriconia^ Salixjlextiosa, Protcoides dafhnogeneoides.
Sassafras acutt'lobum, Laurus j^ltctom'a, Sapindus Morrisoni^
Andromeda Parlatorii^ etc.

The flora has more in common with the middle (Wood-
bridge) stage of the Raritan than with the other layers of
that formation, eleven of the seventeen identical species oc-
curring there, but this is undoubtedly due to the fact that this
horizon is the best known ; the upper Raritan (South Amboy)
layers have not been sufficiently exploited to give us a clear
idea of the vegetation prevalent when they were deposited.
Forty-nine of the Matawan species have not as yet been
found in the Raritan, although two of these are found on
Long or Staten Island in beds probably of Raritan age.
"While this comparison might argue a considerable interval
between the two formations, it remains to be pointed out that
the following ten species are confined to the Raritan of New
Jersey or the Islands and the Matawan formations on this
continent : Chondrites Jlexnosus^ GeinitziaJo7-mosa, Cunning-

(50

hamites elegans, Moriconia cyc/otoxon, Magnolia Woodhridg-
ensts, Laurophylluvi angustifoliuvi, Aralia falmata, Fictis
Woo/soni, Paliiirtis intcgrifoliiis and Cdastro^hylluvi JVeiv-
berryanum, and all except possibly Paliiiriis and Chondrites
with well-characterized remains. Of the numerous species
which are identical with those of the Dakota group of the
West only eight are confined to the Dakota and Matawan
formations. There are twenty-three species identical with
Dakota group forms, but it may be remarked that the latter
horizon is not precisely defined and its flora is exceptionally
well known.

Fifteen of the Matawan species are found in the Raritan
Cretaceous of the islands ; nine occurring on Staten Island,
eight on Long Island, seven on Martha's Vineyard, and four
on Block Island. Eleven of the Matawan species reappear
in the Atane beds of Greenland, and one additional in the
Patoot beds ; of these several are dominant forms of great
vertical or areal distribution, or both, and thus have little sig-
nificance ; such forms are Sequoia Rcichenbachi^ Safindus
Morrisoni^ Laurus -plutonia and Andromeda Parlatorii.
Others are more suggestive ; thus, exclusive of its occurrence
in the clays at Aachen, Moriconia is confined to the ancient
Atlantic coastal plain and Greenland, and its remains are com-
mon ; several species of Magnolia emphasize the similarity
of these floras, as does the presence of the large-leaved Aralia
Ravniana. This species is confined to the Atane and Mata-
wan floras in so far as I can judge from the published descrip-
tions or figures of Aralias, and its remains are unmistakably
characterized. Dczvalquea Grocnlandica is also confined to
these floras. With the Potomac flora as elaborated by Fon-
taine there seems to be no affinity, and the time that elapsed
between them must have been very long.

Of species which occur in the Cenomanian of Europe we
have Geinitzia formosa^ Sequoia Reichenbachiy Cunning-
haviites squajnosus, Cunninghamitcs elcgans, Moriconia cyclo-
toxon, Sassayras acutilobtmi^ Laurus plulonia, Banksia
^usilla, Saf Indus apiculatus and Eucalyptus Geinitzi, a

(52)

total of ten species or over 15 per cent. ; or, on excluding
doubtful species such as those of Banksia, Eucalyptus and
Safindus, and such wide ranging forms as Sequoia Reichen-
bachi, over 9 per cent.

The most striking feature about the Matawan flora is the
entire absence of ferns, which form 5 i^ per cent, of the Raritan
flora, Anemia stricta being commonly found at Woodbridge.
Ferns form i ^ per cent, of the Dakota flora, 1 1 per cent, of the
Atane flora, and about 2 per cent, of the existing New Jersey
flora. In the most recent southern flora with which the Mata-
wan may be compared, that of Alabama,* sixty-two species
of Pteridophytes are listed forming about 2^^ per cent., and
this percentage would be greatly increased if we excluded
herbaceous plants, which as a rule do not occur as fossils.
It is difficult to account for the absence of this order, as the
balance of the flora is proportionally normal, containing nearly
II per cent, of Coniferae against ii-^ per cent, in the Raritan
and 10 per cent, in the Atane beds. Presumably the environ-
ment was unsuited to ferns, although, of course, future dis-
covery may disclose them. Judging by such forms as Dam-
7nara^ Araucarttes, Eucalyptus, Sterculia, Aralia, Myrsine,
Ficus, etc., we may infer that the climate was considerably
warmer than at the present day in this latitude, and at least suf-
ficiently humid to make the absence of ferns remarkable. A
palm {Serejiopsis) occurs at Glen Cove, Long Island,! and
the Raritan furnishes many additional representatives of
genera which are exclusively tropical at the present time, as
for instance Cintiamomum, Bauhinia, etc.

Plants especially abundant in the Raritan formation and
for which we have repeatedly searched in the Matawan are
Thinnfeldia subintegrifolia (Lesq.) Knowlton, Tricarpellites
striatum Newb., and Tricaly cites papyraceus Newb. The
genera Myrica (7 sp.) and Liriodetidron, (4 sp.) which are
abundantly developed in the Raritan, and on the islands, have

* Mohr, Plant Life of Alabama. Contr. U. S. Nat. Herb. v. 6. Jl 1901.
t Dr. Ilollick writes that material recently collected may result in altering
his views as to the botanical affinity of these remains.

{53)

thus far been found wanting. Other Raritan genera which
do not appear in the Matavvan formation are Afe^iis^erfniles,
Dt'ospyros, Cissites, Ilex, Cinnamomum^ Dalbe7-gia, Bau-
htnia, Colutca, Phuiera, Viburnum, Juglans, etc. ; several
of these occur in the upper layers of the Raritan, and future
search ought to disclose some of them in the Matawan.
Cclastrophyllum with abundant remains of ten species in the
Raritan (all horizons) has but two species in the Matawan,
one of these being new and unrelated to any of the known
Raritan species. Widdringtonites is abundant in the Raritan
as are also Salix inaequalis ^.nd Hedera friDiordialis ; Myr-
sine borealis Ileer is one of the commonest leaves at all locali-
ties in the Raritan, as Sequoia hetero^hylla Vel. is one of its
commonest conifers. Celastrus arctica is abundant at South
Amboy, and should extend up into the Matawan. Numerous
specimens of Ophioglosswn granulatiun are also found in the
Raritan according to Newberry (localities not given). The
genus Aralia, so abundantly represented in the Raritan, con-
tinued to develop during Matawan time. We record six
species, the large-leaved Aralia Ravniana emphasizing the
similarity of these Atlantic coastal Cretaceous floras with those
of Greenland. It is of course quite possible, indeed it seems
probable, that these numerous species of Aralia may for the
most part be the variable leaves of a considerably less number
of actual species ; especially is this so of the Raritan species.

This is the extreme northeastern extension of the Matawan,
and the only locality where plant remains have been found,
although the underlying Raritan continues northeastward as
far as Buzzard's Bay and doubtfully on Cape Cod. This
northeastern extension has been much modified by forces
which acted upon it during the Qiiaternary age and is for
the most part entirely covered with drift or totally eroded,
and if the Matawan formerly extended so far north and east
this has been its fate. Professor Lester F. Ward* proposed
the name Island series for the northeastern extension of the
Raritan and makes it the uppermost member of the Potomac,

•Ann. Rep. U. S. Geol. Surv. 15 : 335, 336. 1895.

(54)

unrepresented in New Jersey. Hollick would consider it the
equivalent of the New Jersey Raritan, its distinctive char-
acters being due to morainal action, with which view I entirely
concur.

From the evidence of the flora alone we would consider
the Matawan slightly more recent than the Raritan ; a direct
continuation of the latter, however, with several species added
which are unknown from the Raritan. Thus, aside from the
dissimilar species due to our imperfect knowledge of the flora
of the two formations it remains to be pointed out that the
occurrence of Nelwnho argues a somewhat later age for the
Matawan, as this genus is not found below the Belly River
Cretaceous on this continent.* Time must also have elapsed
for the development or introduction of the various species of
Stercidia which are found here as well as for the changed
species of Aralia. The scarcity (absence) of ferns, the ab-
sence of Brachyphyllum which is essentially a Lower Cre-
taceous genus, and the much larger leaved Moriconia all
point to a somewhat later time than the Raritan.

The Matawan, then, represents the transition period from
the Lower to the Upper Cretaceous, when marine conditions
replaced fresh-water estuarine conditions ; and the flora is
undoubtedly the latest Cretaceous flora of the Atlantic coastal
plain which has been preserved. Professor Ward suggests
that this ancient coastal plain may have extended to Green-
land, but no evidence other than the remarkable similarity of
the floras is known.

Just a word in regard to the remains. In common with the
vast majority of New Jersey Clay specimens, the Matawan
plants were hermetically sealed in the clay and slowly car-
bonized, so that when reexposed to the air, the thick sheet
of lignite dries, becoming cracked, and is soon dissipated,
leaving only a faint impression behind. This has for years
proved an obstacle to the proper investigation of these floras
and it is only with the discovery of leaf-layers carrying con-

*Dr. Hollick has found AV/wwdo on Martha's Vineyard and Long Island,
the latter locality as yet unpublished, which vitiates the above statement.

(55)

ta

3

<J

10

M

m

a

Pt^ fA

^

a

a

nj

Ul

?

01

rt

>

te

OJ

iz;

S

to

•c

a

c-

(U

52^

4J

a

<0

1— 1

c

3

<

O
a
2

t-H

a

.2
'3

Mount V
Rappaha
James Ri

•^.^

jaddfi

aippxH

1 IBSBg

OBraoioj a3itt3M I OBiaoioj^ -ispiO

a,

3 D

Cui

snucH

anuBin-noi^

ifjBiviax

snoaoB^sj^

uiBXd ib;sboo oiiuBpv

OT

O

a.

u

Pi

H

r,

1

u

«

o

B

3

o

•<

(56)

siderably less carbonaceous matter that much progress can be
made. All of my specimens have been sketched immediately,
before becoming dry, so that they are fairly satisfactory ; the
specimens however might almost be thrown away as far as
concerns their value as types.

While usage would sanction the designation of poor speci-
mens of doubtful botanical affinities as " sp." after referring
them provisionally or otherwise to some genus, which prac-
tice is supposed to obviate any undue definiteness on the part
of the describer ; the writer in these notes has followed the
laudable practice of Professor Ward, as quoted above, in be-
lieving that whatever is worth mentioning is worth a name.

Acknowledgment is due Dr. Arthur Hollick, of the New
York Botanical Garden, and Professor W. B. Clark, of Johns
Hopkins University, for material assistance. The specimens
are all deposited at the New York Botanical Garden.

CONIFERAE.
Geinitzia Endl. Syn. Conif. 280. 1847.
This is an entirely extinct genus of the Taxodieae with
several species on both sides of the Atlantic : G. cretacca
Unger (Austria), G. formosa Heer (America and Quedlin-
burg), G. hyfe^'borea (Greenland), G. sp., from the Da-
kota, and G. Jenncyi Font, from the Lower Cretaceous of the
Black Hills. It was founded by Endlicher in his Synopsis
Coniferarum to include certain forms referred by Geinitz to
Sedites and Arancarites and by Corda to Cryftomeria.
Among the former was Arancarites Reichenhachi oi Geinitz,
which Heer in 1868 identified with the living genus Sequoia.
Since that date this plant has been almost uniformly called
Sequoia Reichenhachi, and many place Endlicher's Geinitzia
cretacea under it as a synonym. Others retain the older forms
under Geinitzia. Ward contends that the retention of the
genus Geinitzia logically carries Sequoia Reichenhachi with
it into that genus as the type, while on the other hand the
recognition of Sequoia Reichenhachi logically abolishes the
genus Geinitzia.

(57)

Geinitzia FORMOSA Hccr.

Geinitzia forniosa Heer, Kreidefl. Quedlinburg (Neue
Denkschr. Schvveiz. Ges. 24:) 6. fl. i . /. g; pi. 2.
1871. Newb. Fl. Amboy Clays, 51.//. 9. /. 9. 1896.
(Foliage.) Hollick, Trans. N. Y. Acad. Sci. 16: 129.
pi. 12. f. /, 2. 1897. (Cones.)
Foliage has somewhat the appearance of that of Sequoia
RcichenbacJu\ but the leaves are more crowded. Not col-
lected by me.

Raritan : Woodbridge, N. J. Matawan : Cliffwood, N.
J., foliage not found. Europe: Moletein, Quedlinburg.

Sequoia Endl. Syn. Conif. 197. 1S47.

The genus is unique in that it contains but two dwindling
representatives of its former numerous species, one of which
is the most majestically graceful of trees. These two species
have barely held their own through the vicissitudes of cen-
turies since the glacial period in the little strip of country
where the climate is locally favorable. Many fossil species
have been described, ranging upw^ard from the Upper Jurassic ;
about forty-four from this continent alone, some of them with
a great lateral and vertical range.

Potomac 12, Kootanie 6, Trinity i. Ft. Pierre i, Chey-
enne Sandstone 2, Raritan 7, Island Raritan 2, Dakota 6,
Belly River 3, Montana 4, Vancouver i, Laramie 4, Can-
adian Upper Laramie 3, Lignitic i, Livingston i, Ft. Union
2, Green River 4, Alaskan Eocene 3, Miocene 3, Payette i,
Kome beds of Greenland 5, Atane beds of Greenland 5,
Patoot beds of Greenland 5, Tertiary of Greenland 6, of
which 4 occur in Europe and 2 on the continent of North
America.

Heer records tw^o species from the Tertiary of Siberia, and
Ettingshausen records species from the Tertiaries of eastern
Australia, Tasmania and New Zealand.

Sequoia gracillima (Lesq.) Newb. PI. 48. J". 21 , 22.

Glyptostrobtis gracillimus Lesq. Am. Jour. Sci. IL 46 :
92. 1868 ; Cret. Fl. 52. fl. i. f. 8, 11, iif. 1874; Cret.

(58)

i&Tert. Fl. 32. //././. d,(5<5. 1883. "Cone of Sequoia"
(not described) 111. Cret. & Tert. PI. fl. 11. J. g.
187S. Fl. Dak. Group, z^- 1892.

Sequoia gracillima Newb. Fl. Amboy Clays, 50. -pi. g.
f. 1-3. 1896 ; Later Ext. Fl. 19. ^/. /^. f. 6; fl. 26. f.
p (?).*_ 1898. _

The earliest mention of what is presumably this species is
in a contribution by Newberry, f in which he refers to cones
occurring in the Cretaceous Clays near Keyport (probably
this is the Cliffwood locality) which he referred to Geinitzia^
and associated branches which he referred to Ullmannia
Goepp. In his posthumous Later Extinct Floras, these cones
are referred to Sequoia gracillima; in the Flora of the Am-
boy Clays, which was also issued posthumously, he makes the
same reference and compares them to Heer's Sequoia mac-
rolepis (Fl. Foss. Arct. 7 : 16. fl. 5/. f. ij) considering
them identical.

Hollick describes two cones from Cliffwood as Geinitzia
Jormosa Heer. They are much distorted and incrusted with
pyrites and possibly should be referred to Sequoia gracillima.
Cones of this species are very common, the silicified ones
washing out of the clay on the beach and the lignified ones
occurring in considerable abundance in place in the clays.
I have in my collection the remains of 32 cones, some of
them nearly perfect and but slightly compressed, and I have
refrained from collecting innumerable poor specimens seen.
The largest is 8.5 cm. long, cylindrical, somewhat flattened,
measuring 14 mm. in its shorter diameter and 18 mm. in its
longer diameter. Most of them average nearly this thick-
ness but are somewhat shorter, being about 7 cm. long.

Matawan : Cliffwood, N. J. (foliage not found).

Raritan (?): Keyport, N. J.

Cheyenne Sandstone : Belvidere, Kansas. Dakota Group :
Sioux Cit}^ Iowa. Cretaceous (Dakota?) : Whetstone Creek,
New Mexico.

* Questioned by Hollick, editor of Newberry's work.
tProc. N. Y. Lye. Nat. Hist. 2: 10. 1873.

(50)

Kootanie, British Northwest Terr. (Table of dist., Lesq.
Fl. Dak. Gr., 222.) Forks of Pine River, Northwest Terr.*

Sequoia Reichenbachi (Gein.) Heer. PL ^8. /. ij, 16,

17, 20.

Araticaritcs Reichenbachi Geinitz, Characteristik Schich-

ten und Petrefacten Sachs-bohm. Kreidegebirges, 3 : 98.

fl. 24./. 4. 1842.

Cryftomcria frimaeva Corda ; Reuss, Verstein. Bohm.

kreidef. 89. //. 48./. i-ii. 1846.

? Geinitzia cretacea Endl. Syn. Conif. 281. 1847.

Araucaria Reichenbachi Debey, Entwurf. Geogn.-geoge-

net. Darst. Gegend Aachen 6"^, 64. (Nachtnige) 1849.

Sequoia Reichenbachi Heer, Fl. Foss. Arct. i : 83. -pi. 4J.

f. id, 2b, 5«. 1868; 3: 77, loi, 126. pi. 20. f. 1-8;

fl. 12. f. 7c, d; fl. 28. /. 2. 1875; 4: 50 (Cape

Staratschin, Spitzbergen) ; 6^: 16, S2.fl. 28./. 7. 1882.

Lesq. Cret. Flora, 51. fl. i . f. 10, lob. 1874; ■^^- I^ak.

Group, 35. fl. 2. f. 4. 1892. Dawson, Trans. Roy.

Soc. Can. 21. 1882. Fontaine, Potomac Flora, 243.

//. 118. f. /, 4; fl. iig. /. i-s; fl. 120. f. 7, 8; fl.

122. f. 2 ; fl. 167./. J. 1889. Newberry, Fl. Amboy

Clays, 49.^/. g.f. ig. 1896. Hollick, Trans. N. Y.

Acad. Sci. 12 : 2>^. fl. i . f. 18. 1892; 16: i2d>. fl. 12.

/.3b,s. 1897.

Abietitcs dubius Lesq. Tert. Flora, 81. fl. 6./. 20, 21,

2ia. 1878.
Our specimens from near Cliffwood are possibly related to
the fragments which Hollick refers to Geinitzia formosa
Heer, but I think this is rather unlikely, as his specimens
are much more elongated and more like the cones of Sequoia
gracillima. I feel certain that these cones are those of this
species rather than fragments or immature cones of some
other, for while the specimens preserved are rather frag-
mentary, resembling except in length those cones of Sequoia

* Lesquereux figures (Tert. Fl. pi. 65. /. 5, 5a) fragments of cones from
the Green River group (Tertiary) at Castellos Ranch, Colorado, which are
very similar if not identical with this species.

(6o)

gracillima which have become lignified instead of silici-
fied, I found one nearly perfect cone, which was about 3 cm.
in length by about 2 cm. in diameter, which agreed almost
exactly with Heer's figures of this species.

This was unfortunately smashed in transit, so that I now
have only the recollection of it, which is not apt to be con-
sidered good evidence. However, my first thought on un-
covering it was that it was a cone of S. Reichenbachi. The
foliage of this species is rather common in the clavs at this
point and we would reasonably expect to find the cones ; the
former are very fragmentary. Poorly preserved branchlets of
Cunninghamites squamostis can often be traced for several
inches, but the Sequoia remains are usually not over an inch
in length. The leaves are less closely set than in C. squamo-
sus and longer, often 9 to 10 mm., much more slender and
more spreading.

The best known localities for this species are :

True Laramie and Livingston Beds : Bozeman coal field,
Montana. Montana Formation : Point of Rocks, Wyoming.
Raritan : Woodridge, N. J. Matawan : Clifford, N.J. Belly
River series : Belly River, Canada. Potomac Formation :
Dutch Gap Canal and Fredericksburg, Va. Dakota Group :
Ft. Harker, Kansas. Kootanie : Great Falls, Montana.
Kome Beds : Pattorfik, Avkrusak, Angiarsut, Erkorfat,
Kaersuarsuk. Atane Beds : Unter Atanekerdluk. Europe :
Wernsdorf (Urgonian) Saxony (Cenomanian), Quedlinburg,
Moletein (Senonian), Rainberg bei Salzburg, Brandenberg,
Tyrol, southern France (Turonian), Clays at Aachen, Prussia,
Quadersandstein at Hartz, Bohemia. Cretaceous : Totten-
ville, Staten Island. Lower Cretaceous : Black Hills.

While this species had a wide vertical and areal distribu-
tion ranging from the Upper Jurassic through the Cretaceous,
it is best developed in the Lower Cretaceous.

Sequoia Reichenbachi (Gein) Heer. ? P/. 4-8./. 18.

An oval shaped cone 3 cm. in diameter by 4 cm. long, too
obscure for exact determination. It resembles some of the

(6i)

cones from the Potomac formation which Fontaine refers to
this genus. Is about the same size and character as the cone
of Sequoia Reichcnbachi which Ward figures from the Black
Hills (Ann. Rep. U. S. Geol. Surv. 19': 674. //. 166. f. i).

Araucarites Presl, in Sternb. Vers. 2: 203. 1833.
Araucarites ovatus Hollick.

Araticaritcs ovatus Hollick, Trans. N. Y. Acad. Sci. 16:
128. //. J2. f. ja, 4. 1897.

While these remains are undoubtedly related to the genus
Araucaria, their size would seem to indicate a nearer rela-
tionship to the genus Ai(athis Salisb. {Dammara Lam.). The
only other American post-Jurassic references to Araucarites
are two species of cones from the Potomac which Fontaine
so identifies. The genus Araucaria of Jussieu occurs abun-
dantly from the Jurassic upward ; Fontaine describes three
species from the Potomac formation and Lesquereux a doubt-
ful species from the Dakota Group. Wood of this type has
been identified by Knowlton from the Triassic (?) and the
Lower Cretaceous of South Dakota. All of the foregoing
have very small, more or less imbricated and compressed
leaves, while this species of Hollick's is much larger and
suggests similarity with IVageiopsis, so largely developed in
the Potomac, or various forms referred to Podozainitcs,
Dammara, etc., the exact affinity of which is unknown.

Dammara Lam. Encycl. 2: 259. 1786.
The living species are included in the genus Agathis
Salisb. and are four in number, ranging from the Malayan
Islands and Philippines to Australia and New Zealand.

Dammara Cliffwoodensis Hollick. PL 4.S. f. 8-1 1 .

Dammara {?) Cliffwoodensis Hollick, Trans. N. Y. Acad.
Sci. 16: 128. //. //. /. s-S. 1897.

These problematical remains are very abundant in the
cla3^s at Cliffwood as well as in the Amboy Clays and the Cre-
taceous of Staten Island and Block Island (Hollick). David

(62)

White and HoUick have found them at Martha's Vineyard
and they are known from the Atane and Patoot beds where
Heer differentiates Dammara borealis* D. macrosferma, and
D. microlcfis as well as very similar remains which he de-
scribes as Eucalyptus Geinitzi. These latter remains New-
berry considers are generically the same as those referred to
Davimara and not related to Eucalyptus. However this
may be, undoubted leaf-remains of Eucalyptus have been
found in these various American Cretaceous strata and it
does not seem unreasonable that the fruit should also be
present. It is quite true that various leaves have been re-
ferred to Eucalyptus upon rather doubtful evidence, but
others from both their form and venation, are unquestionably
related to that genus.

The Cliffwood remains are exceedingly common ; often
fragmentary, however, sometimes only a portion of the resin
ducts being preserved ; they are very fragile and crumble
readily upon handling. They vary considerably in size,
some being as small as Damnia?'a borealis and others being
larger than those figured by Hollick from this formation.

PL 4-8^ f. 10 is strikingly like the forms which Heer con-
siders Eucalyptus, but the balance of our collections are
evidently coniferous scales, consisting internally of a rather
central resin-duct enlarged above, with four or five angular
resin-ducts on each side, which seem to descend to the base
of the scale ; externally the scales seem to be more rounded
and finely lined as mf. lo.

Our remains are almost exactly kite-shaped and many of
them seem to have straight ascending sides and are not
abruptly narrowed from above the middle as in Hollick's
specimens (1. c). Neither is there any evidence of the short
mucronate point on the crown ; on the contrary y. lo is evenly
rounded. At the same time it seems best to refer our remains
to Hollick's species, at least until we can be more certain as
to the exact affinity of all these Da7n7nara-\\kQ. remains.

Newberry (/. c.) doubts their relation to Dammara, point-

*This species has been recorded from the Ceuomanian of Bohemia.

(63)

ing out that no Daffimara-Mke foliage has been found asso-
ciated with them and that in the very abundant Amboy Clay
specimens the scales seem to be associated with an extremely
delicate juniper-like conifer ; this association has never been
confirmed, however. He also finds some indications of two
seeds in his specimens, the living Dammara scales being one-
seeded. Merely negative evidence as to the occurrence of
Dammara leaves is not very conclusive, especially in view
of the fact that Lesquereux has described the remains of cer-
tain leaves from the Dakota Group {^Dammar ites)^ which are
undoubtedly related to those of the existing Dammara ro-
btista Moore, of Australia, and various other remains both
in this country and abroad have been referred to Dam-
mar it es.

Furthermore, the remains from Cliffwood which Hollick
describes (/. c.) as Araucarites ovatus are very similar to
those of Agathis Dammara Rich. {^Dammara orientalisYi-d.va. ;
D. alba Rumph.) the existing Dammara of the Malayan
Islands and Philippines.

Fontaine (Potomac Fl. 264. fl. ijj. f. 8-12) describes
wedge-shaped scales under the name Araucarites Aquiensis
from the Potomac Formation near Brooke, Va., where they
are common and always found detached and unassociated
with other remains. While their resemblance to those of
Dajnmara may be considered somewhat far-fetched, their
similar mode of occurrence is suggestive. They are as a
rule larger than Dammara^ but vary considerably in size
and shape and have a transverse furrow on their upper mar-
gin.

CuNNiNGHAMiTES Presl, in Sternb. Vers. 2: 203.
1833.
Pending the discovery of fruit the identification of these
remains with those of the existing genus Cunninghamia is
not beyond question. Cunninghamia R. Br., with a single
species, is at the present day an endemic genus of the China-
Japan region.

( 64 )

CUNNINGHAMITES SQUAMOSUS Hecr. PL 48./. 1 4, IQ.

Cunninghamites squamosus Heer, Beitr. Kreidfl. Qued-
linb. 9. fl. I. f. 5-7. 1872. Hollick, Trans. N. Y.
Acad. Sci. i6 : 129. //. 1 1 . f. 3. 1897. (Not of Hosius
and Von der INIarck.)
Remains of this species are the most abundant coniferous
fossils in the clays at Cliffwood. They occur as twigs about
the size of those figured by Hollick or smaller and demand
no extended discussion. This is the only locality in this
country where they have been found, but additional collec-
tions from the Amboy Clays ought to disclose them. The
two specimens figured are more robust than the majority of
the remains and are ten sevenths of the average size, al-
though several large specimens were collected. On the
usual-sized specimens the leaves are about 6 mm. long, closely
set, stout, incurved, very much crowded in some instances
and quite different in appearance from those of Sequoia
Retchenbacki \m\h. which it is often associated.

CUNNINGHAMITES ELEGANS (Corda) Endl.

Cunninghamia clegans Corda; Reuss, Verstein. Bohm.

Kreidef. gz- P^- 49- f- ^9-31 • 1846.
Cunninghaniites clegans Endl. Syn. Conif. 270. 1847.

Heer, Beitrag. Kreidefl. (Neue. Denkschr. Schweiz.

Gesell.) 12. ^l. j.f. 14. 1869. Schimper, Pal. Veg. 2 :

256. Heer, Fl. Foss. Arct. 7: 17. fl. 53^/. i. 1883.

Newb. Fl. Amboy Clays, 48. fl. 5. f. i-j. 1896.

Hollick, Trans. N. Y. Acad. Sci. 16: 129. fl. 11. f. 2.

1897. Hollick, Bull. N. Y. Bot. Garden, 2: 402. _^/.

41. f. II. 1902. (Knowlton, U. S. Geol. Surv.

Bull. 163 : 29. fl. 5. f. 3. 1900, probably belongs

here.)
Cunninghaniites squamosus Hosius & Von der Marck, Fl.

Westfal. Kreide. 54. //. J7. /. 137-141. (Palaeonto-

graphica, v. 26.) 1880.
Originally described from Moletein in Moravia and Mseno
in Bohemia (Cenomanian), then from the chalk of Westphalia ;

(65)

Heer records it from the Patool beds of Greenland. New-
berry's specimens are from " near Key port "' and are prob-
ably not from the Raritan, however in a table on page 135 he
gives as an additional locality South Amboy, which is within
the Raritan formation. Hollick (/. c.) records unmistakable
remains of this species from the Matawan, but much search
has not resulted in my finding it except one specimen which
is doubtfully referred to this species (too poor to figure).

MoRicoNiA Deb. & Ett. Urweltl. Acrobryen Aachen,

59- 1859.

MoRicoNiA cYCLOToxoN Deb. & Ett. P/. 43./. 4; 48./.

1-4.

Moriconia cyclotoxon Deb. & Ett. Urweltl. Acrobryen
Aachen (Denkschr. Wien. Akad. 17: 239), 59. ^l. y.f.
23-27. 1859. Newb. Fl. Amboy Clays, $$. pi. 10. f.
11-21. 1896. Heer, Fl. Foss. Arct. 3^: 97. fl. 26.
f. 18, under the name of Pccopteris Kudlisctcnsis ; 6^:
¥)'i>l'33'f' 1-9 '^ 7: ^^-P^'SJ'f' 10,10b; pl.34./. ic
(the latter ligure probably represents a B r achy phy Hum).
Originally described from the clays at Aachen, Heer found
it in the Atane and Patoot beds of Greenland and Disco
Island, and Newberry in the Amboy Clays at South Amboy,
N. J., where it is common. Heer's forms have the stem
naked in a majority of cases ; the branchlets are about 21 mm.
long and the widest is 4.5 mm. wide ; one branchlet with the
tip missing is still 36 mm, long but only 3 mm. wide. New-
berry's Amboy Clay specimens have some of the branchlets
long and slender like the Arctic forms, but the majority are
shorter and stouter, being 10 to 12 mm. in length by 4 mm.
in width, and the stems are more uniformly leaved. Speci-
mens from Staten and Block Islands recently reported by
Hollick are also small.

All of my specimens from Cliffwood have the main stem
leaved ; my only complete branchlet is 34 mm. long by 9 mm.
wide, in fact all of my specimens are nearly, or quite, twice
as wide as any of the Amboy Clay or Greenland forms.

{66)

The figure (pi. 4.3- /> 4) shows the appearance of the main
stem of a fragmentary specimen which might readily enough
be taken for the pinna of a fern.

The markings on all the specimens are very obscure and
it is only after the carbonized layer has dried out and blown
away that they show plainly the leaf-markings as shown in
the balance of the figures. No fruit has anywhere been found
associated with these twigs, so that their exact relationship
remains to be determined. Judging from the foliage alone
Heer is inclined to place it among the Cupressineae and near
to Libocedrus.

Lihocedrtis Endl. is unknown from the American Creta-
ceous or later formations, although tjie existing incense
cedar, Libocedrus dccitrrens Torr., ranges from Oregon
southward to southern California and is commonly cultivated.
This typically northern genus reaches Australia through the
East Indian region and penetrates far into South America
along the Andes, thus almost surrounding the Pacific. Heer
has described three fossil species from the Arctic regions,
Libocedrus gracilis from Spitzbergen, Libocedrus cretacca
from the Atane schists (Kardlok, Isunguak), and Libocedrus
Sabiniana from Greenland (Atanekerdluk B, Naujat, Kug-
sinek, Haseninsel, Isunguak) and Spitzbergen.

ARACEAE.

Arisaema Martius, Flora, 14: 459. 1831.
There are about fifty existing species, mostly of temperate
and tropical Asia ; three in eastern North America. Two
fossil species have been provisionally referred here as fol-
lows :

Arisaema cretaceum Lesq. PL 4.6. f. 4.

Arisaema cretacca Lesq. Fl. Dak. Group, 38. fl. 46. f. i.

1892.
Arisaema (P) dubia Hollick, Trans. N. Y. Acad. Sci. 16 :

130.//. 12. y. 6. 1897.
The above species was founded by Lesquereux for a

(67)

monocotyledonous, probably Araceous spathe from the Da-
kota Group of Kansas. In all probability Ilollick's specimen
is of the same species. It is doubtfully a species of Arisacnm,
however, and might equally be a cycadaceous spathe. In
appearance the specimen before us is very similar to some of
Lesquereux's figures of Dammaritcs (Fl. Dak. Gr. fl. i. f,
g-ii) ; the fine lining is about .5 mm. apart as in the existing
Diwimara robtista Moore of Australia, but the texture is very
thin and quite the opposite of the thick coriaceous leaves of
Dammar a and Dammaritcs.

Arisaema (?) Mattewanense HoUick.

Arisacjna Mattewanense Hollick, Trans. N. Y. Acad. Sci.
16: 130.//. 12./. 7. 1897.

Provisionally so referred by Dr. Hollick, as the fruit of
some Araceous plant.

SALICACEAE.

Salix Linn. Sp. PI. 1015. 1753.

The willows are all extremely rapid growers and thrive in
the wettest soil ; they are thus apt to occur in localities favor-
able for fossilization. There are about 160 existing species
widely distributed throughout the northern hemisphere and
arctic zone, a few in the southern hemisphere ; about 80 are
American. There are about 46 fossil American species dis-
tributed as follows: Raritan 5, Island Raritan 4, Dakota 11,
Woodbine i, Montana 3, Vancouver 2, Laramie 4 ( ?), Ft.
Union i, Green River 5, Eocene 7 (?), Eolignitic 3, Tertiary
2, Miocene 5, Pleistocene i (?).

Heer records three from the Island of Sachalin and seven
from the Tertiary of Greenland.

Salix proteaefolia flexuosa (Newb.) Lesq. PI. 4S./.
12; pi. 52./. 2.

Salix jlexuosa Newb. Ann. Lye. N. Y. 9: 21. 1868; 111.
Cret. & Tert. PL //. /./. 4. 1878; Later Ext. Fl. 56.
fl. 2./. 4; -pi. rj./. J, 4; //. 14./. I. 1S98.

(68)

Saltx protcaefolia Jlcxiiosa Lesq. Fl. Dak. Group, 50. />/.
44,/. 4, J. 1892.

Dakota Group : Kansas. Cretaceous : Seacliff, Long Is-
land (Hollick, '94) and Block Island (Hollick, '98).

The smaller leaf is similar to the smaller forms referred to
the above species. It resembles a number of small lanceolate
leaves of varied affinities such as Lesquereux's Aiidrotneda
affints, the smaller forms of Myrica longa Heer from the Da-
kota (but has a narrower base), and Laurtis angtista Heer
as figured in Fl. Foss. Arct. 7: fl. 57./. ib; the latter is,
however, considerably smaller than Heer's figures of this same
species in the same work, v. 6', and also much smaller than
the leaf which Lesquereux refers to this species in the Flora
of the Dakota Group.

Were our specimen [fl. 48./. 12) somewhat more linear it
might be compared to Eucalyptus Dakotensis, but there is
no doubt that it is a Salix. The larger leaves [pi. 52. /. 2)
show the characteristic venation of this species.

Salix Mattewancnsis sp. nov. Pi. 51. f. 5.

A small ovate-lanceolate leaf with an obtuse base and
slender tapering tip, greatly resembling several modern
willow leaves ; secondaries regular, camptodrome.

Except for its small size it is very similar to Salix mem-
hranacea Newb. There is considerable resemblance to
Salix sp. (Fl. Amboy Clays, pi. 42. f. 6-8) only the tip is
more elongated. There is also some resemblance to such
leaves as Leguminosites constrictus Lesq. and to several of
the forms referred to Cassia.

Salix Meekii Newb.

Salix Meekii Newb. Ann. Lye. N. Y. 9: 19. 1868; Later

Ext. Fl. 58. pi. 2. f. 3. 1898. Hollick, Trans. N. Y.

Acad. Sci. 16: 130.//. /j.y. j, 4. 1897.

S. cuncata Newb. 111. Cret. & Tert. pi. i.f. 2, 3. 1878.

S. proteaefolia lanceolata Lesq. Fl. Dak. Group, 50. pi.

64./. 6-8. 1892.
Recorded by Hollick from the Matawan formation near
Cliff wood, N. J. ; not found by me.

(69)

Poi'ULiTKs Lesq. Am. Jour. Sci. 46: 93. 1868.

Founded by Lesquereiix to include leaves from the Dakota
Group which are apparently related to Populus but differing
in their generally entire margin, cordate outline, and cras-
pedodrome venation, the latter character apparently wanting
in our species. Some of Lesquereux's species have since been
transferred to the genera Grezut'ofsis, Hamamclites, Mcnis-
pennites, and C/ssitcs, leaving seven Dakota species and one
species from the Upper Cretaceous of Vancouver Island.

The genus Populus, although containing the oldest known
dicotyledon at Kome, besides ten Dakota species and two
Potomac species according to Fontaine, is so essentially a later
genus that I prefer to include our leaf in the related genus
Populites, thus obviating too great definiteness of relation to
the existing genus.

Populus contains about twenty-five species in the existing
flora, all of which are confined to the northern hemisphere.
Some twelve of these inhabit North America. The fossil
species are numerous.

Populites tenuifolius sp. nov. PL ^g. f. 7.

A leaf exceeding 10 cm. in length and nearly 12 cm. in
width : margin in the upper part apparently entire or perhaps
a trifle undulate ; about 4 cm. of the right lateral margin is
preserved and seems to be slightly crenate, but the indica-
tions are very faint and may be due to the wearing away of
the material. The base is not preserved, but I judge it to
have been cordate. The texture is extremely thin for so
large a leaf. Secondaries four or five on each side, very
thin, alternate, unbranched except the basal ones, leaving
the midrib at an angle of about 45^ and curving upward,
branching near the margin much as in ProtophyUum Stern-
bergii;* basal secondaries evidently much longer than the
others, giving off numerous branches to the latero-basal
portion of the leaf. Areolation ill-defined, angular.

I have been at a loss to correctly determine this leaf ; it
bears considerable resemblance to some of Lesquereux's
species of Pretophylluni, but inasmuch as the latter is a syn-

* Lesq. Fl. Dak. Group, pi. 42. f. /.

(70)

thetic type of uncertain botanical affinities, with mostly cori-
aceous leaves, it has not been considered available. The
only Raritan leaf with which our specimen may be com-
pared is Tiliaephyllum duhium* which it resembles in its
cordate outline, delicate venation, and thin texture ; like the
Raritan leaf ours is represented by but a single specimen
rather poorly preserved. It differs in being larger and in
lacking the dentate margin, and was apparently equilateral
with a straight midrib. In view of the uncertainty of New-
berry's determination it is desirable that we should endeavor
to get an idea of the true botanical affinity of our leaf, which
it seems to me will place it among those leaves ancestral to
the modern aspens or poplars. It resembles several species
of Pofuhis, although the latter are as a rule coriaceous or
subcoriaceous, for instance Pofidus Harheriana Lesq. from
Staten Island f and the Dakota Group. $ It may be com-
pared with Pofuhis balsamoides (}) var. latifolia Lesq.,§
although the latter is a Tertiary species ; the margin is ap-
parently similar and the venation is strikingly similar except
at the margin.

It may also be compared with the Dakota species Pofulites
Lancastricnsis Lesq. || which it greatly resembles in size and
outline ; the secondaries are stouter and straighter in the latter
and the basal one is less branched.

FAGACEAE.

QuERCus Linn. Sp. PL 994. 1753.

About two hundred existing species of the northern hemi-
sphere,H more than fifty of which occur in North America.
The extinct American species number about 127, distributed

*Newb. Fl. Amboy Clays, 109. /»/. 15./. 13.
tHollick, Ann. N. Y. Acad. Sci. 11 : 419. pl- 36./. 8.
X Lesq. Fl. Dak. Group, 44. pl. 46./. 4.
? Lesq. Cret. & Tert. Fl. 158. //. 3^-f-4-
lILesq. Cret. Fl. 58. />/. J./ /•

Tl Recorded by Ettingshausen from Tertiary of New Zealand. (Trans.
N. Z. Inst. V. 23.)

(71)

as follows: Raritan 2, Dakota 20, Montana 2, Vancouver 6,
Laramie 17, Livingston 3, Denver 9, Ft. Union 7, Tertiary
of Yellowstone Park 7, Green River 7, Eolignitic 6, Eocene
12, Miocene 25, Payette 4, Pleistocene 6, Glacial i, Atane
7, Patoot 7, Tertiary of Greenland 15.

Quercus HoUickii sp. nov. PI. 57. f. /, 2.

Leaves subcoriaceous, obovate, obtuse, gradually narrow-
ing to the base; secondaries strong, equidistant and parallel,
alternate, camptodrome, angle of divergence about 50*^ ;
basal one third of the margin entire, above rather irregularly
dentate.

These specimens evidently represent a leaf about 9 cm. in
length by about 4.5 cm. in greatest width.

I have been unable to refer this to any of the described
species of ^lercus although it resembles several ; in outline
it is similar to ^. Wardiana Lesq.* from the Dakota group,
but the latter is 50 per cent, larger and with different vena-
tion. There is also a resemblance to Newberry's ^ cllip-
tica f but the secondaries are straighter, more ascending and
more regularly arched in our specimen. Again, it may be
compared to the Tertiary ^ Olafseni Heer, but the latter
has the leaves more oval, margin more dentate, secondaries
craspedodrome and straighter, greatly resembling Les-
quereux's ^l. 48. f. 4 from the Fort Union Beds ; in the
latter, however, some of the upper secondaries run directly
to the dentate points of the margin. There is, further, a re-
semblance to J^. Nevadensis Lesq., from the Pacific Coast
Miocene, which however has the secondaries straighter and
subcraspedodrome.

Considerable similarity is to be noted with various leaves
referred to Celastrophyllum, as for instance C. grandifolium
Newb., which is common in the Raritan (localities not given) ;
our leaf is considerably shorter and relatively wider, more
obovate in outline, with a more tapering base, straighter mid-
rib and more regular secondaries.

*F1. Dak. Group. />/. 7./. /. 1S52.
t Later Ext. Fl. pi. 20. f. 3. 1S98.

(72)

QuERCus HoLMESii Lcsq. PL 4.8. f. IJ.

Dryofhyllmn {^lercus) salictfolitmi Lesq. Ann. Rep. U

S. Geol. & Geog. Surv. Terr. 1874: 340.//. 8.f. 2.

1876. Name preoccupied by ^lerctis salicifolia Newb.

Dryophyllum (^lercus) Hohnesii Lesq. Cret. & Tert.

Flora, 38. _?i/. ^.y: <?. 1883.
^lercus Hohnesii lu^sQ^. Fl. Dak. Group, 58. 1892.
Upper Cretaceous : Port McNeill, Vancouver Island.
Dakota Group ?: San Juan River, S. W. Colorado.

There are a number of small-leaved oaks which resemble
this species, including ^uc?'cus Montanensis Knowlton (Fl.
Montana Form. ^/. ii.f. i6),Dryophyllumsubfalcatuni\^Q.^c^.
{D. Briineri Ward) of the Laramie and the smaller forms
withrnearly simple margins of Newberry's Tertiary ^lercus
consimilis .

QjjERCus MoRRisoNiANA Lesq.

^lercus Morrisoniana Lesq. Cret. & Tert. Fl. 40. -pL
IJ. f. /, 2. 1883. Hollick, Trans. N. Y. Acad. Sci.
16: 131.//. 13./. II, 12. 1897.
Recorded by Hollick from the Matawan formation near
Cliff wood, N. J. ; not found by me.

Q^jERCus (?) Novae-Caesareae Hollick. PI. 51 . f. 4.

^lercus (?) Novae- Caesar cae Hollick, Trans. N. Y. Acad.
Sci. 16: 131.//. /J.y. p, 10. 1897.

Hollick points out the resemblance of these leaves to those
of ^lercus Myrtillus Heer from Greenland and also to those
of Diosfyros provecta Velen. from Bohemia.

I have found several fragments at Cliffwood which belong
here.

QuERCus sp. '^X.ri. ) ^.'i t 1 )

parous (?) s/., Hollick, Trans. N. Y. Acad. Sci. 16:
131.//. 14./. g. 1897.

Of doubtful affinity. Hollick compares this fragment with
^lercus -poranioides Lesq. and with J^. Thidensis Heer, and

(73)

also suggests that it might be a fragment of Ilex Masoni
Lesq. There are several similar fragments in my collection.

MORACEAE.
Ficus Linn. Sp. PL 1059. 1753.

The existing species number some six hundred shrubs and
trees of the warmer parts of the globe everywhere, chiefly in
Asia, Africa and the East Indian Islands. Two species
enter the limits of the United States, occurring in Florida
and the West Indies. About one hundred species of Ficics
occur in America in the following formations : Potomac 2,
Raritan 3, Island Raritan 5, Dakota 23, Woodbine i, Mon-
tana 16, Vancouver 6, Laramie 21, Livingston i, Denver 9,
Ft. Union 5, Green River 5, Tertiary of Yellowstone Park
4, Eocene 6, Eolignitic 6, Miocene 7, Atane 3, Patoot 2,
Greenland Tertiary i.

Recorded by Ettingshausen from Tertiary of New Zealand
(Trans. N. Z. Inst. v. 23) ; now extinct there, though living
in Australia.

Ficus reticulata (Lesq.) Knowlton. PL 52. f. y, pL jj.
/. /, 4..

Laurophylhmi reticulatum Lesq. Ann. Rep. U. S. Geol.
& Geog. Surv. Terr. 1872: 425. 1873; Cret. Flora,

^e.fl. is.f. 4^5' 1874.
Ficus laurophylla Lesq. Ann. Rep. U. S. Geol. i& Geog.
Surv. Terr. 1874: 342. fl. 5./. 7. 1876. Cret. &
Tert. Fl. 49.//. /./. 12, 13. 1878; Fl. Dak. Group,
85. 1892.
Ficus reticulata Knowlton, Bull. U. S. Geol. Surv. 152 :

104. 1898.
Heretofore known from the Dakota Group of Kansas.
The Cliffwood remains consist of the impressions of the
under surface of leaves rather fragmentary, and indicate an
oblong-lanceolate leaf 2-3.5 ^m. in width, with an entire
margin and rather stout midrib ; secondaries branch at an
obtuse angle, unequal, becoming marginal ; venation some-

(74)

what irreiiular. Tliere is some resemblance to the Greenland
leaves which Heer (Fl. Foss. Arct. 7: fl. 7g./. 4) refers to
Diospyros hrachysepala A. Br.

Ficus WooLsoNi Newb. PL 47' f- 7-

Ficus Woolsoni Newb. ; Hollick, Trans. N. Y. Acad. Sci.
12 : 6. fl. 2. f. /, 2c. 1892 ; Newb. Fl. Amboy Clays,
70. fl. 20. f. J ; fl. 2j./. 1-6. 1896.
This species is thus far only represented by a small frag-
ment from Cliffwood, although common in the Raritan Clays
at Sayreville and Woodbridge, and also reported from Kreis-
cherville, Staten Island, by Hollick.

PROTEACEAE.
Proteoides Heer, Phyll. Cret. Nebr. 17. 1866.
This extinct genus, which is supposed to be allied to the
existing genus Protca, has some eight fossil American species
as follows: Raritan 2, Dakota 4, Mill Creek i, Vancouver
3. Many fossil species of Proteaceae have been described
from the European Tertiary but considerable doubt has been
expressed as to the real affinity of many of them, their re-
semblance to various species of Coniferae, Meliaceae, Sapin-
daceae and Myricaceae being pointed out.

Proteoides daphnogenoides Heer. PI. 51 . f. 6-g.

P?'oteotdes dafhnogenoides Heer, Phyll. Cret. Nebr. 17.
fl. 4./. ^, 10. 1866. Newb. Fl. Amboy Clays, 72.^/.
77./. 8, g; fl. j2. /. II, ij, 14; fl. 33. f. 3; fl. 41.
f. 15. 1896. Lesq. Cret. Fl. 85.//. 15. f. /, 2. 1874.
HolHck, Trans. N. Y. Acad. Sci. 11 : 98. //. j./. /, 2.
1892; 12: Z^. fl. 2./. 4y p, 13. 1893; Bull. Torrey
Club, 21: 52. fl. lyy. f. i. 1894; Ann. N. Y. Acad.
Sci, II : 420. fl. 36. f. 1-3. 1898.
Remains of this species are common in the clays near Cliff-
wood, but very fragmentary. Previously recorded from the
Dakota group at Decatur, Nebraska ; Raritan at Wood-
bridge, Sayreville, etc., N. J. ; the Mill Creek series at Mill
Creek and the Cretaceous on Staten Island and Long Island.

(75)

Banksia Linn. f. Suppl. 15. 1781.
The existing species number about fifty and are confined
to the Australian region. Two fossil species have been iden-
tified from American strata, but whether or'not they are al-
lied to the living Banksiae is not altogether certain.

Banksia pusilla Velen.

Banksia fiisilla Velen. Fl. Boehm. Kreidef. 7 (32).//. /
(9)./. 14-17. 1883. Hollick, Trans. N. Y. Acad. Sci.
16: 132.//. 13' f' 7' 1897-

With the exception of Banksia Helvetica Keer, which
Lesquereux records from the Eolignitic of Mississippi, this
genus is not found elsewhere on this continent, although spe-
cies referred to this and the allied genus Banksitcs occur in
the European Tertiary. It is essentially a later genus, with
upwards of fifty existing species which are all confined to
the Australian region. The above species is very similar to
Santalum Amcricanum Lesq. (Cret. & Tert. Fl. pi. J2. f. 7)
of the western Tertiary.

NYMPHAEACEAE.
Nelumbo Adans. Fam. PI. 2: 76. 1763.

But two living species are known, JV. Nelumbo (L.) Karst.
of eastern Asia and N. lutea (Willd.) Pers. of eastern North
America, giving emphasis to the well-known similarity of
these two floras. The genus appeared in the middle Creta-
ceous and ranges to the Miocene Tertiary, increasing regu-
larly in size. There are one Asiatic, seven European and
nine American fossil species,* all of the American species,
unless it be Heer's from Atane, being from a considerably
higher horizon than our Cliffwood specimen.

Exceptions to the latter statement are unpublished species
from Long Island and Martha's Vineyard, discovered bv Dr.
Hollick.

Nelumbo primaeva sp. nov. PI. 43./. /.

This is undoubtedly a portion of a leaf of Nelumbo — too

*The living and fossil species are enumerated by Hollick in Bull. Torrey
Club, 21 : 307. 1894.

(76)

small a fragment, however, for rigid determination. It rep-
resents a small-leaved species with leaves apparently about
the same size as those of JV^eltimbo Laraniiensis Hollick (/. c.
f. in text) ; the cross venation is obliterated, however, and the
primary veins are only eight in number instead of twelve, as
is the case in N. Laramiensis.*' Nelumho inteTmedia,\
of the Montana formation, is of about the same size, but has
twelve or thirteen weak primary veins. Nclumbo Dazvsoni
Hollick, % from the Canadian Belly River, is also a small
leaf, but has eighteen primary veins. Dawson has de-
scribed,§ but not figured, a small leaf from the Canadian
Laramie, under the name of Nelumhium Saskachtiensis, of
similar size and with only seven primaries.

MAGNOLIACEAE.
Magnolia Linn. Sp. PI. 535. 1753.
There are about fifteen existing species of Magnolia, con-
fined to eastern North America, eastern Asia and the Hima-
layan region. The fossil species are numerous, there being
over forty from continental America besides those from
Greenland. The distribution of the American fossil species
is as follows : Raritan 7, Island Raritan 8, Dakota 11, Atane
4, Mill Creek i. Woodbine 2, Montana 2, Vancouver 2,
Laramie 7, Denver i, Ft. Union 2, Eocene 6, Eolignitic 5,
Miocene 5, Greenland Tertiary 6.

Magnolia obtusata Heer. PL 47./. 4.

Magnolia obtusata Heer, Fl. Foss. Arct. 6^: 90. pi. /j.

/. 12; fl. 21. f. 3. 1882. Lesq. Fl. Dak. Group,

201. fl. 60. f. 5, 6. 1892.

This leaf resembles greatly the smaller of the two figures

which are figured respectively by each of the above authors.

* Dr. Hollick, who has seen a figure of this leaf, writes : " Very close if
not identical with Neluinbo Laratnieusis."

tKnowlton, Fl. Montana Form. (Bull. U. S. Geol. Surv. 163:) 53. pi,

13- /■ 3-5' 1900-

t Brasenia attiiqua, Daws. Trans. Roy. Soc. Cau. 3: sec. 4, 15,/. 1886.
§ Daws, ibid., 5 : sec. 4, 35. 1888.

(77)

Previously recorded from the Atane schists of Greenland
and the Dakota group of Kansas.

Magnolia Woodbridgensis Hollick. PL jj. f. 5; fl.

57' /• 2.

Magnolia Woodbridgensis Hollick, in Newb. Fl. Amboy
Clays, 74. //. 36. f. II ; pi. 57. /. 3-7. 1896. Hol-
lick, Trans. N. Y. Acad. Sci. 16: 133. pi. 14./. 8.
1897 ; Ann. N. Y. Acad. Sci. ii : 60. pi. 3./. 2. 1898.

While the fragments shown on plate 33 have the venation
entirely obscured, I have no hesitation in referring them to
the above species, especially as I have found other smaller
fragments with the characteristic venation of Magnolia.

Judging from the nearly parallel lateral margins of the
fragment shown on plate 37 it would seem to indicate a
rather longer leaf than the typical Magnolia Woodridgensis,
a leaf more like Magnolia Boulayana Lesq. or Magnolia
glaucoides Newb. The incomplete nature of the specimen,
however, and the total obliteration of the venation make it
preferable to place it under M. Woodbridgensis. Leaf-blade
apparently quite thick.

Magnolia tenuifolia Lesq. PI. 47./. 10.

Magnolia tenuifolia Lesq. Am. Jour. Sci. 46 : 100.
1868; Cret. Flora, 92. pi. 21. f. i. 1874; Fl. Dak.
Group, i^S. pi. 24./. I. 1892.

Previously known from the Dakota Group at Decatur,
Neb., Kansas, and Peace River, Northwest Territory.

Our fragment is very similar to /. i. pi. 24, Fl. Dak.
Group, of this species, except that the secondaries are some-
what straighter and more regular.

There is considerable resemblance to Magnolia Capellinii
Heer, which is doubtfully recorded from Sayreville, N. J.,
Glen Cove, L. L, and Martha's Vineyard, as well as from
the Dakota of the West and from Greenland.

Ours, while only a fragment, denotes a more elongated
leaf than that of M. Capellinii.

(78)

LAURACEAE.
Laurus Linn. Sp. PI. 369. 1753.
As commonly restricted, the existing species are but two,
of southern Europe, the Canary Islands and Madeira. The
family is large and chiefly tropical. The fossil species are
numerous, over thirty being found on this continent. It may
be remarked, however, that there is considerable uncertainty
in the generic diagnoses which are based upon nothing but
leaf remains of the Laurineae. Remains referred to Laurus
are common in the European Tertiary and Heer records one
species from the Tertiary of Siberia, besides four at Atane,
three at Patoot, and four in the Tertiary of Greenland. The
American species are distributed as follows: Raritan 3,
Island Raritan 4, Dakota 11, Mill Creek i, Woodbine i,
Montana 3, Vancouver i, Cret. of N. W. Terr, i, Laramie
4, Livingston i, Denver 3, Ft. Union 2, Tertiary of Yellow-
stone Park 2, Eocene i, Eolignitic 4, Miocene 7.

Laurus Hollae Heer. PL 30./. 7, 8 ; fl. 52./. 7, 8.
Laurus Hollae Heer, Fl. Foss. Arct. 6^: 76. pi. 33- /•

1882. Lesq. Fl. Dak. Group,
Hollick, Trans. N. Y. Acad.
1892.

Previously known from the Dakota Group, Kansas ; Cre-
taceous, Staten Island ; Atane schists, Greenland. Ours are
rather poor specimens for positive determination ; as far as
they go they agree admirably with the above species.

Laurus proteaefolia Lesq. PI. 47'/' <)', fl- 49- /■ ^•
Laurus -proteaefolia Lesq. Bull. U. S. Geol. and Geog.
Surv. Terr, i: 393. 1876; Ann. Rep. ibid. 1874:
342. pi. 3. /' /, 2. 1876; Cret. & Tert. Fl. 52. _^/. j.
/, g,io;pl' 16. f. 6. 1883; Fl. Dak. Group, 92. 1892.
Lesquereux's specimens were from the Dakota Group at
Morrison, Col., and Fort Harker, Kansas. The Cliffwood
forms which Hollick refers to Laurus pltitonia Heer are some-
what larger than the majority of Heer's figures of that
species, and our remains which seem to be species of Laurus

13 /

44.

/5h;

45' /• 3-

92.

pi.

12. f.

8. 1892.

Sci

. 12

■ 1-p-

2./. 17.

(79)

are larger still and more ovate in form, with less ascending
secondaries. They are intermediate in size among Lesquer-
eux's figures of L. frotcacfolia.

Laurus plutonia Ileer. PL 50. f. g-ii.

Laurus plutonia Heer, Fl. Foss. Arct. 6^: 75. ^l. /p. f.
id, 2-4; -pi. 20. f. ja, 4-6; -pi. 24.. f. 6b; fl. 28. /.
10, 11; fl. 42. f. 4b; 7: 30. fl. 58. f. 2; fl. 62.
f. Id. Lesq. Fl. Dak. Group, 91. fl. ij. f. j, 6 ; fl.
22. f. j. 1892. Nevvb. Fl. Amboy Clays, 85. fl. 16.
f. 10, II. 1896, Hollick, Trans. N. Y. Acad. Sci.
12: 236. fl. 6. f. I. 1893; Bull. Geol. Soc. Am. 7:
13. 1895 ; Trans. N. Y. Acad. Sci. 16: it,2. fl. ij. f.

5, 6. 1897 ; Ann. N. Y. Acad. Sci. 11 : 60. fl. 4. /.

6, 7. 1898.

Recorded from the Raritan, locality not given ; Mataw^an,
Cliff wood, N. J. ; Dakota group, Kansas and Minn. (?) ; Cre-
taceous, Glen Cove, L. I., Martha's Vineyard and Block
Island; Middle Cretaceous at Atane and Patoot, Greenland.

This is another species which was very common through-
out eastern North America from Greenland to New Jersey
during the Middle Cretaceous. The leaves were rather vari-
able in outline, Newberry's Amboy Clay forms and those
from Cliff wood which Hollick refers to this species being con-
siderably broader than the majority of Heer's specimens.
The leaves from the Dakota which Lesquereux identifies
with this species are, on the other hand, smaller and agree
fairly well with the leaves in our collections which seem to
belong to this species. The latter agree closely with Heer's
figures and differ considerably from Hollick's forms from
this formation. About the same size as the Block Island
leaf which Hollick refers to this species.

Laurus Hollickii sp. nov. PI. j2. f. 4.

A small lanceolate Lauraceous leaf about 8 cm. long and
1.5 cm. in greatest width, which is at a point about half way
between the apex and base, the blade tapering about equally
in both directions ; secondaries four or five on each side,

(So)

leaving the midrib at an angle somewhat less than 45° and
curving upward with a long sweep, becoming nearly parallel
with the margin, along which they form small shallow arches,
the connecting branches to the secondary next above being
small and transverse.

Species dedicated to Dr. Arthur Hollick, who suggested its
Lauraceous affinity.

This species is quite similar to Daphnophylluni Dakotense
Lesq. from the Dakota group, which Lesquereux compares
with Daphne -protogaea Ett.* Our leaf is somewhat nar-
rower and the distal portions of the secondaries are straighter.
There is also considerable resemblance in outline to various
species of Salix and some of the lanceolate species of Ficus.

Laurophyllum Gopp. Tertiarfl. Java, 45. 1854.
An entirely extinct genus containing leaves supposed to be
allied to Laurus and including five American species in the
Raritan, Mill Creek and Upper Cretaceous.

Laurophyllum angustifolium Newb. PL 47./. 1,3,8,

//. 4g.f. 1-5.

Laurophyllum angustifolium Newb. Fl. Amboy Clays, 86.
fl. 17' f' 10, II. 1896.

Remains of leaves, which by their narrow elongated shape,
thick midrib, and coriaceous texture are allied to this species
of Newberry from the Amboy Clays at Woodbridge are very
common at Cliffwood. The remains are very poor, how-
ever, consisting of larger or smaller fragments of the basal
portions of leaves. The petiole (preserved in f. 5) is short
and stout.

Newberry compares his leaves with some of Heer's Myrica
longa, but the latter lack the straight narrowly ascending
basal margins and are rather abruptly rounded. If Arctic
analogues are necessary we may point out the resemblance
to the Tertiary Laurtis Rcussii Ett. or to forms of Laurus
angusta Heer from Atane (Fl. Foss. Arct. 6^: fl. 43,/. ic).

Fl. Bilin. pt. 2, 13. pi. 34. /• f-3- 1868.

(8i)

Sassafras Nees & Eberm. Handb. Med. Pharm. Bot. 2:
418. 1831.

At the present day a monotypic genus of eastern North
America, although inhabiting Europe before the Glacial
period in both the Tertiary and Cretaceous ages. Numer-
ous fossil leaves have been referred to this genus ranging
from the Potomac formation upward. While some of these
are undoubtedly ancestral Sassafras species, others are just
as undoubtedly not related to Sassa/ras ; for instance we
would refer Sassajras Harkerianum, S. obtustim and S.
cretacciim ohlusuni to Cissiles ; we would refer S. disscctmn,
S. dissectum sy^nmetricum^ S. tnirabile, S. paptllostini, S.
recurvatuni, S. dcntatum and S. grossedcntatum to Platanus
or possibly to Protoplatamis. The two dozen or more an-
cient American species occur in the following formations :
Potomac 3, Cheyenne Sandstone 3, Raritan 4, Island Rari-
tan 2, Matawan i, Dakota 14, Vancouver i, Canadian Up-
per Laramie 2.

Heer describes two species from Atane, one of which is
identical with a Dakota form ; one species from Patoot ; and
one from the Tertiary of Greenland. Velenovsky identifies S.
acutilolmm in the Cenomanian of Bohemia, and several species
have been described from the European Tertiary formations.

Sassafras acutilobum Lesq. PL 4S'f- ^1 -•

Sassafras acutilobum Lesq. Cret. Flora, 79. pi. 14.. f. /,
2. 1874 ; Cret. & Tert. Fl. 56. fl. s-f- /, 5. 1883 ; Fl.
Dak. Group, 100. 1892. Newb. Fl. Amboy Clays, 87.
pi. 2s. f. i-io; pi. 26./. 2-6. 1896; Later Ext. Fl. 98.
//. 7./. I. 1898 >^\ cretaceum. Hollick, Trans. N.
Y. Acad. Sci. 12: 236.//. y.f. i. 1893. 16 : 132.//.
14./. 13. 1897. Velen. Fl. Bohm. Kreidef. 4: 2. //.

2./, I. 1886.

S. recurvatum Heer, Fl. Foss. Arct. 6-': 74. pi. sg.f. 4.

1882 ; not Lesq.
There can be no doubt that this rather fragmentary leaf
is referable to Sassajras acutilobum as commonly understood.

(82)

In outline it is almost identical with Lesquereux's typical leaf
from the Dakota Group, the only differences being its slightly
smaller size and somewhat wider median lobe. It is also
very similar to the leaf from the Bohemian Cretaceous which
Velenovsky refers to this species, the only difference being
the less conical lobes. With Newberry's Amboy Clay forms
there is a general resemblance to the more typical specimens.
Hollick, ('97) found a small fragment in this (Clay Marl)
formation which he thought might be referable to this species.
With the question of the proper generic relations of this spe-
cies we are not here concerned. In another place * I have
expressed doubt as to the validity of its reference to Sassafras.

ACERACEAE.

Acer Linn. Sp. PI. 1054. ^753-
There are about one hundred existing species of maples.
The fossil species are also quite numerous, there being some
twenty-six fossil American forms distributed as follows : Rari-
tan I, Island Raritan i, Belly River i, Laramie 5, Denver
I, Ft. Union 3, Green River 3, Eocene 3, Miocene 8, Pleis-
tocene 3.

Heer records one from the Tertiary of Siberia, one from
the Tertiary of Manchuria, three from the Island of Sacha-
lin, two from Patoot, and five from the Tertiary of Greenland.
Ettingshausen records maples in the Tertiary of Australia,
Tasmania and New Zealand.

Acer paucidentatum Hollick.

Acer ^aucidentatum Hollick, Trans. N. Y. Acad. Sci. 16:
132.//. 14../. 2,3. 1897.

As remarked by Hollick this maple resembles several Ter-
tiary species. Acer is only represented by fruit in the New
Jersey Raritan, although a small leaf has been found in that
formation on Staten Island.

Sapindus Linn. Sp. PI. 367. 1753.
Saf Indus is at the present day a chiefly tropical genus of
about ten species of Asia and America. The only existing

* Bot. Gaz. 34 : 438. 1902.

(83)

North American species are S. viarginatus Willd., whicii
ranges from Kansas to northern Mexico and eastward to
Georgia and Florida; ^S*. Safonaria Linn., of the Florida
Keys, West Indies and Venezuela ; and S. Drummondii H. &
A. The American fossil species are numerous, there being
twenty-one or more forms distributed as follows: Island i,
Matawan i, Dakota 2, Denver i, Upper Laramie i, Eocene,
Ky. I, Brandon, Vt. i, Green River 7, Fort Union 5, Ter-
tiary of Yellowstone Park 2, Eolignitic 4, Miocene i, Green-
land 3.

Did we assume that these fossil leaflets should be of uni-
form size and form, as they are in our existing species of
the Southwest, the number of fossil species would be greatly
multiplied.

Sapindus Morrisoni Lesq. PL 47./. 2,j.

Sapmdiis Morrisoni Lesq. Cret. & Tert. Fl. 83. fl. 16.

J. I, 2. 1883 ; Fl. Dak. Group, 158. pi. jj. /• i, 2.

1892. Heer, Fl. Foss. Arct. 6^: 96. fl. 40. f. i; fl.

4i.J.j;fl.4j./.ia,b;fl.44./.7,8. 1882. Hollick,

Trans. N. Y. Acad. Sci. 11: fl. 3. f. 5. 1892; 12:

235. fl. 6. f. 3. 1893; Bull. Torrey Club, 2i : 57.

fl. 179./. 8. 1894; Bull. Geol. Soc. Am. 7: 13. 1895 ;

Ann. N. Y. Acad. Sci. 11: 422. fl. 36. f. 4. 1898.

White, Am. Jour. Sci. III. 39: 99. fl. 2. f. 12. 1890.
The Cliffwood forms are small leaves with a considerably
inequilateral base and numerous somewhat ascending irregu-
larly curved camptodrome secondaries. There is no question
but what this is a species of Safindus ; it agrees quite well
with Lesquereux's Cret. & Tert. Fl. /. 2, and Fl. Dak.
Group, f. .2, and Hollick's specimen from Tottenville,
Staten Island, all of which are rather smaller than the other
figured leaves of this species. Hollick * identifies two fairly
perfect leaves from this horizon with Velenovsky's Safindus
aficulatus from the Bohemian Cretaceous. These leaves are
somewhat smaller than our specimens and less full at the

* Trans. N. Y. Acad. Sci. i5: 133. pi. ij. f. r, 2. 1S97 .

(84)

base on the larger side, otherwise they are quite similar.
They may be either small leaves of Sapindus Morrisoni or
else new forms altogether. It may well be that the leaves re-
ferred to the widely distributed Sapindus Moi'risoni, as indi-
cated above, embrace more than one species.

This is another species which we could expect to find in
the Raritan formation.

Sapindus apiculatus Velen.

Sapindus apiculatus Velen. Fl. Bohm. Kreidef. 3: 6 (53).

pl. 7 {22).f. 1-8. 1884. Hollick, Trans. N. Y. Acad.

Sci. 16: 133. //. jj.f. 7, 2. 1897.

As remarked under Sapindus Morrisoni, this might well be

a small leaf of that species. Recorded by Hollick from the

Matawan ; not found by me.

Celastrophyllum Gopp. Tertiarfl. Java, 52. 1854.
This extinct genus includes leaves related to those of
Colastrus. It has ten species in the Potomac, ten in the
Raritan, one in the Island series, seven in the Dakota, two
in the Matawan, one in the Atane beds and three in the
Patoot beds.

Celastrophyllum elegans sp. nov. PI. 43. f. 6.

A handsome ovate-lanceolate leaf about 6 cm. long and 10
mm. wide at its widest part which is about midway between
the base and the apex ; with a somewhat wedge-shaped base,
an evenly rounded apex and a slightly undulating margin ;
there are eleven shallow indentations on each side, the lowest
about 6 mm. from the base ; petiole rather stout, 10 mm. in
length ; secondaries branch at an angle of somewhat more
than 45° and are straight to within a short distance of the
margin, curving and forming arches only about one milli-
meter from it.

Hollick refers a leaf from this formation to Celastrophyllum
Newberryanum ; ours is a narrower longer leaf with more
regular secondaries. Compared with the Amboy Clay leaves
of C. Newberryanum ours is a longer more slender leaf.
C. Newberryanum was however an abundant and very vari-
able leaf and some of Newberry's figures approach ours quite

(85)

closely, for instance, Fl. Amboy Clays, //. ^9. J. 10.
Our leaf is also somewhat similar to some of the Amboy clay
leaves which Newberry refers to C. crenatum Heer, though
the latter is stouter and averages much larger.

C. f^randifolium Newb. is of somewhat the same propor-
tions but about three times as large. The Dakota species
except C. dccurrens are much smaller leaves. The Potomac
species are mostly smaller broader leaves of rather obscure
affinities.

Celastrophyllum Newberryanum Hollick.

Cclastrop/iylliim JVezt/derryamun HoW'ick ; Newb. Fl. Am-
boy Clays, loi. pL ^p. _/*. 7-^7. 1896; Trans. N. Y.
Acad. Sci. 16: 133.//. 14. f. i. 1897.
This small-leaved Cclastrophyllum is abundant in the
upper Raritan beds at South Amboy and we would naturally
expect it to extend upward into the Cliffwood beds, where it
is recorded by Hollick.

RHAMNACEAE.
Rhamnus Linn. Sp. PI. 193. 1753.

About seventy-tive existing species, mostly north temper-
ate, a few tropical and a few south temperate ; eleven inhabit
North America. Thirty-three species are found fossil on
this continent in the following formations : Island Raritan
2, Dakota 6, Montana i, Laramie 10, Denver 6, Ft. Union
2, Green River 3, Eolignitic 2, Tertiary i.

Heer records one from the Tertiary of Manchuria, one
from the island of Sachalin, two from Atane, one from Patoot,
and eight from the Tertiary of Greenland.

Rhamnus Novae-Caesareae sp. nov. PL 50. y". 5, 6.

These leaves are somewhat similar in outline and venation
to what Hollick calls .^ercus{?) JVovae- Caesar eae from this
same locality, only our specimens are less perfect and consid-
erably larger. Their true affinity seems to be with R/iamnus,
and I have been unable to associate them with any of the

(86)

described species. Remains are fragmentary, but indicate a
simple ovate-lanceolate leaf 7-10 cm. long by about 2.25
cm. wide, with ascending camptodrome secondaries and
transverse tertiaries.

RhAMNUS INAEQUILATERALIS Lesq.

Rhajunus inaequilateralis Lesq. Fl. Dak. Group, 170.
fl. 3J.f. 4.-7. 1892. Rollick, Trans. N. Y. Acad. Sci.

16: 133- /^- 13^/- S. 1897.
The Cliffwood forms are identical with the smaller of
Lesquereux's leaves. Recorded by Hollick from the Ma-
tawan formation ; not found by me.

Paliurus Mill. Gard. Diet., ed. 7. 1759.

There are only two existing species, one confined to
southern China and Japan, and the other to southern Europe
and western Asia. The fossil species are numerous, some
sixteen occurring on this continent ; it is pertinent to remark,
however, that in the absence of fruit Paliurus is practically
indistinguishable from Zizyp/ms or Ceanothus.

Raritan i. Island Raritan 3, Dakota 5, Mill Creek 2, Van-
couver I, Laramie 4, Canadian Upper Laramie i, Denver 3,
Ft. Union 2, Green River 2, Miocene i.

Heer records one from the Tertiary of Siberia, one from the
Island of Sachalin, one from Patoot, and three from the Ter-
tiary of Greenland.

Paliurus integrifolius Hollick (?).

Paliurus integrifolius Hollick, Bull. Torrey Club, 21 : 57.
fl. 177./. 5, «?, 12. 1894 ; Trans. N. Y. Acad. Sci. 16 :
133. ;pL 14./. 10. 1897.

This reference of a fragment from near Cliffwood was only
provisional. The specimen represents the basal fragment of
a leaf which is rather large for Paliurus and lacks the lat-
eral branches of the primaries which ought to be present in
the left hand portion of the specimen. It might well represent
the basal portion of some of the leaves from the Raritan
formation which Newberry referred to Cissitcs formosus Heer.
Not found by me.

(87)

MYRTACEAE.
Eucalyptus L'Her. Sert. Angl. i8. 1788.
There are about one hundred and forty existing species of
great variety of form, foliage and blossom, confined to the
Australian region, none occurring in New Zealand on the one
hand or Asia on the other.* Nine fossil species have been re-
ferred to this genus from American strata, most of which are
doubtfully determined owing to the uncertainty of leaf re-
mains: Island Raritan 2, Dakota 3, Raritan 4, Laramie i,
Green River i, Atane 2.

Eucalyptus Geinitzi Heer. PI. jj. f. j.

Myrtofhyllum [Eucalyptus P) Geinitzi Heer, Fl. Foss.

Arct. 3': 116. pi. 32./. 14-17; pi. 33- f' 6b. 1874.
Eucalyptus Geinitzi Heer, Fl. Foss. Arct. 6^: 93.^/. ig.

f. ic; pi. 63./. 4-g. 1882. Lesq. Fl. Dak. Group, 138.

pi- 37'f' 20. 1892. Newb. Fl. Amboy Clays, no. pi.

32. f. 2, 12, 13^ 16. 1896. Hollick, Trans. N. Y.

Acad. Sci. 11: 98. //. 2./. i. 1892; 12: 34. pi. 2.

f. 3. 1892 ; 236. pi. 6. f. 2. 1893 ; Bull. Geol. Soc.

Am. 7: 13. 1895; Annals N. Y.Acad. Sci. 11: 6o.pl.

4./. 1-3. 1898. White, Am. Jour. Sci. III. 39: 98.

pi. 2. f. 8-11. 1890. Heer, Kreide Fl. Moletein, 22.

pi. I J./. 3, 4.
This reference is only provisional, as the leaf is too frag-
mentary for certainty and the venation is entirely obliterated.
I have thought I detected the characteristic venation at times
but cannot be certain. Previously recorded from the Dakota,
Kansas ; the Raritan, Woodbridge, Sayreville, N. J. ; and the
Cretaceous of Staten Island, Long Island and Martha's Vine-
yard. Also recorded from Greenland and the continent of
Europe.

Eucalyptus (?) dubia sp. nov. PI. 32./. i.

This fragment is referred to Eucalyptus because of its
resemblance to Newberry's Fl. Amboy Clays, pi. 32. f. d,

* Although Ettingshausen records them from the Tertiary of New Zea-
land (Trans. N. Z. Inst. v. 23).

(88)

{B. (?) angusti folia). In all probability neither of these
leaves is related to Eticalyftus.

The secondaries spring from the midrib at an angle of 45*^
to 50° and are approximately straight and parallel, about
1.5 mm. apart, some opposite, others irregular; intermediate
tertiaries give alternate branches to each secondary. The
venation is not characteristic of JSucalyptus , and resembles
somewhat that of Lmirofhyllum reticulatum Lesq., but our
specimen is a smaller, relatively narrower leaf. A second
specimen shows a marginal vein connecting the secondaries
about .5 mm. from the margin.

STERCULIACEAE.
Sterculia Linn. Sp. PL 1007. 1753.

Nearly one hundred existing species of the tropics of both
hemispheres. American fossil species are some seventeen
in number, distributed as follows: Potomac i, Cheyenne
Sandstone i, Raritan i, Island Raritan 2, Dakota 8, Creta-
ceous of British Columbia i, Denver i. Green River i.

None have been recognized in the Atane beds of Green-
land, although one species occurs at Patoot. One species has
been recorded from the Island of Sachalin ; the genus is
present in the upper Cretaceous of Europe and some sixteen
species have been described from the European Tertiary de-
posits, although the American Tertiary is practically barren
of these plants.

Sterculia Cliffwoodensis sp. nov. PI. 4.3./. 5.

Sterculia sp. (?) Hollick, Trans. N. Y. Acad. Sci. 16:
133. i>l. 14../. 4-7. 1897.

It is to be hoped that more and fuller remains may be dis-
covered of this beautiful species. The present specimen in-
dicates a trilobed leaf with entire margin and decurrent base ;
lobes diverging at an angle of about 45° or less, with nearly
parallel margins and acute tips (?); the secondaries branch
at a wide angle and their tips are joined by rather straight
arches.

(89)

Lesquereux separates the Dakota leaves of Sterculia from
Aralia merely on account of the " primary divisions and pri-
mary nerves from the top of the petiole." This character,
which I do not consider diagnostic, would refer this leaf to
Aralia as the lateral primaries branch from the midrib a con-
siderable distance above its base. The venation is some-
what similar to the Dakota Sterculia reticulata Lesq. (Fl.
Dak. Group, pi. 34. f. 10), and also to that of Aralia trans-
z'crsincrvia Sap. & Mar. described by Hollick from Oak-
neck, Long Island (Bull. Torrey Club, 21 . 54. pi. 176./. i .
1894) which leaf he does not consider an Aralia.

In outline this leaf resembles Sterculia lugubris Lesq. ex-
cept that the primaries are not basal ; whether the lobes were
produced to the length they are in that species is of course
conjectural. Our fragment is also somewhat similar in out-
line to the fragment (Fl. Amboy Clays, ^/. 26./. 2) referred
by Newberry to Aralia quinquepartita Lesq., in which how-
ever the venation is unfortunately obliterated. Hollick's
Sterculia sp. (/. c.) probably belongs here ; his f. 4 is the
fragment of a much smaller leaf, but his larger fragments
(y. 5-7) might well be the acutely tipped lobes of our leaf,
the venation of the two corresponding very well.

Our leaf also has somewhat the appearance of Aralia
Jorgenseni Heer (Fl. Foss. Arct. 7: pi. loi. f. /) but the
sinuses are not quite so deep. It might further be compared
to Aralia Wellingtoniana Vaughanii Knowlton from the
Woodbine formation of Texas (Dakota). The latter is tri-
lobed, the lobes slender and entire ; not figured, however.
(Knowlton; Hill, Ann. Rep. U. S. Geol. Surv. 21: 318.
1901.)

Sterculia Snowii bilobata var. nov, PI. 4.3./. 7.

Sterculia Snowii \s known from the Dakota Group of Kan-
sas and New Mexico and the Cheyenne Sandstone at Belvi-
dere, Kansas. The specimen from the Matawan here figured,
in its outline approximates Liriophylhim Beckzt'ithii Lesq.,
from the Dakota Group, but the venation is radically differ-

(90)

ent. We have supposed its relationship to be with Stei-ctdia
Sno'ivli not onl}'^ because the latter already has a remarkable
biiobate form (var. disjunctd)^ but also because we can readily
imagine a leaf like the leaf of S. Snozuii shown on Fl. Dak.
Group, fl. jj. y. J with a deeper sinus which would then
make it correspond with our specimen.

This reference is far from satisfactory ; it would seem that
if this is a variety of S. Snowii the latter ought to be present
as well or at least in the Raritan, although it has not as 3'et
been discovered, except a doubtful specimen from Tottenville,
Staten Island (Hollick, Ann. N. Y.Acad. Sci. 11 : 422.
i>i-37'f' 4' 1898). Although Lesquereux in his Report
on the Clay Deposits of New Jersey (1878) recognized un-
determined species of Stcrculia at three different localities,
no decisive remains of this genus have thus far come to light
in the Raritan formation in New Jersey. Dr. Hollick, to
whom a sketch of this leaf was sent, is disposed to compare
it with Fl. Dak. Group, ;pl. 21. f. 5, which Lesquereux
refers to Heer's Cissites formosus. Inasmuch as Lesquer-
eux's determination is doubtful, because the secondary system
is quite unlike that of Cissites in looping along the margins
and more like that of Sterculia^ I prefer to consider it more
nearly related to the latter, at least provisionally. It may also
be compared withy, j of Aralia concreta Lesq., as figured
by him in Cret. & Tert. Fl. fl. g. Stcrculia limhata "V'elen.,
the Bohemian representative of S. Snowii^ has sometimes
4-lobed, 4-veined leaves.

Sterculia mucronata Lesq. , PL 4J./. j.

Stcrculia mucronata Lesq. Fl. Dak. Group, 182. fl. jo.
f. 1-4. 1892.

In the absence of the apical and basal portions, and be-
cause of the obliteration of the venation, it is with consider-
able hesitation that I refer this small bilobed leaf to the above
Dakota Group species. That it is referable to Sterculia is I
think probable, but just which species to associate it with is
doubtful. It is about the same size as Lesquereux's smaller

(90

specimens (y. j, ^), but differs in having the midrib branch
at some distance above the base. Both ancient and modern
Stcrculia leaves vary considerably, and StcrcuUa Snozuii has
a bilobed form.

There is considerable resemblance to the smaller leaves
from the Raritan at Woodbridge, which Newberry refers to
Sassafras aaUilohiim Lesq. (Fl. Amboy Clays, //. 2§.f. ^,5,
(5, id). Leaves of the living Sterculia diversifolia occiden-
talis Benth., from interior Australia, contained in the Meis-
ner Herbarium, are very similar to Sterculia mucronata.

ARALIACEAE.

Aralia Linn. Sp. PL 273. 1753.

The genus Aralia has never been precisely defined for the
paleobotanist, the custom being to follow precedent and refer
a variety of polymorphic leaves of synthetic types to this
genus — leaves having a variety of affinities, Platanoid, Sas-
safroid, etc., as well as leaves allied to Cissus and Hedera.
The existing flora includes some twenty-seven species of
North America and Asia, six of which are American ; only
one of these, however (^4. spinosa L.), is arborescent. The
ancient flora contains numerous leaves that have been referred
to this genus, besides several that have been referred to the
allied genus Araliaefhyllum. The distribution of the Amer-
ican fossil species is as follows:* Potomac i, and 4 sp. of
Araliaephyllian^ Raritan 8, Matawan 6, Island Raritan 5,
Dakota 13, Mill Creek 3, Laramie 3, Denver i, Ft. Union 5,
Tertiary of Yellowstone Park 3, Green River i, Eocene 4,
Miocene 5, Atane beds 2, Patoot beds i, Greenland Tertiary 2.

Velenovsky enumerates two species from the Cenomanian
of Bohemia, A. decurrens being apparently identical with A.
Safortanea of the Dakota, and the other, A. coriacea, reap-
pearing at Martha's Vineyard. From the European Tertiary
some thirty-two species are recorded, none occurring in the
existing flora of Europe.

* Ettingshausen (Trans. N. Z. Inst. 19: 449) records Aralia in the Tas-
manian Tertiary.

(92)

Lesquereux (Fl. Dak. Group, 249) characterizes these
leaves as follows: "Base decurrent, primary nervation pal-
mately trifid and supra-basilar," but he repeatedly fails to
conform to his definition. Thus his A. acerifolia^ A. tenui-
vervis, A. dissecta and A. subemarginata lack the decurrent
base, as do also five of Newberry's Raritan forms. The pri-
maries are generally subbasal and are basal in Lesquereux's
A. quingticpartita, A. notata, A. dissecta and A. Masoni.
I have been at a loss to distinguish between Sterculta and
Aralia in the Matawan material ; no mutually exclusive
characters can be gathered from the published descriptions
or figures, and as it would be useless to attempt a revision
without an examination of all the collected material, I have
been forced to follow the pernicious precedent above al-
luded to.

Aralia Towneri Lesq.

Aralia Towneri Lesq. Bull. U. S. Geol. & Geog. Surv.
Terr, i : 394. 1876 ; Ann. Rep. ibid. 1874 : 349. pL 4.
/. 3. 1876; Cret. & Tert. Fl. 62. //. 6./. 4. 1883 ; Fl.
Dak. Group, 132. pi. 2j. f. jy 4; fl. 31./. i. 1892.
Hollick, Trans. N. Y. Acad. Sci. 16: 132. pi. 14./.
II, 12. 1897.
Described originally from the Dakota Group of Kansas,
Hollick has doubtfully referred two fragments from the clays
near Cliffwood to this species. These fragments are so in-
complete that the form of the leaf is more or less conjectural.
The secondaries are straighter than in Lesquereux's speci-
mens and branch from the primaries at a wider angle. At
the same time they seem to differ from my specimens from
this formation which I have referred \.o Aralia Ravniana Heer.

Aralia Ravniana Heer. PI. 46./. 7 ; pi. 53. f. 2 ; pi. 57./. i •
Aralia Ravniana Heer, Fl. Foss. Arct. 6-: 84. pi. 38./.

/, 2. 1882.
The specimens from Cliffwood figured above are identical

with Heer's Aralia Ravniana from the Atane schists at

(93)

Igdlokunguak, Greenland, except that the basal primary
forks a considerable distance from its base. I was at first
disposed to refer them \.o Aralta Tozuneri Lesq., particularly
as Heer compared his leaves with that species and Les-
quereux suggested * that the two were identical.

While the occurrence of two such large-leaved species of
Aralia in the Matawan formation may seem anomalous,
especially as they had much in common, I fail to see their
identity. Aralia TozvncriWiH^ a palmately five-lobed leaf with
a decurrent base and obtuse lanceolate lobes. Aralia Rav-
niana on the other hand was probably a six- or seven-lobed
leaf of large size, for while in no case is the apex preserved,
I cannot conceive that such a leaf as the discovered frag-
ments evidently represent could have had an undivided
terminal lobe. If they had, they would differ from all other
species of Aralia in its size, and from lobed leaves in gen-
eral. They would have had a lobe wider than long, with an
area greater than the balance of the leaf, the deep lateral
sinuses almost cutting it off from the rest of the blade. In
the specimen figured at //. SJ,/. i (one fourth natural size)
the midrib more than half way to the tip gives off a strong
lateral branch which it seems reasonable to suppose formed
the midrib of a lateral lobe. Furthermore, A. Ravniana
differs from A. Toivneri in having stouter primaries, nar-
rower sinuses, more ovate lobes, the basal ones widely spread-
ing, and the base but slightly or not at all decurrent. If we
may judge from the obsolete venation, it was a more cori-
aceous leaf.

Aralia palmata Newb. PL 4.4..

Aralia pahnata Newb. Fl. Amboy Cla3's, 117. ^l. jg. f.
6, 7; ^l. 40./. 3. 1896.

It is easier perhaps to criticise others than to escape criti-
cism oneself, at the same time in considering the leaves which
seem referable to Aralia in our collections from near Cliff-
wood and in comparing them with the Raritan forms referred

* Cret. & Ten. FI. 105.

(94)

to this genus by Newberry, we are struck with the range of
variability, not only within each species as defined by him,
but in the whole group, and the thought forces itself upon us
that perhaps it would be an advantage to cut down the total
number of species. Newberry has described seven species
from the Woodbridge horizon alone and only one from the
higher beds at South Amboy. The occurrence of a variety
of Ara/ia-liVie leaves in the Matawan shows that the imper-
fection of the record is probably responsible for their absence
in the intervening beds : and it seems rather incredible that
each form represents an ancient species that flourished on the
New Jersey coast in Cretaceous days.

The leaves before us, while not uniform, seem to more
nearly represent Aral/a fahnata than any other known form.
Our f. 5 may be compared with Newberry's f. 6. While it
is about one sixth smaller, other fragments from Cliffwood
indicate a somewhat larger size ; the lobes are a trifle more
slender and the main sinuses somewhat deeper. The lower
margins were sometimes undulate and the latero-basal lobes
short. The lobation was, however, somewhat variable, as it
was also in Newberry's leaves. Oury. 6 might be compared
with a variety of forms, such as Ficus (?) Alaskana Newb.
(Later Ext. Fl. //. 51 . f- i) and Hedera obliqua Newb. or
Hedera ^rimordialis Newb. of the Amboy Clays. It has
the secondaries straighter than is usually the case in this
genus; in its tertiary venation it agrees with Newberry's^, j
of Aralia -palmata. The secondaries were distant and were
joined at their tip by widely arching loops.

Other species with which our leaves may be compared are
Cissites ingens Lesq. and Liquidamhar integrifol'mm Lesq.
(Cret. and Tert. Fl. fl. 14./. 3).

Aralia Groenlandica Heer. PL ^5. /. 4.

Aralia Groenlandica Heer, Fl. Foss. Arct. 6^: 84. pi. jS.

/• 3; P' 39' f' i^ t^- 46'/- i^^ n- 1882 (/. /; is
Aralia Ravniana) ; pi. 3()- /• 3 of Sassafras rccurvatuni
Lesq. is in all probability this species. Lesq. Fl. Dak.

(95)

Group, 134- /^- i-/-y^ i-J- 1892. Hollick, Bull. Geol.
Soc. Am. 7: 13. 1895. Nevvb. Fl. Amboy Clays, 116.
//. 28. /. 4. 1896.

A widespread species recorded from the Atane schists,
Greenland; Dakota Group, Kansas; Raritan, Woodbridge,
N. J. ; and the Cretaceous at Martha's Vineyard.

If Newberry has correctly identified Fl. Amboy Clays, pi.
28 if. /, as the above species then our fragment undoubtedly
belongs to the same species. It is the same size as New-
berry's leaf ; the angle of divergence of the lateral primaries
is a trifle greater however, and the primary venation is
stronger, agreeing with Lesquereux's and Heer's leaves in
the latter particular. Both the New Jersey leaves are smaller
than the Dakota and Greenland specimens and have relatively
narrower lobes. Unfortunately the basal portion of the Cliff-
wood leaf is gone, so we do not know whether or not there
was an extra pair of laterals springing from the base of the
midrib. This is a feature of all the leaves which Lesquereux
has referred to this species, but is wanting in Heer's pi. j8y
f. J, and is also wanting on one side in the Raritan leaf.

The leaf which Newberry describes as a new species (/. c.
11^. pi. 28. y. j), under the name of A7'alta patens^ should
in all probability be considered as a form of his A. Grocn-
landica with deeper sinuses and more divergent lobes, as he
suggests. Our leaf might also be compared with Cret. &
Tert. Flora, pi. ^. f. i, which Lesquereux considers Sassa-
fras actitilobum ; it is also much the same form of leaf as
Sterculia aperta Lesq., but larger ; and there is considerable
resemblance to the leaf which Heer refers to Sassafras Fcr-
retiana Mass. (Fl. Foss. Arct. 7: //. 97. y. 5).

Aralia Mattewanensis sp. nov. PI. 4.3./. 2 ; pi. 46./. 6.

A palmately four- or five-lobed leaf ; lobes oblanceolate in
outline (tips missing), with rather narrow sinuses nearly to
the base ; primaries rather stout ; a majority of the seconda-
ries branch at a wide angle and are nearly straight to within
a short distance of the margin, along which they arch. Leaf
coriaceous, if we may so judge from the obsolete venation.

(96)

These leaves have a distant resemblance to Lesquereux's
Cisst'tes /ormosus Heer (Fl. Dak. Group, ^/. 21. f. ^) but
bear no resemblance to the Amboy Clay leaves which New-
berry refers to that species. Our leaves also suggest some
forms of Aralia such as A. quinqiiefartita Lesq., but the
base is apparently not decurrent and the primaries branch
from the midrib at the same place, the lateral ones at nearly
right angles.

Aralia Brittoniana sp. nov. PL ^5. f. j.

I have been unable to identify this with an}^ known species
of Aralia and therefore add another to the long list of diver-
sified leaves of this genus which have been found in the
Raritan and Matawan formation. In size and outline it re-
sembles Aralia acerifolia Lesq. of the Fort Union beds of the
West, but the secondaries are stronger and more regular.
The specimen denotes a leaf which was trilobed with an evi-
dent tendency to produce an extra latero-basal lobe on each
side ; with a broadly truncated base which curves upward
for about half the distance to the tip to form a point above
which the margin is concave ; lobes presumably acute ; ter-
minal lobe broad with moderately convex sides ; sinus to be-
low the middle, rounded ; primary and secondary venation
strong, but tertiary venation entirely obsolete ; lateral pri-
mary could not have branched far from the base and forms
an angle of about 45° with the midrib, leaving room for a
secondary below ; secondaries regular, leaving the primaries
at a wide angle and running straight to within a short dis-
tance of the margin and then curving to join the secondary
next above. Our only specimen was evidently not bilaterally
symmetrical.

ERICACEAE.

Andromeda Linn. Sp. PI. 393. 1753-
At the present time a monotypic genus of the north tem-
perate and subarctic zone. Many fossil leaves have been
referred here, some twenty-five species in this country alone.

(97)

The generic determination of Ericaceous leaves is always
however a matter of extreme uncertainty, which is fully
shared by the following distributed American forms : Rari-
tan 6, Island Raritan i, Dakota 9, Woodbine i, Montana i,
Laramie 2, Livingston i, Denver i, Green River 2, Eocene
2, Miocene i, Pleistocene 2, Atane 2, Greenland Tertiary 5.

Andromeda Parlatorii Heer. PL 50. f. i-^.

Andromeda Parlatorii Heer, Phyll. Cret. Nebr. 18. //. /.
/. 5. 1866. Lesq. Cret. Flora, 88. //. 23. f. (5, 7; //.
28. f. /J. 1874; Fl. Dak. Group, 115. -pL ig. f. i ; fl.
52./. 6. 1893. Newb. Fl. Amboy Clays, 120. pi. ji.
f. i-y; pi' 33' /• i, 2, 4,3. 1896. Heer, Fl. Foss. Arct.
^: 112. pi. 32. /. 1,2. 1875; 6^: ']<^. pi. 21. /. ib^ii ;
pi. 42. /. 4c. 1882. Hollick, Bull. Torrey Club, 21 :
54. pi. 173- /• 2, 3. 1894; Ann. N. Y. Acad. Sci. 11:
420. pi. 37'/. 7. 1898. White, Am. Jour. Sci. IH. 39 :
97. pi. 2./. 4. 1890.
Prunus ? Parlatorii Lesq. Am. Jour. Sci. H. 46: 102.

1868.
This species is recorded from the following localities :
Dakota Group: Kansas, Nebraska, and New Ulm, Minn.
Cretaceous: Sea Cliff, L. L, Tottenville, Staten Island
and Martha's Vineyard. Raritan : common at nearly every
locality opened. Greenland: Atane beds at Atanekerdluk,
Isunguak and Igdlokunguak.

Newberry (/. c.) is inclined to doubt the reference of all
these leaves to Andromeda, pointing out that generic determi-
nations of most Ericaceous leaves are always doubtful. Heer
compares his Greenland specimens to Ettingshausen's Laurus
cretacea from Niederschona. In oury". 4 the finer reticula-
tion is very minute and five- or six-sided.

MYRSINACEAE.

Myrsine Linn. Sp. PL 196. 1753.
Fossil American species occur in the following formations :
Raritan 3, Island Raritan 3, Dakota 2, Green River i, Atane
I, Greenland Tertiary 2.

(98)

The family is a large one of the tropics of both hemispheres.
In the recent monographic revision by Carl Mez (Engler,
Pflanzenreich, Heft 9, 1902) nine hundred and thirty-three
species are enumerated distributed among thirty-two genera
and nine fossil genera are enumerated. Four species, all
arborescent, enter the United States, one of them a true
Myrsine, the others referred to the genera Icacorea {Ardisia)
and Jacquinia. They range from southern Florida through
the West Indies, Central America, Mexico and northern
South America.

Myrsine crassa Lesq. PI. 53. /. 6.

Myrsine crassa Lesq. Fl. Dak. Group, 114. fL 52. f. 2,
3- 1892.

The single leaf which I have referred to this Dakota species
was lost after the hurried sketch which is here reproduced
was made and the reference can therefore be only provisional
unless additional specimens are discovered.

The outline and venation suggest this species although it is
a somewhat smaller leaf. I was at first inclined to refer it
to Liriodendrofsts, which it greatly resembles, but in the
absence of the apex our reference of it to this species of
Myrsine is warranted.

Of Uncertain Affinities.

Dewalquea Groenlandica Heer (?). PL SJ.f. 3.

Dewalquea Groenlandica Heer, Fl. Foss. Arct. 6^: 87.
fL 29./. 18, ig; fl. 42./. s, 6; fl. 44./. 11; 7: 37.
fl. 62. f. 5, 6. Newb. Fl. Amboy Clays, 129. fl. 41. f.
.?, J, 12. 1896. Hollick, Ann. N. Y. Acad. Sci. 11 :

A,iZ'i>l' 3^-f' 7' 1898.
Obscure leaf-remains of uncertain botanical affinities ; in-
cluded by Heer in the Ranunculaceae. Leaves (or leaflets)
with very tapering base, thick midribs, and short petioles ;
apparently rather coriaceous in texture and with the venation
entirely obliterated. They agree fairly well with the figures
of this species as cited above. This is another species which

(99)

occurs in the New Jersey Raritan (localities not given) ; on
Staten Island ; and in the Atane and Patoot beds of Green-
land. The genus was founded by Saporta & Marion* and
embraces several European species of which Dewalquea m-
signis reappears in both the Atane and Patoot beds of Green-
land, and on Staten Island ; while D. Haldemiana reappears
in the Patoot beds, in Utah, and on Staten Island. The
Dakota group furnishes two additional species.

PoDOZAMiTEs MARGiNATUS Heer. PL 46./. J-J.

Podozamites marginatils Heer, Fl. Foss. Arct. 6^: 43. fl.
16. f. JO.

Similar remains are common in the Raritan (three species).
Hollick (Bull. Torrey Club, 21 : 62. pi. 180. f. 4) records a
fragment from Glen Cove, Long Island, and also from Chap-
paquidick Island, Mass. (Bull. N. Y. Bot. Gard. 2 : 401. pi.
41./. 8, p. 1902).

The genus was founded by Fr. Braun, in Miinster, Beitr.
Petrefacten-Kunde (Heft. 6. 28. 1843) and is chiefly Upper
Triassic (Rhetic) and Jurassic, becoming decadent in the
Cretaceous. The latter has yielded, however, seventeen
species on this continent (including Greenland), nine of which
existed as late as the mid-Cretaceous, all described from
rather fragmentary and somewhat doubtful leaf-remains.

Our specimens would appear to be fragments of Podozam-
ites tnarginatiis? Heer, which occurs at Woodbridge in the
Raritan clays (Fl. Amboy Clays, 44. pi. 13. f. j. 6);
originally described by Heer from Atane, Greenland.

Phragmites (?) ClifFwoodensis sp. nov. PI. 46./. 5.

A terminal, sharply pointed fragment of a monocotyle-
donous leaf, 12 cm. long and 5.5 mm. broad, finely parallel-
veined.

The remains of Phragmites usually consists of leaf frag-
ments or rhizomes, all of rather doubtful affinity, although a
single palet of P. Oeningensis A. Br. is described by Heer
from Greenland.

*Mem. Cour. & Sav. Etraugers Acad. Belg. 37 : 55. 1873.

( loo)

The Matawan remains are too small to be definitely referred
to Phragmites, and may be compared to those referred to
Poacites^ Cyj>e7-ites, etc.

Chondrites flexuosus Newb. (?)

Chondrites jlexuosus Newb. Fl. Amboy Clays, 34. ^l. i.
f. 7, 4. 1896.

Obscure remains from near Clifford (not figured), of doubt-
ful botanical affinities, may be compared with the above
species which occurs at Sayreville, Woodbridge, etc., in the
Raritan clays.

Carpolithus juglandiformis sp. nov. PI. 46./. 8.

Has a superficial resemblance, but no botanical affinity
with some of the fruits referred to the genus Cycadeosfermum.
Is evidently not a seed-bearing scale but seems to have been
a small nutlet which has been compressed and transformed
into lignite ; bears considerable resemblance to Juglans cos-
tata (Presl) Brongn. as figured by Lesquereux (Cret. & Tert.
Fl. ^l. 3g- f' 5) from the Green River group at Florissant,
Colorado.

Carpolithus Cliffwoodensis sp. nov. PL 48. f. 6.

This specimen resembles a number of seeds figured by
Hear from the Arctic regions, as for instance Lamprocar-
■pites nitidus (Fl. Foss. Arct. 6^ : pi. 8. f. 12-14) ^"^ Car-
■polithes najaditan {ibid, i : pi. 2j. f. /j", /J*^), although with
the exception oif. 14 our specimens are about twice the size
of any of those of Heer.

Carpolithus dubius sp. nov. PL 48. f. 7.

This appears to be a thick, inequilateral, oblong scale. It
is about 2 mm. in thickness and the surface is irregularly
roughly lined. Botanical affinity vague.
Carpolithus Virginiensis Font. (?) PL 48. f. 5.

Carpolithus Virginiensis Font. Potomac Flora, 266. pi.
134' /• 11-14-'^ t^- 135- f' r, i; pL 16S. f. 7, ya.

(lOl)

1889. Ward, Ann. Rep. U. S. Geol. Surv. 19- : 693.
//. i6().f. 16. 1899.
Without a comparison of specimens this reference is only
provisional, although from the published figures our specimen
is almost exactly like the detached specimens from the Po-
tomac formation described as above. Fontaine considers
them as probably belonging to some species of BatcropsiSy a
Lower Cretaceous genus which does not occur in New Jersey ;
nor is it at all likely to have persisted as late as the Middle
Cretaceous. Remains are abundant in the Potomac formation
attached to stems ; also found detached in the Lower Creta-
ceous of the Black Hills ; and in the Kootanie at Great Falls,
Montana. Remains indicate a small, smooth, and hard
nutlet.

Carpolithus drupaeformis Hollick.

Carpolilhus drupaefornns Hollick, Trans. N. Y. Acad.
Sci. 16: 134. pi. II. f. ^, 4a. 1897.

Apparently the seed of some drupaceous fruit as the name
indicates. Recorded by Hollick from the Matawan forma-
tion near Cliffwood, N. J. ; not found by me.

Strobilites inquirendus Hollick.

Strobilites inquirendus Hollick, Trans. N. Y. Acad. Sci.
16: iT,o. pi. ii.f. I. 1897.

Remains of doubtful affinity, possibly a distorted and
somewhat macerated cone. Recorded by Hollick from the
Matawan formation near Cliffwood, N. J. ; not found by me.

PiTYOXYLON HoLLiCKi Knowlton.

Pityoxylon Hollicki Knowlton ; Hollick, Trans. N. Y.
Acad. Sci. 16: 134./. /, 2. 1897.

Recorded by Hollick from the Matawan formation near
Cliffwood, N. J. ; not found by me.

(I02)
Description o1 Plates.

Plate 43.

Fig. I. Nelutnbo pritnaeva sp. nov.

Fig. 2. Aralia Mattewanensis sp. nov.

Fig. 3. Sterculia mucrotiata Lesq.

Fig. 4. Moriconia cyclotoxon Deb. & Ett.

Fig. 5. Sterculia Cliffwoodensis sp. nov.

Fig. 6. Celastrophyllum elegans sp. nov.

Fig. 7, Sterculia Snoivii bilobata var. nov.

Plate 44.
Aralia palinata Newb.

Plate 45.
Figs. I, 2. Sassafras acutilobum Lesq.
Fig. 3. Aralia Brittoniana sp. nov.
Fig. 4. Aralia Groenlandica Heer.

Plate 46.
Figs. 1-3. Podozaniites marginatus Heer.
Fig. 4. Arisaema cretaceum Lesq.
Fig- 5- Phragniites (?) Cliffwoodensis sp. nov.
Fig, 6. Aralia Mattewanensis sp. nov.
Fig. 7. Aralia Ravniana Hear.
Fig. 8. Carpolithusjuglandiformis sp. nov.

Plate 47.

Figs. I, 5> 8. Lanrophyllum angusti/oliujn Newb.

Figs. 2, 3. Sapindus Morrisoni Lesq.

Fig. 4. Magnolia obtusata Heer.

Fig. 6. Quercus sp.

Fig. 7. /^^V«5 IVoolsoni Newb.

Fig. 9. Laurus proteaefolia Lesq.

Fig. 10. Magnolia tenui/olia Lesq.

Plate 48.
Figs. 1-4. Moriconia cyclotoxon Deb. & Ett.
Fig. 5. Carpolithus Virginiensis Font. (?).
Fig. 6. C. Cliffwoodensis sp. nov.
Fig. 7. C. dubius sp. nov.
Figs. 8-11. Dammara Cliffwoodensis Hollick.
Fig. 12. Salix proteaefolia flexuosa (Newb.) Lesq.
F'g- 13- Quercus Holniesii Lesq.
Figs. 14, 19. Cunning hantites squamosus Heer.
Figs. 15, 16, 17, 20. Sequoia Reichenbachi (Gein.) Heei.
Fig. 18. Sequoia Reichenbachi (Gein.) Heer. (?).
Figs. 21-22. Sequoia gracillivia (Lesq.) Newb.

( I03)

Plate 49.
Figs. 1-5. Laurophylluin angustifoliutn Newb.
Fig. 6. Laurus proteaefolia Lesq.
Fig. 7. Populites tenuifolius sp. nov.

Plate 5°-
Figs. 1-4. Andromeda Parlatorii Herr.
Figs. 5, 6. Rhamnus Novae-Caesareae sp. nov.
Figs. 7, 8. Laurus Hollae Heer.
Figs. 9-1 1. Laurus Plutoiiia Heer.

Plate 51.
Figs. I, 2. Quercus Hollickii s'p. nov.
Fig. 3. Undetermined.

Fig. 4. Quercus (f) Novae-Caesareae Hollick.
Fig. 5. Salix Mattewanensis sp. nov.
Figs. 6-9. Proteoides daphnogenioides Heer.

Plate 52.
Fig. I. Eucalyptus (f) dubia sp. nov.
Fig. 2, Salix proteaefolia flexuosa (Newb.) I,esq
Fig. 3. Undetermined.
Fig. 4. Laurus Hollickii sp. nov.
Fig. 5. Ficus reticulata (Lesq.) Knowlton.
Fig. 6. Myrsine crassa Lesq.
Figs. 7, 8. Laurus Hollae Heer.
Fig- 9- Quercus sp.

Plate 53.
Figs. I, 4. Ficus reticulata (Lesq.) Knowlton.
Fig. 2. Aralia Ravnia7ia Heer.
Fig. 3. Eucalyptus Geinitzi Heer.
Fig. 5. Magnolia Woodbridgensis Hollick.

Plate 54.
Boulders of clay on beach near Cliff wood, N. J., yielding plant remains.

Plate 55-

Showing how face of bluflFnear Cliff wood, N. J., is obscured by landslips.

Plate 56.
View of bluff near Cliffwood, N. J., showing alternating layers of sand
and lignite. '

Plate 57.
Fig. I. Aralia Ravniana Heer. ClifiFwood specimen restored, % nat-
ural size.

Fig. 2. Magnolia Woodbridgensis Hollick.
Fig. 3. Dewalquea Groenlandica Heer. (?)

Bill. N. V. Ht)T. (i.\Ki).

\<)L. III. PL. 43,

NKLIMIU). .\R.\L1.\. STI;RCI"LI.\. .M( )R ICC )N1.\.
CELASTROIMIYLLUM.

I>1 1 1.. N. \'. lIoT. (i AHI).

\'<)i.. Ill Pi.. aC.

PODOZAMITKS. AR1S.\I:M.\.

iMiRA(,Mrn;s i .- . .\r.\i,i.\. LWRPoLiriirs.

r.ri.i.. N. ^■. 15oT. Gakd.

Vol.. 111. Pi.. 47.

LAUROPHVLLLM, SAPINUUS, MACNOLIA,
QLERCUS, FICUS, LAURUS.

Br 1. 1.. N'. ^'. Hot. (iARi).

\ oi.. 111. Pi.. 48.

MORICOXIA, CARIHJLITIILS, DA.MMARA, SALIX,
(ilKRCUS, CUNNINGIIAMITES, SEQLTOIA.

Bill, X. ^^ i;,,t (iAKu

\<'L 111 Pl.

49-

LAlROPIlVLLrM. LAIRIS. POPILITES.

Bri.i.. X. V. But. Gaki

Vol. Ill Pi.. 50.

AXDROMKDA. RIIAMMS, L.MRTS.

Bull. N. \'. Hor. Gard.

\ (.1.. HI. Pl. 51.

Ql ERCUS. SALIX, PROTEOIDES.

Bii.i.. N. V. BoT. Gakd

\nl.. HI. Pi..

EUCALYPTUS, SALIX. FICUS,
MVRSINE, LAURUS, QLERCUS.

Bl 1,1. N. \' r.oT. (iAKI)

\()i.. III. Pi.. 53.

FICUS. AR.MJA. ElCALVPTrs. M.\( ,N( )I.I A.

c

Hi LI.. X. V. BoT. (i AKi

Vol.. III. Pi.. 57

.\RAIJ.\. M.\(,X()LI.\, 1)I-:\V.\L(4JEA

QE924.B4165 1903 gen

Berry, Edward Wilbe/Flora of the Matawan

3 5185 00094 7141