Tables of Contents:
Volume I
Volume III
Volume II
Atlas    Index
All Four Volumes:
Scans and Webpage
©George Langford III, 2011


flat membranaceous margins or wings CARDIOCARPUS ELONGATUS Rhabdocarpus mamillatus


Stems slender, leaves narrow, linear, subcoriaceous, of various length, forking, or dividing in filaments in the upper part, marked with a few thick primary nerves and intermediate nervilles more or less immersed into the epidermis.

These leaves, as remarked by Grand'Eury, are inserted around small branches, upon tumescent small bolsters, whose disposition is in regular spiral order, with a rhomboidal shape recalling that of the scars of
Lepidodendron, but formed by the fleshy base of laterally decurring leaves like those of Conifers. The coaly envelope of the branches is thick, the foliaceous bolsters are soon effaced upon it as if the bark had increased in thickness in contact with a ligneous increasing body, as in the dicotyledonous.

DICRANOPHYLLUM DICHOTOMUM, Sp. nov., Plate LXXXVII, Figs. 9, 9a, (this volume.)

Stems slender, dichotomous; leaves erect, mostly at the top of the branches, fiat, nerved and undivided at the base, forked upwards or disjoined into thin filaments fixed together or loose; stem of the same character as described for the genus.

Except the dichotomous divisions of the branches the generic description resumed from the remarks of the French author is in entire coincidence with the characters of this plant. The leaves, crowded at the top of the branches, measure at base a little more than two millimeters in diameter, their length varying from five to ten centimeters. The primary nerves, three to four, are distinct and thick near the base, with three or four intermediate veinlets,
this volume f. 9a, more or less immersed into the epidermis. They are soon effaced upwards, either diverging or subdivided in thin filaments in such a way that from the middle of the leaves upward the surface is either very thinly lineate with indistinct veinlets, or divided into one or more laciniae, bundles of filaments of the veins. The stem is slender, four millimeters at the base, the branches slightly narrower, three millimeters, the leaf-scars very distinct, tumescent, of the same character as those of Cordaites costatus, only smaller, their base inflated and decurring forming irregular coarse striae upon a thick cortex, which is generally, like the whole stem, transformed into hard, shining coal.

This plant has some of its characters different from those of
Cordaites, but it is merely a form of the costate Cordaites, which might be separated as a genus, but which, from our specimens at least, cannot represent a different order.

Habitat—Cannelton, Penn'a.  Mr. I. F. Mansfield.

DICRANOPHYLLUM DIMORPHUM, Lesqx., Plate LXXXIII, Figs. 1, 2, (3?).

Proc. Am. Phil. soc., 1878, p. 329, Pl. LIV, f. 1, 2, (31).

Stem or branches small, the largest seen not quite two centimeters wide (flattened), apparently articulate at the point of divergence of the branches and there slightly narrowed, covered with a coaly bark about half a millimeter thick; stem leaves oblique or in right angle to the branches, narrow, linear, lacinate at the top; ncrves distinct under thc epidermis, four or five primary ones near the base, unequally distant; intermediate veinlets indistinct, not perceivable with the glass; top leaves broader with the facies and characters of leaves of Cordaites.

The top of the stem,
Atlas f. 2, which is not figured, bears a reniform scar like those of Cordaites costatus; it is probably that of a top leaf like those of Atlas, f. 1.

The stem,
Atlas f. 1, seems like articulate by a depression across its whole diameter at the point of attachment of the branch, Atlas f. la. The top of this branch terminates abruptly in an obtuse apex to which is attached a somewhat thick leaf like that of Cordaites, with parallel nervation. The top of the main stem bears a tuft of three leaves of the same character, flat, linear, one centimeter broad. The two on the left side are somewhat thick, their veins are immersed into the epidermis; that on the right side is represented by the impression of its lower surface, with primary nerves distinct to the eye, three in one millimeter, with two or three distinct intermediate veinlets. The other leaves attached along the stem are Dicranophylloid, with the nerves rendered more or less obsolete by the thick epidermis. They are of the same character as those of f. 2, which appear to be borne upon a tumescent base. We have here, therefore, in the abrupt termination of the lateral branch, Atlas f. 1, and the large leaves at the top of the main branch, the characters of Cordaites, while the stem leaves have evidently those of Dicranophyllum.

The ramifications of
Atlas, f.1, confirms the observations of Grand'Eury—that some of the branches are very long and simple; that others bear a single isolated branch, while others have them by two opposite, and even it seems sometimes many branches are placed nearly at the same level as in the Conifers. This variation explains the dichotomous ramification of the former species.

One of our specimens,
Atlas f. 3, represents a small seed, oval and similar in form to the ovules which Grand'Eury has represented in the axils of the leaves of Dicranopyhllum, Pl. XXX, f. 3, of his flora. The seed is somewhat larger, with flattened borders and of a thick texture; at least its surface is covered by a pellicle of coal as thick as that upon the leaves of Dycranophyllum. Under it, or as enclosing it in its axil, is a leaf of Dycranophyllum, four millimeters broad, soon splitting twice and separating in three narrow branches hamulose in their curve, and dividing again in filiform laciniae at their extremities. The character of nervation, four primary distinct nerves in one of the laciniae, as seen where the thick epidermis is destroyed, are exactly the same as in the leaves, Atlas f. 2. Hence I believe that we have here positive evidence that these organs are referable to the genus Dicranophyllum as established by the celebrated French author, and also that this genus is related to the order of the Cordaiteae. The ovule at the base of the leaf may be there by casual occurrence.

Habitat—Cannelton, Penn'a. Mr. I. F. Mansfield.


Stems slender; leaves narrow, sublinear, gradually enlargcd from the base, single and sparse, or joined three or four together and fasciculate at the base; surface of stem and leaves irregularly ribbed lengthwise by prominent large bundles of nerves buried under the epidermis which is thick, irregularly granulose, by splitting of the coaly layer.

From the coincidence of characters in the surface of the leaves, I was inclined to consider this peculiar branch as referable to the
Taeniophyllae. It, however, greatly differs by the agglomeration at their base of some of the leaves coming out in bundles from a common button-like point of the stem which there appears irregularly articulate. Some of these leaves are separate and joined single by a semi-lunar base to the stem. But the base of the inferior leaves is not perfectly distinct and appears rather truncate than semi-lunar and embracing. The point of attachment of the leaves in bundles is, however, clear. Therefore, the lower leaves may have been separated from a common point and scattered along the stem, where by compression they seem joined to it. The round points showing scars of bundles of leaves are seen all along the stem and at equal distance from each other, even to the very base, and, therefore, the separate distribution of the others in the intervals seems anomalous. On this subject Prof. Heer remarks that the disposition of these leaves has an affinity to that of the leaves of Salisburia, which, generally attached upon short button-like branches, and then close to each other at base, or like fasciculate, are also often scattered single upon the branches.


Proc. Am. Phil. Soc., p. 333, Pl. LIII, f. 1.

Specific characters same as those of the genus.

The figure is an exact representation of the specimen as far as it can be observed. The stem, a little more than one centimeter thick and flattened, seems to have been, if not fistulose, at least soft and flexible. Its surface has the same appearance as that of the leaves, the epidermis being of the same texture, and the bundles of nerves being also indistinctly discernible by the irregular vertical ridges, or the more or less obscure and always obtuse wrinkles. The leaves appear long; none of them is preserved entire. They are sessile, two or three millimeters broad at their point of attachment to a circular scar, gradually and equally enlarging upwards to about one centimeter at the point where they are broken. To the naked eye, the leaves and stems appear smooth, rather shining, but with the glass, the surface is seen rugulose.

Habitat—Same as the former. Mr. I. F. Mansfield.


Stems or branches of large size, tapering up to a conical point; bark thin, covered with leafy scales; leaves of various size, sub-linear, narrowed or enlarged to the point of attachment, forking or dividing upward in two or more laciniae; nervation distinct with the glass only; primary nerves parallel, about three in two millimeters, buried in the epidermis, generally inflated or half round, intermediate veinlets very thin, distinct upon the decorticated face.


    Proc. Am. Phil. Soc., p. 334, Pl. LIV, f. 5; Pl. LV, f. 1-1a.

Schizopteris anomala? Brgt., Hist. d. veg. foss., p. 384, Pl. CXXXV.

Specific characters same as for the genus.

The fragment of stem figured is six centimeters broad at the base, conical at the top. The bark is a thin coating of coaly matter covered with sparse, distinct, foliaceous, oblong-lanceolate pointed or acuminate scales, marked near the base by a short inflation like a midrib.

The lower surface of the stem, where this thin bark is destroyed, shows round scars of various size, from one to two millimeters in diameter, irregularly distributed, which represent either the base of the scales or that of small leaves. Short small leaves, narrowed to the base, are attached upon the stem, mixed with the scales. On the borders, the leaves are more enlarged at base, a few only of them are narrowed, others are seemingly broken and compressed upon the stem, others still are enlarged as mere diverging and detached fragments of the stem. They vary in diameter from three to ten millimeters, divide by an anomalous forking in acute sinusses, either from near the base or from above, and are marked lengthwise by parallel equal and equally distant primary nerves and very thin intermediate veins.

I consider the species as identical with that of Brongniart described in great detail, l.c., especially from the figure of Gein., Verst., Pl. XXVI, f. 2, which has the divisions of the leaves of this species somewhat broader than figured by Brongniart. In my specimens the leaves are still broader. I must say, however, that in another specimen in my possession, which is like the top of the fig., Atlas,
Plate LXXXIV, the stem, whose scars of scales are marked upon the bark in elevated round points, bears, mixed with the scales, linear leaves as narrow as those represented by Brongniart. Though there may be some doubt of the specific identity between the American and the European plants, they evidently belong to the same group and are referable to the same genus.

Specimens of
Schizopteris anomala are very rare. After Brongniart none have been found, or at least described, but that of Geinitz. Brongniart in considering his species admits it as probably referable to Ferns. Geinitz joins it to Aphlebia, Presl., Rhacophyllum, Schp., a genus which, as seen from the species described in this flora, is a compound of mixed types whose affinity is not positively ascertained, and which Schimper considers as representing primary fronds of Ferns. From this genus this species is positively removed, not only by its peculiar stem, but by the character of its ribbon like equally nerved leaves. On the true relation of the plant to any of the present time, I can say nothing. Like Cordaites it has some analogy to the Cycadeae, for it seems evident, from the shape and the nervation of its leaves, that it is related to the Cordaites.

The stem is of a peculiar character. It appears to have been rather of a soft than of a hard texture. The bark is so thin that, after erosion, some of the scales and young leaves are left attached to the lower surface of the stem, as seen in the upper part of
Plate LXXXIV. On another side, large leaves, especially seen upon my specimens, are decurring at the base along the stem, and join it by a division of their borders, or come to it in a more or less open angle of divergence without any diminution of their width and without apparent separation at their point of union, just as if they were lacineae split from the stem. The epidermis of the leaves is also thin; the primary nerves are half buried into it, and then appear distant as in f. 2, of Brongniart; but under the epidermis the primary veins are less discernible, sometimes totally unobservable, the intermediate ones, very thin veinlets, covering the whole surface.

Habitat—Communicated by Dr. J. H. Britts from the Clinton coal of Missouri.


The number of fruits represented, Plate LXXXV, and the diversity of their characters, may give an idea of the great variety and richness of the vegetation of the coal. And yet this plate has merely a small part of the seeds known until now from the Carboniferous, and they refer only to a class of vegetables, the Gymnosperms, scantily represented in the coal flora by leaves and stems.

The fossil seeds of the American Carboniferous are generally found flattened by compression in beds of shale, or preserved in their original shape in ferruginous clay, or in sandstone, where their whole texture is transformed into amorphous homogeneous compounds. Of course no analysis of the texture of the fruits preserved in that manner can be made. The characters considered for their classification and determination are merely taken from the outside shape of their teguments; and as these are often composed of divers superposed layers, Sarcotesta or fleshy envelope, Pericarp, Endocarp, or inner and outer Testa, easily separated from each other or partly destroyed or flattened in the process of maceration; and also as the surface of each of the layers is of a different character, the determination of these fruits is subject to a great degree of uncertainty.

In France, Grand'Eury has lately discovered in beds of conglomerate sandstone of the Upper coal measures of St. Etienne, a quantity of silicified vegetable remains, among them a large number of seeds, which have been anatomically prepared for microscopical examination by Mr. Renault, of the Museum of Paris, and determined by Prof. Brongniart. In his memoir on the subject,*

* Le grainer fossils, etc. Ann. d. Sci. Nat., Bot., Ser. V, v. XX.

the celebrated Professor recognizes seventeen genera, and twenty-four species of seeds, all from the same formation in the St. Etienne and the Giers coal basin, which occupies an intermediate place between the Middle Carboniferous of France and the Permian.

These seeds are divided into two essential groups:

Seeds more or less compressed, bicarinate, composed of two symmetrical parts. [Cardiocarpus, Rhabdocarpus, Diplotesta, Sarcotaxus, Taxospermum, etc.]
Seeds divided in three, six, eight segments, radiating from a central axis, circular by horizontal cross-section. [Stephanospermum, Trigonocarpus, Codonospermum, etc.]

To the 1st group the author refers the genera Cardiocarpus, Rhabdocarpus, Diplotesta, Sarcotaxus, Taxospermum, etc. To the 2nd, Stephanospermum, Trigonocarpus, Codonospermum, etc.

Other subdivisions are still considered in the memoir. It is evident that with the materials we have at hand and able to consider merely, for their determination, the external character of the seeds, it would be impossible or at least confusing, to admit here the above classification. It is rather advisable to follow, for the present, the traces marked by the exploratory who have found the fruits in the same state of preservation as we have them, and who have described them from their outside characters.

The classification of the fruits in four generic divisions,
Cardiocarpus, Rhabdocarpus, Trigonocarpus, and Carpolithes, is that of Sternberg, admitted also in the first works of Brongniart. It has been generally followed until now, and with few exceptions, which are remarked in their descriptions, all the seeds known from the American coal measures enter into this systematic arrangement.


Seeds of various shape, composed of a compressed, generally cordiform or oval nucleus, surrounded by a flattened, fibrous border, or a membranaceous wing.

This definition is not that of Brongniart, who, in the Prodromus, describes the seeds of this genus as cordiform, compressed, short pedicillate at the emarginate base, acute or acuminate, with a hard pericarp or a fleshy perisperm. In his memoirs on silicified fossil seeds, quoted above, the author reviewing the characters of these fruits, recognizes in them, from anatomical and microscopical analysis, two different groups.

Cardiocarpus (Sclerotesta)—Seeds with a very hard pericarp clearly limited on the outside.
Cardiocarpus (drapaceus)—Seeds whose pericarp, a very dense and hard zone at its internal surface, gradually passes outside to a soft tissue composed of large transparent cells.
He however recognizes the generic affinity of these two groups, as he has observed seeds, with a compact testa homogeneous in its whole thickness, though covered by a thin stratum of transparent cellular tissues.

In the seeds described here as
Cardiocarpus, we have also two groups or sections which probably represent different generic divisions. One for species whose nucleus is superposed to or enclosed into a kind of Samara or thin membranaceous scale, like the seeds of some Conifers.

The other for seeds bordered by a generally narrower margin, which seems more compact or composed of a fibrous tissue.

In the first section should be placed the seeds f. 38, 41, 45-51, of Atlas,
Plate LXXXV, under the name of Samaropsis, Goepp., a genus admitted by Weiss, Dawson, Grand'Eury, etc. But if we consider the transitional forms between the seeds quoted above and those with narrow compact borders, we find Atlas, f. 32, 39, 43, and a number of others not figured, which, evidently alate, partake of the characters of Samaropsis, though the borders are not quite as broad, and apparently rather fibrous than membranaceous. In these, the nucleus is certainly of a more compact texture than the borders, generally convex, perfectly distinct in outline as in the species of Samaropsis. Preserving, therefore, the genus in its integrity, I have separated it in two groups, according to the differences remarked above.

§ 1. Species with flat membranaceous margins or wings.


Ann. of Sci. of Clevel., 1853, 1, p. 152, f. 1.

Geol. Rept. of Ohio, Paleont., I, p. 375, Pl. XLIII, f. 11, 11a.

Nucleus round-cordate; base marked with the cicatrice of the point of attachment; wings emerging from, near the base, gradually enlarging upwards to above the middle, rounded at the apex far above the nucleus, the inside borders inclining towards the top of the fruit where they are joined.

This species is the most remarkable of the genus. The nucleus is nearly round, two centimeters in diameter both ways, slightly striate; the wings, coming out from near the inflated point of attachment, are oval, oblique to the nut, three and a half centimeters long, nearly two centimeters broad in the middle with the inside border joined at the apex of the nucleus, appearing like two open wings of the fruit.

Habitat—Shale over coal No. 1 of the Ohio section, Talmadge, Ohio.


Geol. Rept. of Ohio, Paleont., II, p. 425, Pl. XLVI, f. 2.

Nucleus somewhat heart-shaped, pointed at top, obtuse at the base; lower part distinctly punctate; wings large, broader on the sides, rounded at top and joining at the apex of the nucleus, split or emarginate at base, horizontally striate.

This description is made from the figure. The author says that the nut is covered with a thin epidermal coating, which, when removed, shows underneath the smooth body of a nut or seed, with vertical striae toward the apex. This epidermis is covered with irregularly placed dots. The width of the whole, both nucleus and wings, is about four centimeters; that of the nucleus seventeen millimeters, the wings on the sides being twelve millimeters broad, narrowed to four at the base where they are split or divided as to a point of attachment of the nucleus to the stem.

This seed is unlike the many forms figured by Dr. Newberry. It resembles in its wings
Cardiocarpus Bayleyi, Daws., from the Devonian of New Brunswick; but the nucleus is wider and more acuminate at the apex.

Habitat—Shale in the sub-conglomerate Waverly sandstone, Perry county, Ohio, with
Megalopteris, etc.
CARDIOCARPUS INGENS, Lesqx., Plate LXXXV, Figs. 34, 35.

Geol. Rept. of Arks., II, p. 311, Pl. 4, f. 4a.

Schp., Paleont. Veget., II, p. 223.

Nucleus cordiform, narrowed at the apex into a micropyle passing up to the sinus of the wing; wing broad, fotlowing the outlines of the ovule, somewhat larger toward the apex, where it is deeply cut into a narrow sinus.

The nucleus is two to two and a half centimeters broad, and about two centimeters in length; the wings, five millimeters broad in the middle, narrow at the emarginate base point of attachment, are gradually broader up to the acuminate apex, where the parallel borders are joined in a narrow obtuse sinus descending to near the top of the nucleus or to the upper end of the micropyle.

Habitat—Male's coal bed, sub-conglomerate, Arkansas.


Lesqx., Geol. Rept. of Arks., II, p. 311, Pl. IV, f. 5.

Schp., Paleont. Veget., II, p. 224.

Seed much smaller than those of the former species, broadly ovate-cordate, rapidly acuminate, marked at the base by a tumescent large dot like a chalaza, surrounded by a narrow ring.

The nucleus is one centimeter in diameter both ways, rounded at the lower part, joined to a tumescent mamilla from which appear to originate three fasciles of vessels, the middle straight up, the lateral ones diverging, following the borders and effaced in the middle of the seed. Under the apex, the nucleus bears two diverging appendages like the base of a tubular micropyle. The margin, a little more than one millimeter broad, follows the outline of the nucleus. As the top is broken, it is not possible to see if it is split, emarginate or connate. The details of the internal structure of the fruit are nearly as clearly exposed by the splitting of the seed as it would be if its compound was silicified.

Habitat—Same locality; found upon the same piece of shale as the former.

CARDIOCARPUS ANNULATUS, Newby., Plate LXXXV, Figs. 36, 37.

Newby., Ann. of Sci. of Clevel., 1.c., p. 152, f. 2.

Geol. Rept. of Ohio, Paleont., p. 374, Pl. XLIII, f. 8, 8 a.

Nucleus heart shaped or short ovoid-acuminate, faintly striate, marked at base by the cicatrice of the pedicel, surrounded by an annular border slightly emarginate at the summit.

The nucleus is fourteen to fifteen millimeters in diameter, subtruncate at base, rounded up to a short acumen; the wing, five millimeters broad, slightly narrower in the lower part, follows the outlines of the nucleus and is cut at the top into a short obtuse or nearly round sinus.

Habitat—Shale over coal. No. 1, Youngstown, Ohio.


Seeds of medium size, nucleus subcordate or round, sub-truncate at base, rounded up to a short acumen base of the tube of a micropyle; borders flattened like wings, prolonged upward and connivent to the orifice of the tube, and also prolonged downward to the point of attachment.

The nucleus varies, on different specimens examined, from fifteen to eighteen millimeters long and from eleven to thirteen millimeters broad; the margin, either a flattened testa or a wing is five millimeters on the sides, prolonged upward to eight millimeters above the apex of the nucleus where both sides are curved down and connivent, leaving at the top a short angular space between them, like the opening or enlarging of a micropyle. The wing is also prolonged to eight millimeters below the base of the nucleus, lanceolate-acuminate to the point of attachment by an inside curve of the borders.

This species has the characters described by Brongniart as illustrative of his genus
Pachytesta, Ann. d. sci. l.c., p. 16, P1. XXII, f. 4. From the description, the testa surrounding the seeds is entirely compact., probably very hard, often broken by compression. In the specimen I have examined, the margin is quite flat like the nucleus, of an apparently fibrous texture, but not broken. It is continuous or like a wing bordering the nucleus all around.

Habitat—This species is represented by six specimens in the collection of Mr. R. D. Lacoe, from the sub-conglomerate ledge of Pittston, Pa.


Lesqx., Geol. Rept. of Ill., IV, p. 493.

Seeds small, alate; nucleus ovate-lanceolate, acuminate, bordered by a large margin in prolonged and narrowed downward into a long pedicel, continuous to the apex of the nucleus, where the two extremities are diverging outward and acuminate or horn shaped.

The nucleus is small, flattened though convex on the surface, seven millimeters long, three millimeters broad. near the rounded base, hence gradually tapering up into a sharp acumen. The wings are formed of a thin somewhat membranaceous substance, from which the nucleus is easily separated. Aside of the seed, the margin which follows its borders is two millimeters broad, prolonged downward by a curve into a narrow pedicel, whose end or point of attachment is seven millimeters below the base of the nucleus. From their point of union at the apex of the ovules, the wings take an outward opposite direction and are rapidly acuminate.

These seeds have a marked affinity to those of some Conifers,
Abieteae or Taxodiaceae, like Taxodium distichum, for example, which has the scales of its seeds with one sided horn, and prolonged below the base of the nucleus though far less than in this fossil seed. Sciadopitys verticillata, Sieb. and Zucc., also, of Japan, has seeds with the nucleus placed in the middle of a scale prolonged downward about half a centimeter, with the borders joining at or above the apex of the ovules and there obtuse. The scale is, therefore, emarginate as in the fossil species, but the borders are obtuse at the apex instead of acuminate.

As the close relation of these seeds with those of some Conifers seemed to authorize their separation under a different general division, I proposed for it the name of
Ptilocarpus, Geol. Rept. of Ill., l.c. But as many other seeds, Samaropsis especially, which were then unknown to me, show a great analogy of characters with this one, its separation is not more appropriate than that of a number of others, as seen in the general remarks on the genus.

Habitat—Found by Rev. H. Herzer, in the shale above the coal of Coshocton, Ohio, and kindly communicated in a number of specimens.


Ann. of Sci., l.c , p. 153, f. 3.

Geol. Rept. of Ohio, Paleont., 1, p. 372, Pt. XLIII, f. 3.

Nucleus broadly ovate, acute, marked with a cicatrice at the subtruncate base, smooth; borders of the same shape, continuous, marginate and narrower at base.

The seed has the same form as that of Atlas,
Plate LXXXV, f. 38, being much larger and more sharply acuminate. The nucleus is fourteen millimeters both ways; the margin, three millimeters broad in the middle of the fruit, enlarges to five millimeters at its acuminate conjoined apex.

I refer to this species a fine specimen No. 344 of the Mansfield collection. It differs in nothing but in the obtuse, not acuminate top of the margin.

Habitat—Roof of coal No. 1, Cuyahoga Falls, Ohio; Cannelton, Penn'a.


Newby., Ann. of Sci., l.c., p. 153, f. 4.

Geol. Rept. of Ohio, Paleont., 1, p. 372, Pl. XLIII, f. 4.

Nucleus smaller but of the same shape as in the former species, with a basilar cicatrice and traversed by an elevated line passing through to the apex of the margin.

The nucleus is six millimeters in diameter, eleven millimeters long; the margin which surrounds it is less than one millimeter broad at the rounded base, three millimeters at the obtusely pointed apex.

Habitat—Bituminous shale below coal 1, Cuyahoga Falls, Ohio.


Newby., Ann. of Sci., l.c., p. 153, f. 6.

Geol. Rept. of Ohio, Paleont., 1, p. 373, Pl. XLIII, f. 5.

Seeds small, nucleus ovoia, acuminate, surrounded by a margin which is very narrow at the base, much prolonged beyond the apex of the nucleus, obtuse at the top.

The nucleus is five millimeters in diameter, one centimeter long, narrowed above into a tubular prolongation of the micropyle traversing the margin to the apex. At the base of the nucleus, the margin is extremely narrow, appearing joined to the nucleus as seen in the figure copied from the author.

But for the tubular prolongation of the micropyle, this fruit would be comparable to the winged seeds of

There are in the collections of Mr. Lacoe a number of specimens with the characters of this species. I refer them to it, though they differ by the marginal border continuous and as large at the base as on the sides, only abruptly emarginate to the point of attachment. The margin ascends high above the apex of the nucleus, where it measures five millimeters in width, and there abruptly curves to the apex of the tubular prolongation. It is only one millimeter on the sides.

Habitat—Shale of the coal No. 1, Youngstown, Ohio. Sub-conglomerate ledge of Pittston, Penn'a.


Seeds small, nucleus ovate, acuminate; margin narrower at the base, enlarging up to the apex of the nucleus and there emarginate.

The nucleus is joined by one or two concentrical circles appearing as double envelopes of a small internal ovule of the same shape. In its whole, the seed is from five to seven millimeters in diameter, the margin, one millimeter, or a little broader in the upper part. The apex of the nucleus passes up into a thin tube of a micropyle, which ascends to the head curve of the borders.

Habitat—Sub-conglomerate ledge of Pittston, Mr. R. D. Lacoe.


Ovule smaller, broadly margined, borders conjoining at the apex.

The nucleus is cordiform, abruptly pointed, four millimeters in diameter both ways, continued in a distinct line or micropylar tube upward to the apex where the margins are confluent. The borders, three millimeters broad at the base, four and. an half at the top, follow the outlines of the nucleus.

Habitat—With the former.


Seeds small, round or broadly oval, nucleus ovate, obtuse or emarginate at base, border comparatively broad, marginate at the apex.

The species resembles
Cardiocarpus zonulatus, but differs by the surface quite smooth, the nucleus distinct, without any concentrical zone, ovate, rather obtuse at the top, obscurely marked there with a very small micropyle. The nucleus is three to five millimeters broad and four to six millimeters long. In f. 50, the micropylar tube is indicated by a narrow line between the inside border of the margin which has the same diameter three and a half millimeters all around the nucleus.

Perhaps these three forms, separated as species, represent the same, though the differences appear evident. It will be only possible to judge the variability of these seeds when they are found in connection with their supports, probably enclosed many together in some involucral scales.

Habitat—With the two former species, Mr. R. D. Lacoe.


Newby., Ann. of Sci., l.c., p. 153, f. 5.

Geol. Rept. of Ohio, Paleont., I, p. 374, Pl. XLIII, f. 10.

Seed small, nucleus orbicular, smooth, with a minute scar at base, entirely surrounded by a narrow border.

From the description of the author, the seed is figured overturned, as the small cicatrice is placed at the top where the nucleus is slightly emarginate. It is six millimeters in diameter both ways, the continuous margin only one millimeter.

Habitat—Shale over coal No. 1, Cuyahoga Falls, Ohio.

§ 2. Species with narrow compact margins.


Seed very small, ovoid-acute, margin continuous, equal all around.

This small fruit is only two and a half millimeters long and one and a half broad, inflated or convex on the smooth surface. The very small lenticular nucleus is often seen separate or deprived of its borders, its convex, polished, apparently hard testa raising over the surface of the shale. It thus resembles
Rhabdocarpus minutus, Lesqx., whose testa is, however, ribbed lengthwise.

Habitat—Sub-conglomerate ledge of Pittston, Pa.; Mr. R. D. Lacoe.

CARDIOCARPUS FASCICULATUS, Sp. nov., Plate LXXXV, Figs. 30, 30a.

Seeds of various size, ovate, acuminate, margined; nucleus slightly lineate lengthwise, margin conforming to the nucleus, continuous, or slightly emarginate and bicuspidate at the apex.

These fruits are evidently in racemes, as seen by their position, placed as they are in number and in corresponding direction to branches, with which, however, I have never distinctly seen the point of connection.  F. 30a represents the fruit enlarged. The nucleus is seven millimeters long, scarcely half as broad, smooth, convex, or inflated, marked with some obscure longitudinal lines, bordered with an equal sized flat margin one millimeter broad, notched at the top or broken. It is not possible to see if the notch is casual or natural. The characters of this seed refer it as well to
Cardiocarpus as to Rhabdocarpus.

Habitat—Sub-conglomerate ledge of Pittston; Mr. R. D. Lacoe.


De fruct., p. 23, Pl. II, f. 32.

Seed small, nucleus oval, traversed from the base by a medial line passing up to the emarginate or mammillate apex of the borders.

The seed is represented overturned. The nucleus is exactly oval, four millimeters broad, five long, the margin equal, emerging from the basilar point of attachment, a little enlarged to the emarginate apex, the whole seed being ovate. The authors represent the apex as bicuspidate by the enlarging or dividing of the medial costa. In the specimens examined, some have the costa nearly effaced, and the bicuspidate apex is seen as formed by the acute border of the margin cut near the point or emarginate, as it is generally the case in species of this genus. Except for that medial costa the species is a

Habitat—Same as the former.

CARDIOCARPUS? MAMILLATUS, Lesqx., Plate LXXXV, Figs. 32-33a.

Rhabdocarpus mamillatus, Lesqx., Geol. Rept. of Ill., IV, p. 461, Pl. XXXI, f. 12-15.

Seed small; nucleus oval, mammillate at the base (or apex), regularly and deeply striate, surrounded by a flat margin or flattened testa.

From the specimen,
Atlas f. 33, the species should be referable to a Rhabdocarpus, but specimen, Atlas f. 32, which I consider as a larger form of the same, has a flattened margin which, however, may be merely the fragment of a pericarp imbedded into the stone. The nucleus is oval or nearly round, ten to twelve millimeters long, seven to ten millimeters broad, deeply regularly striate lengthwise, with a small protruding mamilla in the middle of a small round smooth surface at its top.

I should have preserved the original name of this species and described it as a
Rhabdocarpus, if I had seen only the seeds from Illinois. But the Pardee Museum, at Easton, Penn'a, has a number of very fine specimens nearly exactly round, like Atlas, f. 32, some of them covered with a thin pericarp obscuring the striae, others free of it with a margin; and in the collection of Mr. R. D. Lacoe, there is a number of oval specimens slightly larger than Atlas, f. 33, but of the same oval form, all apparently referable to the same species which is intimately related to Cardiocarpus lagenarius, St., Fl. d. Vorw., I, P1. VII, f. 16, or perhaps identical with it. This last figure shows the fruit margined, the border ascending higher than the apex of the nucleus to the orifice of the tube of the micropyle.

Habitat—Concretions of Mazon Creek, the specimen
Atlas, f. 33; Hazleton mines, Atlas, f. 32. The specimens of Mr. Lacoe are from Ontario colliery, Pittston. I am not certain that all the specimens represent one species only.

Carpolithes regularis, St., Fl. d. Vorw., 1, Pl. VII, f. 2.

Carpolithes ellipticus, St., ibid, f. 1.

Seed small, nucleus oval, surrounded by a broad, fleshy ? pericarp, regularly striate lengthwise.

The nucleus is only five millimeters long and half as broad. It is surrounded by a pericarp one millimeter thick on the side, broader toward the base, striate lengthwise, the lines being parallel on the whole seed, and covering the testa. The borders are not, therefore, margined. This seed seems to belong to a same generic division as the former, intermediate to
Cardiocarpus and Rhabdocarpus.

Carpolithes regularis
, St., resembles this fruit by its outline; but the nucleus is not marked upon it. I believe, however, that some of the seed described above represents the same species; for I have on a same plate of shale from Cannelton a large number of seeds of the same general character, with or without a border, all narrowly striate when seen with the glass, some with an oval, convex nucleus, others quite flat; they average five millimeters broad. seven long, and may be referable to both Carpolithes regularis and Carpolithes ellipticus, St.

Habitat—Cannelton, Penn'a. Mr. I. F. Mansfield.


Fl. Carb., p. 236, Pl. XXVI, f. 21.

Seeds small, cordiform, more or less inflated, smooth, without margin.

Seeds of this kind somewhat varied in shape between broadly ovate-obtuse, and cordate-acute, four to eight millimeters in diameter both ways, are not rare upon the shale of the sub-conglomerate ledge of Pittston. I have seen them also on specimens from Cannelton. They resemble detached ovules, like the central parts of
Atlas, f. 46, 49—even 48 and 50, and may represent different species.

Habitat — Cannelton coal; Pittston, sub-conglomerate ledge.


Verst., p. 40, Pl. XXII, f. 24-27.

Carpolithes marginatus, Artis, Antedil. Phytol , Pl. XXII, f. B.

Fruit broadly oval or circular, surrounded by a narrow margin prolonged downward like a small pedicel.

These fruits, twenty to thirty millimeters in diameter, are surrounded by a border one to two millimeters, continuous, and of the same width, only prolonged abruptly downward into a short pedicel two to three millimeters long. The substance of the fruit and of the border is hard, compact; for though flattened, the borders are often broken transversely and the cauda generally destroyed. These alterations render the identification difficult.

Habitat—Seen in many specimens, all upon a piece of shale from Trevorton, Penn'a, low coal.


Carpolithes bicuspidatus, St., Fl. d. Vorw. 1, Pl. VII, f. 8.

Lesqx., Geol. of Penn'a., 1858, p. 877.

Cardiocarpus bicuspidatus, Newby., Geol. Rept. of Ohio., Paleont., I, p., 373, Pl. XLIII, f. 9, 9a.

Seeds of medium size, broadly cordiform, abruptly acuminate, borders narrow, continuous, prolonged downward under the base of the nucleus into a short pedicel.

This species especially differs from the former by its smaller size and the cordiform shape of the nucleus, which generally abruptly acuminate or cuspidate, varies in diameter from one to one and a half centimeters and has an equal margin one and a half to two millimeters broad. As in the former species the downward prolongation of the border into a pedicel is rarely observable but the nucleus and its borders, as in
Atlas, f. 42, are not rare. I have, however, not seen any specimen like Atlas, 43, which seems to represent a different species. It is copied like Atlas, f. 42 from Dr. Newberry, l.c. This species is closely allied to Cardiocrpus Gutbieri, Gein., and has been identified with it by Gutbier.

Habitat—Upper coal strata of Penn'a, Salem vein etc. Roof shale of coal No. 1, Cuyahoga Falls, Ohio; Dr. Newberry.