■^^^si; ^^ /2cd^. /t: Vol. XIV.] [Part II. PROCEEDINGS UK TllK YORKSHIRE GEOLOGICAL AND POLYTECHNIC SOCIETY. Ki.iTHi. I'.v \V. LOWER CARTER, M.A., F.G.S. AM) WILLIAM CASH, F.G.S. IQOl. LEEDS : CHORl.FA' AXn nrKKRSfJITJ., TTfK KI.KCTKIC I'RKSS. 189 TIIK FLORA OF TIIK CAUBONIFEKOUS PERIOD. BY ROBERT KIDSTON, F.R.S.K., F.O.S. ^^^ VORK 80:^,, : • GARDEN. LIBf?ARY riRST rAPER. (,;v/»n i, iOTi.i,,, biven Oy ARTHUR HOLLICL I have pleasure in complying with the request of your Council to read before your Society a short account of the Flora of the Carboniferous Formation, and in so doing shall, as far as possible, avoid technical language, as I address myself more specially to those who, though they have not previously given serious study to the subject, may have a wish to know more about the Fossil Plants which formed such a prominent feature in Carboniferous times, and who, one would fain hope, may be induced to give some attention to a branch of botany than which there is none that would more repay careful observation. There has long been undoubted evidence of the occurrence of Algje and Fungi in Carboniferous times in Britain, and recently I have met with a fossil in rocks of Calciferous Sand- stone age so similar in appearance to Fegatella, that the Liverworts must now be added to our Carboniferous plants. I shall not, however, enter into a detailed description of these fossils, which are of rare occurrence, but pass to those groups which occupy a more prominent place and of which there is more certain knowledge. A fossil which has been referred to the mosses was described from the French Coal Measures by MM. Renault and Zeiller, but hitherto no representative of this class has been met with in Britain. In the present paper we shall therefore reserve our remarks to the Ferns, Fquisetites, and Calamites, leaving the Lycojjods, Sj)henophylls, Cordaitea', and Co7iifer(e for a future time. Before proceeding further, it is necessary to point out that many fossil plant genera are quite provisional, for palaeobotanists have seldom the data for the definition of a genus in the clear E 190 kidston: the flora of the carboniferous period. and full maimer which one demands in the case of genera founded on existing species. Notwithstanding the difficulties of the subject, by careful collecting and study much has been done in elucidating the structure and form of Carboniferous plants, and in some rare cases our knowledge is little less perfect than if we had been able to study the growing plant. Such results have only been attained by much study and careful observa- tion, and are generally the result of the united labours of sevei-al workers — one laying the foundation and another building thereon. Thus the science of palaeozoic botany has grown and, I doubt not, wall grow. I. — Ferns. If we only consider the mere form of the frond and the arrangement of the veins, many fossil ferns have a considerable superficial resemblance to certain recent species ; still this resem- blance must not be regarded as affording any evidence on which to presume a generic relationship. The fact, however, remains that the same type of pinnule form and nervation which is found amongst Carboniferous ferns is seen amongst those exist- ing at present, and also the same mode of circinate vernation (Plate XXVI., fig. 1). In Carboniferous ferns the main rachis sometimes divided into two arms, as in C alymmatotheca bifida L. k, H. sp. (Plate XXV., figs. 2, 3), and this dichotomous division even more frequently occurs in the pinnse, which are once forked, or end in a pair of forks. This character is rare in recent ferns in their native condition, but fretjuent in cultivated forms, resulting in the dichotomous or crested varieties of garden origin. Among Carboniferous ferns the principal families are the Sphenopteridece., JVeuropteridece, and Pecopteridece. These will be briefly described. Sphenopteride-e. Considerable latitude of character is shown by the ferns included in this family. The pinnules may be more or less oval, entire or lobed, and contracted at the base into a short KIDSTOX: THE FLORA OK TIIK CARBONIFEROUS PERIOD. 191 stalk, or cuneate, and even almost tiiiform. Nervation radiating from the base of the pinnule and frequently dichotomising. The cliief genus is Sphenopteris Brongt. Spiienopteris Brongt. The general characters of one of the sections of the ferns commonly included in Sphenopteris is well represented by Sphenopteris obtusiloba Brongt. (Plate XXV., fig. 1). The pinnules are oval entire, lobed, or divided into 3-5 segments — their form varying according to their position on the i)inna. The dichotomising veins radiate fan-like from the base of the pinnules (Plate XXV., fig. la). Anotlier section of Sphenopteris has pinnules with more or less cuneate segments, of which Sphenopteris furcata Brongt. may be taken as a typical representative (Plate XXVII., fig. 2). The segments of the pinnule are narrow, linear, with a veinlet running into each tooth (Plate XXVII. , fig. 2a). In both of these sections there are some species with very small pinnules. Although one must be very careful in generalising, still it seems as if tlie linear or cuneate pinnuled forms were more characteristic of Lower Carboniferous rocks, while those with rounded lobed pinnules were more typical of the Upper Car- boniferous. Both types, however, occur together in all the divisions of the Carboniferous Formation. Many species originally included in Sphenopteris have had special genera provided for their reception. In some cases the characters are derived from the mode of division of the pinnje — characters dependent on the vegetative system. It appears to me very doubtful if any real advantage is derived from the creation of such genera, as they cannot be regarded as other than provisional, and personally I prefer retaining the ferns placed in these genera in Sphenoptej-is. As examples of the genera to which I refer, Palmatopteris Potonie (of which Sphenopteris furcata is the type), and Diplothmema Stur may be mentioned. The other class of genera which have been taken from Sphenopteris Brongt. hold, however, a very different position, as they are founded on characters which are derived from their 192 KIDSTON: TlIK FLOKA OF TIIK CARBONIFEROUS PKHIOD. fructification, but before referring to these more fully it is necessary very sliortly to ('onsidcr the fructification of existing ferns. Recent ferns are divided into two great classes — the Iso- sporous Ferns, or those with one kind of spore, and the Heterosporous Ferns, or those with two kinds of spores — macrospores and microspores. With the latter class, however, we have nothing to do at present. Returning to the Isosporous Ferns, these again form two great sections. Fii'st, those whose sporangia are provided with a prominent ring of cells, called an Anntdus (Plate XXVIII., fig. 2), and those whose sporangia are destitute of this structure. The first section contains the great majority of recent ferns, of which the common Polypody and Male fern may be men- tioned as well-known representatives. The second group con- tains the Marattiacea', which comprises few genera and a small number of species, all of which are natives of more or less tropical areas. In Carboniferous times both these groups are represented, though the exannulate ferns seem to have outnumbered those with annulate sporangia. Let us now return to Sjyhenopteris. Many of the species originally included in that genus have in recent years been found showing their fi-uctification, and for these new genera have been created. Among British Sphenopteroid forms a few are known to possess annulate sporangia, and of such are Corynepteris Baily and Oligocarpia Goppert. In the former the sporangia are placed in groups of five or six, united at the base around a common centre, and collectively form a globular mass or sorus ; in the latter they form little circular heaps composed of a number of independent sporangia. Isolated annulate sporangia are frequent in the Yorkshire and Lancashire " Coal Balls," and also occur in the material from Pettycur, Fife, which is situated in the lowest division of the Carboniferous Formation (Calciferous Sandstone Series). KIDSTOX: THE FLORA OF THE CARBONIFEROUS PERIOD. 193 The exannulate form of fructification is illustrated by several genera, Mhich are characterised by the form and arrangement of the sporangia. Among these may be mentioned RenauUia Zeiller, where the small oval sporangia are situated on the veins towards the margin of the pinnules (Plate XXVIII., fig. 4). They open by a longitudinal cleft. The fruiting pinnules are little modified from those of the barren frond. In Urnatopteris Kidston the barren (Plate XXIX., fig. 1) and fertile (Plate XXIX., ngs. 2 and 3) fronds are dissimilar, that is, only some of the fronds bear sporangia, and on these the pinnules are entirely deprived of the limb — ^the sporangia being arranged in two rows, one on each side of the rachis. The sporangia are pointed-oval, and open at the summit by a small round pore (Plate XXIX., fig. 6). Each sporangium is free, but in their structure they have considerable resemblance to the sporangia of Dmuvxi only in that genus the sporangia are united to each other to form a synangium. Though I only mention these two Sphenopteroid exannulate types, others are known. NEUROPTERlDEiE. The most important genus of this family is Neuropteris Brongt. (Plate XXVIII. , fig. 3. Neuropteris gigantea Sternb.). The pinnules ai'e generally more or less oval or tongue-shaped, and articulated to the rachis, from which they are easily detached. Each pinnule had a central vein, from which are given off lateral divided veinlets (Plate XXIX., fig. 4). On some species of Neuropteris, possibly on the majority, between tlie points of insertion of the lateral pinnse or towards the base of the frond, immediately below the pinnse, the main rachis bore reniform or orbicular pinnules (Plate XXVIII., fig. 3, a, «), sometimes of large size ; these, before their true origin was known, were supposed to belong to a distinct plant, and were named Cyclopteris by Brongniart. The fructification of Neiiropteris is imperfectly known, but in the case of Ne^iropteris heterophylla Brongt. it was borne on long pedicels which terminate the pinnse. 194 KIDSTOX: TIIK FLORA OF TlIK CAHIJONIFKKOUS PERIOD. Linopteris Presl. {Dicfi/opti'rif! Gutbier not Lamoui'oux) tliuugli rare in Britain, must not be omitted. In the form of the frond and pinnules it is similar to Xearopteris, and specimens not showing the nervation might easily be overlooked as belonging to that genu.s, but it is at once distinguished by the nervation, the veinlets of which unite among themselves to form a net-like reti- culation (Plate XXX., fig. 2. Linopteris obliqua Bunbury sp. x 3;\). Though this genus is certainly not common in Britain, it may be more common than supposed through being passed over for Neuropteris. PECOPTERIDEiE. This family holds an important place among pakeozoic ferns. Pecopteris Brongt. is the chief genus and contains many large and fine species. It is chiefly represented in tlie Upper Coal Measures, and Pecopteris arhorescens may be regarded as the type (Plate XXVII., fig. 3). The pinnules in Pecopteris are attached to the rachis by the whole of their base. They have straight sides and rounded apices. The pinnules are sometimes united among themselves at the base and possess a strong central mid-rib, from which — according to the species — are given off simple or dichotomously divided veins which extend to the margin. The fructification of many of these ferns consists of four or five exannulate sporangia arranged in a stellate group, from which circumstance the genus Asterotheca has been proposed for them, though not generally adopted (Plate XXVII.. fig. 4). In the Middle Coal Measures, Peco^iteris [Asterotheca) is represented l)y few species, but Pecopteris (Asterotheca) Miltoni is fairly plentiful. Though this species also occurs in the Upper Coal Measures, it is there associated with many other Pecopterids which are not found below that horizon. The fronds of Pecopteris were of very large size and most jjrobably some of the tree fern stems were the trunks of Pecopteris. The Pec. plAcmosa Artis sp. ( = Pec. dentata Brongt. Plate XXVII., fig. 1, Plate XXXI., figs. 1-4), so common in the Middle and Upper Coal Measures, forms the type of the genus KIDSTOX: TIIK FLORA OF THE CARBONIFEROUS PKRIOD. 195 Dactylotheca Zeiller. This is characterised by the ovoid-pointed sporangia, which are placed singly on the veins and open by a longitudinal cleft (Plate XXXI., fig. 3). The barren pinnules vary greatly in foi'iii, being entire, lobed, or crenate, according to the position they hold on the frond. On the main rachis, at tlie point of insertion of the pinme, are curious, much-divided outgrowths, called aphlebia (Plate XXXI., fig. 2). These were originally supposed to be a climbing fern (Schizopteris adnascens L. ct H.) which had used the frond of Dactylotheca as a sup- port. These aphlebia are an integi'al part of the frond on which they occur, and are found on other species of ferns belonging to various genera. Another Pecopterid genus, Mariopteris Zeiller, is extremely common in the Lower and Middle Coal Measures, but very rare in the Upper Coal Measures. The fructification is unknown, but the fern is distinguished by a double bifurcation of the rachis of the primary pinnae. The leathery texture of the pin- nules, difiicult to describe but easily learnt from an examination of specimens, as well as the nervation, appears to me to add a chai'acter to the genus, which I would be inclined to restrict for Mariojderis (Pecoptei'is) muricata, Schl. sp. (Plate XXXII. , figs. 1 and la) and one or two close allies, but from which I would exclude such species as Sphenopteris latifolia Brongt. The double liifurcation of the primary pinna;, which occurs in this species, does not alone seem to me to be of much systematic importance. Alethopteride^. The Alethopteridece. are closely related to the Pecopteridece, but the pinnules are generally obliquely placed on the rachis, the prominent mid-rib joins the rachis near the upper margin of the piimule, and thus gives a somewhat decurrent character to the mode of their insertion on the I'acliis. The lateral dichotomously divided veins are very numerous and run to the margin at almost right angles with the mid-rib. The common Alethopteris lonchitica Schl. sp. well illustrates these characters (Plate XXXIL, figs. 2 and 3). 19G KinSTOX: TIIK FLORA OF TIIK CARBOXIFEROUS PERIOD. The fructification of Alethopteris is imperfectly known, but what is supposed to be a fruiting specimen of Alethopteris Serlii Brongt. has been described by Zeiller. This most interesting example appears to show that the sporangia were globular and arranged in rows along the veins. The fronds of Alethopteris attained to large size. Tlie genus Lonchopteris Brongt. (Plate XXYL, fig. 2) holds the same I'elationship to Alethopteris tliat Linopteris does to Neuropteris, having the same form of growth and pinnule cutting as Alethopteris, but is easily distinguished at first sight by the net- like reticulation of the veins (Plate XXVI., figs. 2a + 3). Odontopteride^. The only genus of this family to which reference requires to be made is Odontopteris Brongt., which, however, is very rare in British Carboniferous rocks, and appears to be restricted to the Middle and Upper Coal Measures (Plate XXVIII., figs. 1 and la — Odontopteris alpina Presl. sp.). The pinnules are more or less tongue-shaped and attached to the rachis by their broad base. They have no true mid-rib — several veins passing into the pinnules direct from the rachis, where they bifurcate once or twice. Tree Fern Stems. Some of the palfeozoic ferns had stems like our modern Tree Ferns and must have attained to a considerable height. In Britain the two following genera of fern trunks occur : — Caulopteris L. & H. (Plate XXXIII., fig. 1 — Caulopterls auglica Kidston). The frond scars are arranged in vertical rows placed close to each other. They are oval and contain, a short distance within the margin, a closed oval or horse-shoe-shaped band, which corresponds to a tract of sclerenchymatous or much indurated tissue. Within this band and near its upper end is placed the vascular bundle scar. The outer surface of the stem is usually densely clothed with aerial rootlets. Megaphyton Artis. (Plate XXXIII., fig. 2 — Megaphyton sp. allied to M. anornalum Grand' Eury.). The fronds are attached kidston: the flora of the carboniferous period. 197 to these stems in two opposite rows, tlie frond on one side of the stem alternating with tliat on the other side. The stem, excej^t at the part to which the fronds are attached, was densely covered with aerial rootlets. Caulopleris in its general aspect would much resemble one of the recent Tree Ferns, but Megaphyton, with its two opposite rows of fronds would have a very different aspect from any of the Tree Ferns at present existing. Before passing from this brief consideration of the more im- portant groups of paheozoic ferns, a few remarks must be made on their internal organisation, though this subject can only be touched on very slightly here. The stems or rhizomes of recent ferns have no exogenous growth, that is, when the vascular bundle is once fully formed no new elements are subsequently added to it. Hence Tree Fern stems when once fully developed retain the same diameter of trunk for years. Among fossil ferns whose structure is known, a few, generally of small size, possess the same structural peculiarities, but there is another type of palaeozoic fern structure where, among other characters, an exogenous increase to the vascular system takes place. In these, after the formation of the primary vascular bundles, whose size is limited as in the first type, a cambium layer appears from which an outer circle of exogenously developed vascular tissue arises. This ring of secondary xylem or wood may increase indefinitely in size by additions from the cambium zone, the ultimate size of the stem being limited only by the life of the plant. These Fern Stems with exogenous growth present certain anatomical characters intermediate between ferns and Cycads, and are now placed in a group to which Potonie has given the name of C ycadojilices. There is reason to believe, though little is known of their fructification, that they may be ferns, though in their anatomy they possess certain characters not found in existing members of this group. This discovery is one of the most interesting and important advances recently 198 KIDSTON: TUK FLOHA of TIIK CAKHOXIFKKUUS PEUIOU. iiKule ill the study of pahrozoic botany, and to the Cycadofilices are known to belong certain ^phenopteris, Alethopteris, and Xeuropteris. As an example of how step by step our knowledge of paheozoic- l)otany is l)uilt up, it may be mentioned that the petioles described by Williamson as Rachiopte7'is aspera were subsecjueutly found to belong to the stem named Lyginodendron Oldhamium by the same author, and furtlier it has been dis- covered that Lyginodendron Oldhamium is the stem of the well-known Sphenopteris Iloeninghausi Brongt. {Plate XXIX., fig. 5). Could any better example be found of the result — or reward — of patient, plodding work, or of the provisional nature of genera founded on the vegetative organs ? Calamarie^. Equisetites Sternberg. A few fossils have been found in Carboniferous rocks which from their great external resemblance to the recent Equisetum or Horsetails, have been placed in a genus called Equisetites by Sternberg. These fossils are extremely rare, and as far as at present known do not go furtlier back than the Coal Measures. One of the most interesting examples of the genus is the Equisetites Hemingwayi (Plate XXXIV., fig. 3), which was dis- covered by Mr. Hemingway, Barnsley. The cones are oval, about one inch long and rather over half an inch broad. The outer surface of tlie cone is covered with hexagonal scales about one-fifth inch in diameter, with a small central point, indicating probably the place of attachment of the little pedicel by which the peltate shield was united to the axis of the cone. Nothing is known of the inner structure and arrangement of the sporangia, but the external appearance of Equisetites Hemingwayi is so like that of the cones of recent Equisetum (of which a figure is given for comparison, Plate XXXIV., fig. 4, Equisetum hyemale), that the atfinities of Equisetites Jlemingwayi Kidston with Equisetum is probably very close. KIDSTOX: THK FLOKA OF TIIK CAIIHON'IFEIIOUS PERIOD. 199 A specimen in the British ^luseum shows that the cones of Equisetites llemingwayi were apparently sessile and borne at the nodes of a very Equisetum-like stem. MM. llenault and Zeiller have descriy)ed from the Comentry Coal Field an Equisetaceous stem, with distinct sheaths, under the name of E. Moni/i. Some other small specimens from the Carboniferous have been ascribed to Equisetites, but their reference is in manv cases doubtful. Calamites Suckow. The Calamites form one of the most prominent types of vegetation in Coal Measure times. True Calamites tlo occur in Lower Carboniferous rocks, that is, below the Millstone Grit, but there tliey take a very unimportant place and are of very rare occurrence. It is onl}^ when we roach tlie Upper Carbon- iferous that they attain their importance, both in numbers and diversity of form. Calamites reached to arborescent dimensions. When dealing with the Calamites, we are under the necessity of placing the stems, foliage, and fructification of the plants comprised in this group in separate genera, as in few cases can the foliage and fruit he referred to the parent plant. In fact, even in the genus Calainites in which the stems are placed, there are almost certainly included plants which belong to different genera. One is led to infer this from the structure of Calamitic cones which show among themselves important structural differences. This fact must not be lost sight of, and the genus Calamites should be I'egarded more in the light of a group than of a true genus, but for practical purposes some system of classification, even if pro\isional, must be adopted. The late Professor "Weiss divided Calamites into three groups : — I. — Calamitina. In Calamitina the branches are borne in verticils, but between each verticil there is one or more nodes from which no branches are developed. II. — EucALAMiTES. The stems placed in this group bear branches from every node. 200 kidston: tiik flora of tiif, carboxiferous pkriod. in. — Stylocalamitks. Here tlie stems are either unbranclied or, if lateral branches occur, tliey are developed very irregularly. In all these divisions the ribs on the pith-cast alternate at the nodes. There is a fourth division which, however, only occurs in the Lower Carboniferous : — IV. - --Asterocalamites Schimper. In these plants the ribs do not alternate at the nodes, and the branch scars are irregularly produced. Before considering these groups more fully it is desirable to make a few general remarks on the Calamites as a whole. The majority of the fossils referred to Calamites have ribbed exteriors, such as the figures of Calamites ( Stylocalam^ites ) Suckowii given by Brongniart and others (Plate XXXV., fig. 3). These do not represent the exterior of the plant as originally supposed, but are merely the casts of the pith cavity. This is well seen in the figui'e of Calamites (Calamitina) app?-oxim,atus Brongt. given on Plate XXXV., fig. '2, where the cast of the pith cavity is seen at a, and the vascular portion of the stem at b. Plate XXXIV., fig. 2, also shows the same characters. The true outer surface of the stem of Calamites is rarely preserved, and though very young stems may show faint ribs the older stems have almost invariably smooth barks, though on rare occasions a ribbing of the outer surface seems to occur as in some examples of Calamites (Calamitina) verticillatus L. k. H. (Plate XXX VI., fig. 4). The stems of Calamites (except possibly in the very young condition) were hollow except at the nodes, where a more or less complete diaphragm of cellular tissue extended across the cavity. The pith cavity was surrounded by a zone of vascular wedges, in the inner angle of which is a carinal canal. This woody zone increased indefinitely in size by additions from a cambium ring. The vascular wedges are separated by broad medullary rays, anfl the whole is enclosed in a thick cortex. Modifications of this structure occur in different members of the group, but all conform in tlieir outstanding features to this type KIIXSTON': THE FLORA OF THE CARBONIFEROUS PERIOD. 201 of stem. The aerial stems of Calamites spring from creeping rhizomes as seen in the figure of Calamites Suckoivii, given on Plate XXX., fig. 1, as well as from the subterranean portion of the aerial stems (Plate XXXV., fig. 3). Let us now return to the consideration of the three groups of stems, to which reference has just been made : — I. — Calamitina. Calamites varians Sternb. may be taken as t3'pical of this group (Plate XXXIV., fig. 1. Calamitina varians var. inconstans Weiss). The internodes vary in length, and the nodes bear closely- placed transversely oval leaf scars. The bark is smooth but frequently shows slight longitudinal clefts or cracks, which vary in their length and distance apart. These longitudinal cracks or lines probably arise through the splitting of the bark from the increase of the stem in girth. Between each branch-bearing node several bi'anchless nodes intervene, their number varying, not only in the same species, but even on the same specimen. The foliage of some Calamitince consisted of acicular leaves, but whether all possessed such foliage is not known (Plate XXXVI. , fig. 1). II. EUCALAMITES. Calamites ramosus Artis. is representative of this section (Plate XXXVII., figs. 3 and 4). Each node gives rise to two branches, one on each side of the stem. The branches are super- posed, and, though these again bear lateral branches, the plant would possess the form of a triangle. The surface of the stem is smooth. Plate XXXVII., fig. 4, shows the cast of the pith cavity ; fig. 3 shows the outer surface of the species with smooth bark. The foliage of Calam,ites raviosus consisted of lanceolate leaves, arranged in whorls and united by their bases to form a very narrow ring round the stem. This foliage was named Annularia radiata by Brongniart before it was discovered to be the foliage of Calamites ramosus (Plate XXXVII., fig. 1). '20'2 kidston: tiii': p^lora of thk cahbonifkrous period. The fruit of Calamites ramosus is in tlie form of small cones which terminate the branchlets. Their structure is that of CalamostachT/s, which will be presently described. In Calamites [Eucalamites) cruciatus Sternb. each node bore a verticil of somewhat distant branches. III. — Stylocalamites. The Calamites in this group very rarely produced branches. Calamites Suckowii Brongt. is a good example (Plate XXXV., fig. 3). The outer surface of the stem was smooth, and if the nodes gave rise to branches they must have done so very rarely. In Calamites Cistii Brongt., another member of this group^ small scars occasionally are found on the nodes, but these probably are the scars, at least in part, of short stalked cones. IV. — Asterocalamites Schimper. This gi'oup is of generic value, and ranks in importance with the genus Calamites ; it not only differs from Calamites in the ribs not alternating at the nodes, but also in the foliage being dichotomously divided. The fructification consisted of narrow cones, fully five inches long, which are periodically divided into sections by interposed barren whorls, so that the cone appears as if composed of a number of oblong segments resting on each other, and between which is a whorl of leaves. Each segment therefore consisted of a barren whorl, which is succeeded by 10 or 12 fertile whorls. Though specimens of the fruit and foliage are very rare in Britain, fragments of the stems are not uncommon. The genus is characteristic of the Lower Carboniferous. Owing to our inability in the majority of cases of associating the isolated foliage branches of Calamites with the stems to which they belong, they are placed in the two following genera — Calamocladus and Annularia. In Calamocladus Schimper [Asterophyllites Brongt.) the leaves are arranged in whorls. They are narrow linear or setaceous, single nerved and placed closely together. One of the commonest species is Calamocladus equisetiform,is Schl. sp. (Plate XXX., fig. 3). kidston: the floua of the carboniferous period. 203 Annularia Sternberg contains those forms with whorled, single- nerved lanceolate leaves, widest near the centre like Annularia radiata Brongt. (Plate XXXVII., fig. 2), or with spathulate leaves like Annularia sphenophylloides Zenker, sp. (Plate XXXVII., fig. 1). The leaves unite at the base and form a xQvy narrow collar round the stem. Fructification of Calamites. The fructification of Calamites consists of narrow linear cones, attaining in some species a few inches in length, though in most cases they are of smaller size. The arrangement of the sporangia in many of these cones is still unknown, but of some a \ery complete knowledge is possessed. For their reception several genera have been founded, as hitherto it has been generally im- possible to refer them to their parent stems. The more important of these genera may be briefly described. Calamostacuys Schimper. (Plate XXXVI. , fig. 2). The cone is composed of alternating whorls of barren leaves or bracts and sporangiferous scales. The basal portions of the bracts unite to form an almost horizontal collar which surrounds the axis, while the free parts of the bracts rise up almost at a right angle, the whole forming a saucer-like structure. Between each of these barren whorls is a fertile whorl. This consists of slender pedicels or sporanglophores, which spring from the axis at riglit angles and terminate in peltate shields, on the inner surface of which are borne four sporangia. Both homosporus and heterosporous cones occur in Calamostachi/s. Pal.«ostaciiya Weiss (Plate XXXIV., fig. 5; Plate XXXVI., fig. 3. Palceostachya pedunculata Williamson). The cones placed in Faloiostachya differ from those of Calamostachys in the sporangiophores springing from the axis immediately above the axils of the bracts and forming with the axis an angle of about 45 degrees. For cones possessing the general appearance of Calamostachys and Palceostachya, but in which the arrangement and position of the fertile whorls are unknown, the genus Volkmannia Sternb. may be conveniently employed. 204 kidston: tiik kloka of thk cauhoxifekous period. Macroiitdchija Scliiniper is anotlier genus of Calamitic cones. These attained to considerable size, and are much larger and broader than those of the three preceding genera. The cones are composed of alternating closely placed verticils of many bracts, united to each other tlirougliout the greater portion of their length ; only the short upturned extremities of the bracts remain free. Each whorl of bracts thus forms a saucer-like collar which surrounds the thick axis of the cone. The an-angement of the sporangiophores has not been clearly made out. Other types of Calamitic cones are known, but those mentioned are the principal forms which occur in British Carboniferous rocks. Occasionally specimens of Calamites are found showing the remains of their rootlets. These are — in whole or in part — the fossils for which Lindley and Hutton founded the genus Pinnularia (Plate XXXV., fig. 1 — Pinnularia columnaris). They consist of roots pinnately giving off lateral roots, which in turn bear the rootlets, apparently in the same plane. That the Calamariece and Equisetacece are closely related is beyond all doubt, and there seems to be no satisfactory reason why they should not be united in one family under either of these names, preferably under that first mentioned. The genus Calamites seems to have entirely disappeared with- out leaving any modern representative, while the less important palaeozoic genus Equisetites is probably the ancestor of the recent Eq7iisetum. KIDSTON : TIIK Kl.nHA OK TIIK CA ItHDMFKKOl'S I'KUIdD. 205 ExiM.ANATlDN OF Pl-ATKS. Platk XXV. l-'ii,'. I. Sp/irnnpfrri.f <>/)fi/fii/(>h(i lU'otiirt. Grall^e Colliery, Kil- iiiai-iiock. T.owei- Coal .Measures [15G0]. Tlirce-tifths natural size. Fi^. ]((. Sjiht'iinpti'r'Di obtusilnhd BroiiLjt. Portion of a pinna sliyhtlv enlari^ed to show tlie nervation of the j)innules. S]>eeinien i'eeei\'e(l from the I5ev. I ). Landshoi'ouiili, D.I). \^y()fi'. — Tlie fi.i,'ui'es enclosed in brackets give the i-e^istration nunibers of the specimens in the Authors coUecrion. ] Fig. :,'. Ciil iinnintlofht'cd liijidd L. it H. sp. Lewis Bui'ii, about ■_'()(! vards belnw Lewis Purii Colliery, North Tynedale, Northumberland. Calciferuus Sandstone )Serie.s (Lower Cai-boiiifei-ous) [7l'S]. Tliree-fiftlis natural size. Basal jiortion of fi-ond showing bifurcation of rachis. J''ig. 'A. C<(l ;/)ii)n(it(ithfc<( hij'uln L. it H. sp. Purdiehouse, Mid- lothian. Calciferous Sandstone Series [~1T]. Collected by the late Mr. C. W. Peach. Three-fifths natural size. Cpper portion of frond showing bifui'cation and j)innte. i'"ig. 3rt. Pintnde eidarired. a- KIDSION : Till-: FLORA OK TllK t'AHUONl FEIIOIS I'KHIOD. *207 Platk XXV L Viis. I. S/>ir(>/i(/'ris. Braysdown, near Radstock, Somerset. Upper Coal Measures [510]. Youiiijj frond, probably of Fecop- teris in circinatc vernation. Kather less than half natui'al size. Fiu. -. Loiirlnijitcris ruyosa. Brongt. St. P^loi, Mariemont, Belgium. Coal Measures. Natural size [2634]. Specimen communicated by Rev. Father G. Schmitz, S.J., Louvain. Fig. '1(1. Portion of pinnule showing tlie reticulate nervation. Magniiicd about '!}, times. KIIISTON : TlIK FLORA ill' IlIK CAHHCiN I KKKOUS riCHlOl). '-'O'J I'LATi: XXV J I. Fig. 1. I>iicti/liitliic(i /t//tnifis(t Artis. sp. Mcmcktoii Main Colliery, near IJai'iislcv. M iddlc Coal Measures [2107]. Natural size. Collected 1)V Mr. W. Jleiiiiiigway. Fig. 1'/. Pinnule enlai-ged foui' times. Fig. '2. Sphf'iiojifrris j'nrctita l'>rongt. Ci'amlingtoii, Northumber- land. Lowei- Coal Measures [259]. Natural size. Collected by .AIi-. .1. Sim. Fig. '1(1. Pinnule enlai'ged to show nervation. Fig. 3. Pecojtti'rix arlioresrens Schloth. sp. liadstock, Somerset. Natui'al size. Upper Coal Measures [-^-^'i]. Fig. '^u. Pinnule enlai-ged to show nervation. Fig. t. Pecopteris [Axterotheca) M'lltoni Artis. sp. (after Zeiller). Enlai'ired twice. KIDSI'ON : llIK FLOHA OF THK CA l{ni >N I FKHOl'S I'KHIOD. 1*11 ri.ATi: XX VII r. I''ii;". 1. 0(i/>f''ris (djiiud Prcsl sp. ^loiickloii Main C'olliei'v, near ISai'iislt'V. Middle Coal Measures [1902]. Tlii'ee- tiftlis natural size. Collected by .Mr. W. Heiniuifwav. V'\\l. \. Xi'tn-ojifi'i-iK (/uidntpa iSternb. Hill Top Colliery, Skegby, near Hiechnael-undei'-Hutliwaite, Notts. [20G]. Three- tifths natural size. Middle Coal Measures. Collected by ]Mr. E. Wilson. On tlie main rachis, as at r?, are seen the small cyclopteroid pinnules. Fig. 4. Iif ltd III fill iiiicrociirpa Lestpx. Plairpoint, near Dysart, Fife. Lower Coal Measures [773]. Collected by ]VIr. James Bennie. Pinnule showing the fructification enlarged. Fig. ^a. Sporangium more highly enlarged to show the structure. >i 05 KID.STON : THK FLORA OF THK CAKUOXIFEROUS PERIOD. - 1 .'5 Plate XXIX. Fig. 1. U}')iafi>jit>'ris ti'ui'Jln Vyv^m'^t. sp. Furnace lluiik, S;iuchie, near Alloa, ClackinaiiiiaTishire. Lower Coal MtMisures [1983]. Natural size. Portion of barren frond. i'^JL,'. -. Uriiatopteris tenella Brongt. sp. Furnace Bank, Sauchie, near Alloa, Clackmannanshire. Lower Coal Measures [1988]. Natural size. Fruiting frond. l'"ig. .'3. r nitif(tj)tn-is tt'iu'lla Prongt. sp. Ellisinuii', Baillicston, Laiiarkshii'c. Power Coal Measures [2450]. Enlarged about twice. Collected by Mr. P. Jack. Fi^. 1. .\ rii ro/ifrrts (jhjdiifea Sternb. Coseley, near r)udley. .Middle Coal Measures [212]. Pinnule enlarged to show nervation. I'^ig. 5. Sp/ifHopfffis I/ittii ii(//iKitsi Brongt. Tullvgarth, near Clackmannan. ■ Lower Coal Measures [938]. Natural size. Fig. Orf. Pinnule eidarged [930]. Fig. f). J'rnatojiteris tenelUt Brongt. sp. Furnace Bank, Sauchie, near Alloa, Clackmannanshire. A few sporangia en- larged and showing terminal pore [1970]. ii.tcuiapli.-.l l.y K. Ki.Kt I'roc. Yorts. Col. n„ol,/t,r. Sor., Vol. XIV., Piatt XXIX. KIDSTON: THE FLORA OF THE CARBONIFEROUS PERIOD. I'lo Plate XXX. Fiir. 1. Calaniites Sucko>vii Brongt. Ellismuii-, Ballieston, Lanark- .sliire. Lower Coal Measures. Four-fifths natural size. Towards the centre of the .specimen the rhizome gives rise to three aerial stems. At the lower end of the speoiiiicn another .stem is given off. From the dii'ection in which the stems l)eiid, it is apparently the under surface of the rlii/.oine which is exIiihitecL Collected by Mr. P. Jack. Fig. '2. Linopferis olil'uiiut Biinhury s[). Pittston, Pa.. L\.S.A. Specimen i-eceived from the late Mr. W. I). Lacoe [1348]. Pitumlc cidargcd about 3.', times to show tiie nervation. Fig. .3. Calavwcladus equi,sptiJ'ormis Schl. sp. Cadeby Collierv, Conisborough, Yorkshire. Middle Coal Measures [1536]. Collected by Mr. W. Hemingway. 'hotosraplu.M.v K. Ki,lst.,M J>,o,. Yorkx. (Icol. >ni(l. Pohjtcc Sac, Vof. A' /I'., /'/n/i^ XXX. KIDSTON : THE FLOHA OF TlIK (A lUJOMFEROl'S PKKIOI). Plate XXXI. Fii:. 1. Dactylotheca pluvumi Artis. sp. Monckton Main Colliery, near Earnsley. Middle Coal Measures [210-")]. Natural size. Colleeted hy .Mr. W. Heinin'f<( j)hn)i(is(( Artis. sp. Monckton Main Colliery, near Barnsley. Middle Coal Measures [2008]. Piniiule sho\vini>; sporangia x S. Collected by Mr. W. Hcniiny- Wil V. Y\ii. 3r?. Sp(irani,duni x 25. Fi- iv/"- •' r KIDSTOX : TIIK FI.OKA oK TIIK ('A Klii >NIKKKOU.S I'KlJKiU. Platk XXXIII. Fig. 1. C'aulopteris cyclostiyma Lesqx. Braysdowii Colliery, Had- stock, Somerset. Upper Coal Measures [i'72]. Three- fifths natural size. a. Vascular sear contained within the sclerenchymatous band b of the frond scar. Fig. 2. Mi'(ja])h]jto}i sp. allied to Meij((iih!/ti>ii (nmnialani Grand Euiy. WooUey Colliery, Darton, near Barnsley. Middle Coal Measures [2159]. Three-fifths natural size. Collected by Mr. W. Hemingway. Portion of stem showinij one of the i-ows of frond scars. KIDSTON : THE FLORA OF XriK CAKBONIFKUOIS )'i;iilOD. -223 Platk XXXIY. Fig. 1. Calaniitina Goppi'rti Ett. sp. Woolley Colliery, Darton, near Barnsley. Mifldle Coal Measures [1199]. Natural size. Collefted by Mr. W. Heiniiigway. At a and b are seen two wliorls of bi-ancli scars. Fig. 2. Calamite from "Coal Ball," Hard Bed, Halifax. Lower Coal Measures. Specimen in the collection of tlie late Mr. Spencer, Halifax. a. Cast of pith cavity. h Ik Vascular axis with structure preserved. Natui-al size. Fig. 3. Equisefifes Ileminywayi Kidston. Monckton Main Colliery, near Barnsley. Middle Coal Measures [1678]. Natural size. Collected by Mr. W. Hemingway. Fig. 4. Cone of E(jiiisi'fum h>/f))uile, natural size, for comparison with Kl..„. /.,.,,,.^ Yorks. C-ol.ajnl l'„h/l,r. So,-., Vol. XIV., I'lol, XXXIV KIDSTON : TIIK KL()I{A OK THE CAHHONI I'KKOIS PKKIOD. Pl.ATK XXXV. FiiT. 1. /'nn(i(/((i-i(i cdlamiuiriii Artis. sp. Croplieatl Pit, iSauchie, near Alloa, Clackniannan.sliire. Lower Coal Measures [2815]. Thrcc-tifths natural size. Pi-obably the I'ootlets of a CdluDiite. Fis;. 2. Cdlamitnui approxinuUa Brongt. sp. Woodhill Quarry, Kilniaurs, Ayrshire. Lower Coal Measures [1551]. Three-fifths natural size. At a is seen the cast of the jiith cavity, and at h tlie impression of the vascular cylinder. Fig. '^. Cdlaniifes ^^acko/vii Brongt. Oaks Colliery, near Barnsley. Middle Coal Measures [2218]. Three-fifths natural size. Collected by Mr. W. Hemingway. Pith cast of stem n, giving (ttl' another stem /i, also only represented by the pith cast. At c are seen some rootlets. KIDSTdN : TIIK FLOKA OF TIIK CAliHONlFKliOUS PKKIOD. Platl: XXXVI. F'li^. 1. Tei'iiiiiiation of ;i Culamito heloiigini!; to the aoction Calannfi iiK, showint; the lonij; narrow foliage. Tliree- fifths natiii'al size. Dolly Lane, Leeds. Middle Coal Measures. Coll(>cted hy Mi'. J. W. Bond. Fig. 2. Diagiviiiunatic I'epi-esentation of Calanw^Uirliys, showing barren and fertile whorl. Fig. ."3. I )iagraiiiniatie representation of I'd/troxfnc/ii/d. showing barren and fertile whoi'l. Fig. 4. Caiamitina verticilhita L. k, H. sp. Oaks Colliery, near Barnsley, Yorkshire. Middle Coal Measures [2148]. Three-fifths natural size. Collected by Mr. W. Hemingway. At the upjoer end of the spcH'inien is seen a verticil of branch scars. The fossil shows a ribbed exterior which in this species appears to represent the outer surface of the plant. KIDSTON : TIIK KKOKA ol' TlIK CAUIJONI FEHOUS I'KKIOI). 'JlM) I'l.ATK XXXVII. Fiy. 1. All II iil((ri