IMAGE EVALUATION TEST TARGET (MT-3) // 1.0 I.I 1.25 ■^ Im 111112 2 llf 1^ = !!: 1^ 12.0 1.4 6" 1.6 V] <^ v: /^ Photographic Sciences Corporation 4- ,\ « :\ \ # "^.^ 6^ 23 WEST MAIN STREET WEBSTER, N.Y. 14580 (716) 872-4503 ) &r Jt? ^^ M /K^ ^ €/. ^ h CIHM/ICMH Microfiche Series. CmM/ICIVIH Collection de microfiches. Canadian Institute for Historical Microreproductions / Institut Canadian de microreproductions historiques Technical and Bibliographic Notes/Notes techniques et bibliographiques The Institute has attempted to obtain the best original copy available for filming. Features of this copy which may be biblioqraphically unique, w.'iich may alter any of the images in tha reproduction, or which may significantly change the usual rr.ethod of filming, are checked below. D D D D Coloured covers/ Couverture de couleur r~~l Covers damaged/ Couverture endommagie i Covers restored and/or laminated/ Couverture restaur^e at/ou pellicul^e r~~| Cover title missing/ Le titre ce couverture manque □ Coloured maps/ Cartes gAographiques an couleur □ Coloured ink (i.e. other than blue or black)/ Encre de couleur (i.e. autre que bleue ou noire) □ Coloured plates and/or illustrations/ Planches et/ou illustrations en couleur nBouni Relii Bound with other material/ avec d'autres documents Tight binding may cause shadows or distortion along interior margin/ Lareliure serree peut caujer de I'ombre ou de la distorsion le long de la marge intdrieure Blank leave:i added during restoration may appear within the text. Whenever possible, these have been omitted from filming/ II se peut que certaines pages blanches ajout6«s lors d'une restajration apparaissent dans le texta, mais. lorsque cela dtait possible, ces pages n'ont pas iti filmikes. (.'Institut a microfilma le meilleur exemplaire qu'il lui a ete possible de se procurer. Les details de cet exemplaire qui sont peut-^tre uniques du point de vue bibliographique, qui peuvent modifier une image reproduite, ou qui peuvent exiger une modification dans la m^thode normals de filmage sont indiquds ci-dessous. r~~] Coloured pages/ D Pages de couleur Pages damaged/ Pages endommag^es Pages restored and/oi Pages restaurees et/ou pelticuldes I I Pages damaged/ r~~] Pages restored and/or laminated/ Pages discoloured, stained or foxed/ Pages ddcolorees. tachetees ou piquees □Pages detached/ Pages d^tachees SShowthrough/ Transparence □ Quality of print varies/ Quality inigale de I'impression □ Includes supplementary material/ Comprend du materiel supplementaire < edition available/ le Edition disponible □ Only edition available/ Seu Th« to Th« PO! of filn On bet the sio otfi fira sio or Tht sha Tl^ wh Ma difl ent beg rigl' reqi mai Pages wholly or partially obscured by errata slips, tissues, etc.. have been refilmed to ensure the best possible image/ Les pages totalement ou partieilement obscurcies par un feuillet d'errata, une pelure, etc., cnt it^ film^eis i nouveau de facon a obtenir la meilleure image possible. Additional comments:/ Commentaires supplbmentaires: Pagination is as follows : [6261 - 641 p. This item is filmed at the reduction ratio checked below/ Ce document est film^ au taux de reduction indiqu^ ci-dessous. 10X 14X 18X 22X L 26.^ 30X 12X 16X J 20X 24X 28X 32X I tails t du odifier une mage Th« copy filmed here hee been reproduced thenks to the generosity of: Library, Geological Survey of Canada The images appearing here are the best quality possible considering the condition and legibility of the original copy and in keeping with the filming contract specifications. Origiinal copies in printed paper covers are filmed beginning with the front cover and ending on the last page with a printed or iiluatrated imprea- sion, or the back cover when appropriate. All other original copies are filmed beginning on the first page with a printed or iiluatrated impres- sion, and ending on the last pacja with a printed or illustrated impression. The iaat recorded frame on each microfiche shall contain the symbol ^c^ (meaning "CON- TINUED"), or the symbol V (meaning "END"), whichever applies. Maps, platea. charts, etc., may be filmed at different reduction ratioa. Those too large to be antireiy included in one exposure are filmed beginning in the upper left hand corner, left to right and top to bottom, as many frames aa required. The following diagrama illuatrata tho method: L'sxemplaire fiimi fut reproduit grdca i la ginitosit* do: Bibliothique, Commission Geologique du Canada Lee images suivantee ont M reproduites avec le plus grand aoin. cornpta tenu do la condition et de la nettot* de I'exemplaire film4, at an conformity avec lea conditiona du contrat da fiimage. Lea exomplairea originaux dont la couverture an papier eat imprimio sont filmte en commen^ant par le premier plat et «n terminant soit par la derniire page qui comporta une smpreinte d'Impression ou d'iliustration, soit par le second plat, salon le cas. Tous lea autrea axismplairaa originaux sont filmte an commenpant par la premiere page qui comporte une smpreinte d'impreaaion ou d'illustration et an terminant par la derniire page qui comporte une telle empreinta. Un doe symbolea suivants apparaitra sur la damiire image da cheque microfiche, selon le caa: le symbols — ^ signifie "A SOIVRE", le symboie V signifie "FIN". Las canea. planches, tableaux, etc. , peuvent itrm filmte i das taux de rMuction diffirents. Lorsque le document est trop grand pour dtre reproduit en un saul cliche, il est film* A partir do {'angle sup4rieur gauche, de gauche i droita, et de haut en baa. an prenant le nombre d'images n^cassaire. Las diagrammes suivants iilustrent la m^thode. rrata :o selure, 1 a 3 32X 1 2 3 1 2 3 4 5 6 V p^ "I '■' (,(. >■ OH ''i THE VEGETABLE STRUCTURES IN COAL. i» W' By J. W. DAWSON, LL.D., F.G.S. &c., parrciPAL op m'gill college, Montreal. [From the Quarterly Jottrnal of the Geological Society, ^ Feb. I860.] I ' I On the Vegetable Structures in Coal. Ey J. W. Dawson, LL.D., F.G.8., etc., Principal of M'Gill College, Montreal. (Plates XVII.— XX.) ' Mineral Contents Introrliiction. I. Struct iiros preserved in Cliarcoal." A. Tissues of Crypt ogamous Plants in the state of Mineral Char- coal. li. Tisssues of Gymnospcrmous Plants in the state of Mincnil Charcoal. r. Ti^^sues of Uncertain Origin. II. Structures preserved in the Layers of Compact Coal. III. General Conclusions. Introcjnctlon.—Xcce^tmg as established concisions the vegotdble origin of coal and the accumulation of its materials by growth in situ, rather than by driftage, there still remain some questions regarding Its production, to which as yet no verj- satisfactory answers have been given. _ One of these relates to the precise genera and species ot plants which have contributed the vegetable matter required ; an- other to the causes (whether differences in the plants themselves or m the manner ot their preservation) which have produced the different quahties of coaly matter observable in the different parts of the same bed, or in (hfferent beds in the same coal-field. In aid of the solu- tion of these questions, and incidentally in illustration of the struc- ture and affinities of the more important coal-plants, I desire to offer the results of a series of microscopical observations on +he bi^aminoua coals ot Nova Scotia. The observation of the beds associated with the coal, and of their contained fossils, has already furnished data which, inforentially at least, might dispose of these questions. A fundamental fact is the almost constant occurrence of Stirpnaria in the underclays. first as- certained by Sir W. E. Logan, especially when taken in connexion with the further observations of Mr. Binney and Mr. Brown*, that Stnjmana is the root of Siwberry and Lcsquereux in those of the United States,_and the exploration of the w^onderful coast-sections of the South Joggins and Sydney by Sir W m. Logan, Sir C. Lyell. Mr. Brown, and the author— have all contributed facts and conclusions tending inevitably to certain results * See also papers by the author. Quart. Journ. Geol. Soc. 1840 and 18.^3. f> I -^ DAWSON — VEGETABLE STBIICTUHES IN COAL. 627 j. { 4 respecting- the materials of coal, which, however, it appears to me, those geologists not immediately engaged in the study of the carbo- niferous sj'stem have hecn slow to perceive. The direct investigation of the tissues preserved in the coal itself has also been pursued to some extent by Witham, Hutton, (ioeppert, Brongniart, Bailey, Hooker, Quek(>tt, Harkness, and others. Two difficulties, however, luive impeded this investigation, and have in some degree prevented the attainment of reliable results. One of these is the intractable character of the material as a microscopic object, the other the want of sufficient information in regard to the structures of the plants known by impressions of their external forms in the beds of the coal-formation. Terplexed by the uncertain and contradictor^' statements arising from these difficiilties, and impressed with the conviction that the coal itself might be made more fully to reveal its o^vn origin, I have for some time been engaged in experi- ments and observations with this object, and believe that I can now offer definite and certain results in so far as relates to the particular coals examined, and, I have no doubt, with some slight modifications, to all the ordinary coals of the true coal-measures. In ordinary bituminous coal we recognize by the unassisted eye lamina) of a compact and more or less lustrous appearance, separated by uneven films and layers of fibrous anthracite or of mineral charcoal. As these two kinds of material differ to some extent in origin and state of preservation, and in the methods of study applicable to them, I may aiTange the subject under the following heads : — 1st. The structures preserved in the state of " mi" oral charcoal." 2nd. The structures preserved in the layers of compact coal. I. Structures itresefved as " Mineral Charcoal^ The substance known by the very appropriate name of "mineral charcoal" consists of fragments of prosenchymatous and vasiform tissues in a carbonized state, somewhat flattened by pressure, and more or less impregnated with bituminous and mineral matters de- rived from the surrounding mass. We cannot suppose that this substance has escaped complete bituminization on account of its original constitution ; for we have abundant evidence that this change has passed upon similar material in various geological periods. A substance so intimately intermLxed with the ordinary coal cannot be accounted for by the supposition of forest- conflagrations or the action of subterranean heat. The only satisfactory explanation of its occur- rence is that afforded by the chemical changes experienced by woody matter in decay in the presence of air, in the manner so well illustrated by Liebig. In such circumstances, wood parts with its hydrogen and oxygen and a portion of its carbon, in the forms of water and carbonic acid ; and, as the ultimate result, a skeleton of nearly pure charcoal, retaining the form and sti-uctm-e of the wood, remains. In the putrefaction of Avood imder water, or imbedded in aqueous deposits, a very different change occurs, in which the prin- cipal loss consists of carbon and oxygen ; and the resulting coaly 628 PHOCEEDINGS OP THE GEOLOGICAL SOCIETY. [March 9 profluot oontaina proportionally more hydrogen than the original wood. This IS the condition of the compact bituminous coal. This last may, by the action of heat, or by long exposure to air and water, lose its hydrogen in the form of hydro-carbons, and bo con- verted into imthracitc. In all the ordinarj' coals we have the pro- ducts, more or less, of all these processes. The mineral charcoal results trora subaerial decay, the compact coal from subaqueous pu- trefaction, more or less modified by heat and exposure to air. As Dr. Newberry has very well shown, in coals, like canncl-coal, which have been formed whoUy under subacjueous conditions, the mineral charcoal is deficient*. A consideration of the decay of vegetable matter in modem swamps and forests shows that all kinds of tissues are not under ordinary circumstances susceptible of the sort of carbonization which we hnd in the mmeral charcoal. Succulent and lax parenchymatous tissues decay too rapidly and completely. The bark of trees very long resists decay, and, where any deposition is proceeding, is likely to be imbedded unchanged. It is the woody structure, and espe- ciaUy the harder and more durable wood, that, becoming carbonized and splitting along the medullary rays and lines of growth, affords such fragments as those which we find scattered over the suifaces of the coal t. These facts would lead us to infer that the mineral char- coal represents the woody debris of trees subjected to subaerial decay, and that the bark of these trees should appear as compact coal along with such woody or herbaceous matters as might be im- bedded or submerged before decay had time to take place. We shall have an opportunity of testing the accuracy of these views when we consider the textures actually visible in the coal. In examining the mineral charcoal, I have, after many trials, adopted the following process of preparation. Specimens were selected containing the tissues of only a single plant. Fragments or portions of stems of this character can be obtained by careful mani- pulation from most coals. They were placed in marked test-tubes and treated with strong nitric acid, in which they were heated to the boiling-point, and kept in that condition so long as dense fumes of nitrous acid were disengaged, or until, on looking through the tube the matenal could be seen to have a brown colour and a certain de- gree of transparency. In many cases, boiling in this manner for a short time is sufficient to render the fibres flexible, and as transpa- rent as slices of recent wood when slightly charred. When ready lor examination, the charcoal was allowed to settle, and repeatedly washed ^vlth pure water before removing it from the tube. It was then examined m water, with powers of from 50 to 300 diameters drawings of the structures observed being made with a camera • and when It appeared desirable, specimens were j)ut up in balsam for liu-ther examination. Some refractory specimens were found to re- S;"GeAo?K.''"" '"''"' -'^^ Fossils fro. XovaScoL.- Q.t". t Sep paper of 184(1 prcriousiT cited 1&59. DAWSON — VEOETABLK SiRlCTUKES IN COAL. 629 (luire iiltornate wiushing and Ijoilinj? in hydrooliloric and nitric acids before their structures could be made out ; but in tlie jjreparation of more than two hundred specimens from various kinds of coal 1 have scarcely met Avith any that resisted all these processes*. I may obsciTC here that the object is not to decarbonize the coal and obtain what has been termed a siliceous skeleton. The change effected consists in the removal of bituminous matter, which is oxidized and dissolved by the acid, and of mineral matters, especially of the sulphuret of iron, which is one of the principal causes of the'^brittle- ncss and opacity of the crude mineral charcoal. The prepared material is nearly pure carbon, burning without flame and leaving scarcely any ashes. It represents the cell-wall and its ligneous Iming, or i)erhaps in some cJises only the latter, in a state of per- fect integrity, appearing under the highest powers (piite smooth and continuous, and with all its minute markings in excellent preserva- tion. The methods of incineration of the charcoal and of polishing its firmer portions I have found to be, in comparison with the nitric acid process, of little value. The first gives no adequate idea of the real character of the tissues. The second gives merely a rude outline of the more minute markings, and is chiefly valuable as af- fording cross-sections and a better view of the general arrangement of the tissues than can be obtained from the shreds of woody matter resulting from the i)rocess above described. It is further necessary to state that, to compare specimens of coal with the structures of mineralized plants from the accompanj-ing beds, it is not sufficient to have slices of the latter. It is necessary also to have specimens prepared by removing the mineral matter by an acid. Most of the coal-fossils showing stnicture are mineralized by the carbonates of lime and iron ; and on removing these, the cell- walls will be found intact and sometimes apparently not even car- bonized. Diluted hydrochloric acid suffices for this ; and structures by no means to be found in the comparatively rude slices prepared by the lapidary can be distinguished in these isolated cells. Pyritous fossils, so intractable as slices, can always be resolved by the treat- ment with nitric acid, though in some cases they require a prelimi- nary roasting, or, w4iat is better, exposm-e to the weather until the pyrites begins to crumble. A. Tissues of Cri/jifor/amoiis Plants in the state of Mineral Char- coal.— Vndev this head I notice, in the first place, coarse scalariform tissues apparently identical Avith those of Lepidodenclron and Uloden- dron, in both of which genera the vascular axis consisted of a cylin- der of scalaritorm vessels of great size, exceechng in this respect not only those of Sigilhma, but those of its kStigmaria-roots. Fig. 1 represents a portion of one of these vessels from the axis of a Ulo- dendron, i^roh'My C. minus, from the Pictou main-seam. Fig. 2 ia similar tissue from the rachis of Lepidostrolms, of the t3-pe of L. vari- ahilis, from the Joggins. For the sake of comparison', I have drawn these to the same scale (300 diameters) as that of the vessel of Sticj- * This nitrip acid proc'ef!» is T believe. uoHrk the same with that reeommejided Hr Goe)i))evt and Morris, 630 pnocEEDiNos OF TUT. OEOLooiCAL sociETT. [March 9, marla in Fig. 10, as well as of tho other .seulurifonn voshcLs repre- Bented in various figures. It will thus ho Hcen that they are of great comparative size ; and this, so far as my ohsorvation extends, is cha- racteristic of these giant Lycopothaceous phuits. Tissues of this kind are by no means rare in the mineral charcoal, hut are much less abundant than some other varieties. They are often difficult of re- cognition, owing to tho wide tubes being cnished and appearing as masses of fibrils, which represent the broken transverse bars. In this state thoy constituto mineral charcoal of that powdery and incoherent character which does not distinctly show the; fibrous structure. A second and very remarkable form of Crji)togiimous tissue appears in the minute " needles" that often lie in immense (luantities in the planes of lamination of the coal, appearing to the luiked eye like hairs or bristles. In the lower part of the Pictou main-seam, through several yards of thickness, nearly every plane of lamination presents these bodies more or less abundantly ; and they also occur in the Sydney coal. For some time they baffled my attempts to ascertain their structure ; but I have at length been able to resolve them into the two kinds of tissue represented in figs. 4 a to /. One of these is distinctly scalariform, though in some instances, owing to tho obliquity of the bars, presenting the appearance represented in fig. 3. The other is a finely punctui'ed woody tissue, sometimes with very fine spiral lines. Each needle-like filament is, in short, a bundle of scalariform vessels, enclosed in a sheath of woody fibres. This is precisely the stnictm-e of the vascular bundles of the petioles of ferns*, which may be obtained by maceration in water in the same condition with these fossils. That plants so abundant as the ferns should have left quantities of their debris in the coal is by no means surprising, though, with tho exception of a few cases of the impres- sions of leaves mentioned by Goeppert, I am not aware that the fact has previously been ascertained. It is possible, however, that these vascular bundles may not have belonged to ferns alone. The long narrow leaves of Sigillariaf were strengthened by nerves no doubt composed of scalariform and woody tissue, and they must have fallen in great quantities in the coal-swamps. I have no certain means of making the distinction; but tho forms of the scalariform vessels woiild induce me to suspect that those in figs. 4a, h, may have belonged to Ferns, and those in figs. 4 cl, e, to Slg'dlarice. Further, the quan- tities of these fibres lying together and nearly parallel indicate that they may consist in part of the vessels of root-stoclcs or of bundles of stipes in situ. B. Tissues of Gi/mnospcnnous Plants in the state of Mineral Charcoal. — Modern gymnospermous plants, whether Conifers or Cycads, differ both from Cryptogams and Angiosperms in the pos- session of discigerous wood-cells, — the discs occurring on the sides fronting the medullary rays, and consisting of depressed portions of * Brongniart, V^g^taux fossiles, fig. 30. f Cyperites of Lindley and Hutton. See also Sigillaria hpidodendrifolia, Brongn. ; and leaf of 8. scutellafa, in fig. 22. I 1859. J DAW80N — VEOETABLK HTRUCirRES IN COAL. G31 the cell-wall enclosing thin spots or pores*. In true Conifers these peculiar wood-cella are associated with spiral vessels in the medul- laiy sheath and leaves. In Cycads, so far as I have boon able to ascertain, the associated vessels are of scalariform t)i)e. While similar in general structure, the discs or bordered pores of Conifers and Cycads admit of several varieties of form, and in different plants and in different circumstuncos may present very diverse appearances; and, since in the mineral charcoal these minute structures are those on which wo must chiefly rely, it may save time in the first place to state the true nature of the gj-mnospermous bordered poro or disc, an example of which in all its parts, and as preserved in coal, is repre- sented in Hg. o. 1. The central part of the strxicturo consists of a thin portion of the cell-wall, usually callcil a " pore," which it actually is (owing to the disappearance of the original cell- wall) in much of the mineral charcoal. It may bo circular, or more or less in tho form of a slit ; and in tho latter case tho pores on two adjoining cell- walls are often placed obli(iucly to each other, so that the two pores belonging to tho opposite sides of the double cell-wall present a cruciform appearance. This is seen in some recent Conifers, as Sallsbaria, and in Cycads, as in Cifcas revohitaf, and is not infrequent in mineral charcoal, in fossil Conifers, and, as 1 believe, also in Siyillaria. 2. The pore is placed in tho centre of a depressed disc or areole in the cell-wall, which under the microscope appears as a circle, either, as in most recent Pines, much larger than the pore, or, as in some Cycads and some ancient Conifers, in ISh/ilktHce, and in Calamo- dendra, scarcely larger than the pore when the latter is circular. 3. \Mion the discs are crowded, the intervening portions of the cell-wall, especially if thickened by ligneous deposits, appear as a hexagonal network, or as transverse bars enclosing the discs. This occurs in Araucaria, in many fossil Conifers, and in Sig'dlaria and Ca la mode m Iro n . It is important to observe that, in tissues in which the whole of this complex structure exists, it often appears only in part. Speci- mens of fossil wood of the same species may in different states of preservation, or prepared in different ^^t'ays, exhibit to one observer hexagonal reticulation, to another circular discs or apparent pores, to another simple or cruciform sUts, to another all these structures united. This has been a fertUe cause of error in the microscopic examination of fossil plants. It is also worthy of notice, that when the pores are oval and transverse, and placed closely together, they form with their separating bars a kind of scalariform tissue not in- frequent in Cycads and in fossil plants allied to them, and perhaps characteristic of these plants. * The occurrence of such pores in WlnterccB and some other angiosperms is an exceptional fact ; and the identity of structure of these with the true gymnosper- mous disc is not certain. t Quekett, Lectures on Histology, vol. feren. Goeppert, Monographie der Coni- 632 PttOCKKDrNOa OF THE OEOI.OOICAI, NOCIETT. [Mttfch 9, I Hhnll now proceed to dcHcribo Hoverul forma of tissue that boar the above-mentioned ehaructcriHtics of gyniiionpermouH vegetation, and to consider the evidence that can be obtained as to tho plants from which they have been derived. 1. One of the most commf)n forms of coal-tissue, especially in the great Pictou main-seam, is that represented in fif,'s. H a to h and m to 0. It con.nists of slender wood-cells, with one or two rows of discs on the sides of t'ae cell-walls frontinj^ the medullarj' rays, which, when they can be diHtin{?uishcd, are sliort and (lonsi.st of one or two rows of cells (figs, 8 a, <>), These discs have either largo round perforations with scarcely any arcoles, or present transverse or decussating slits, or have the pores distinctly inscribed in a hex- agonal or rectangular network. Examples of all these forms will bo Been in the figures. Fig. 8 m shows the reticulations and jjores, as seen in the same specimen under different lights. On some cells the discs are crowded, but on others they are thinly scattered (fig. 21). It must be obseiTcd, however, that the cells of the mine- ral charcoal are Hattftned by ])ressuro ; -md cons( quently, when a specimen lies with the medullary rays parallel to the lamination of the coal, the pores are well seen ; but v^hen in the opposite direc- tion, it is difficult to perceive them, and the tissue might be regarded as aporous. Woody tissue of the kind above described is nearly always asso- ciated with the scalariform vessels seen in figs. 8 /, I; I, in such a manneil as to show Lhat both occurred in the same plant. This ia an important observation (so far as I am aware, not hitherto made), and leading to the more precise determination of the plants from which the charcoal has been derived. It will bo observed that these vessels are mu(;h more slender than those previously referred to, and that some of them present the appearance of elongated transverse pores with areoles. Mineral charcoal of this type is usually tough, fibrous, and silky. It generally appears in rectangidar pieces, but sometimes in sterna much flattened. One of the latter, from the Pictou main-seam, is two inches broad and one line thick, and consists extenially of mi- neral charcoal, with a plate of compact coal in the centre, and within this a thinner plate of mineral charcoal. It is, I have no doubt, a hollow cylinder of wood, much decayed both within and without, and flattened by pressure. Of eighty carefully selected specimens from the Pictou and Sydney coals, twenty-seven exhibited wood- cells of this type ; and in more than half of that number these were associated with scalariform vessels. The tissue just described has been observeu by various microsco- pists, and is indeed one of the first to attract attention in most samples of coal, though in the methods usually pursued it is very imperfectly seen*. * Witham, Fossil Vegetables. Reade, Philos. Journ. 1837. Hooker, Memoirs of G^eol. Survey. Quekett, on Torbanehill Mineral, Microsc. Journal. Goeppert, Essay on Coal. Harkness, Ed. New Phil. Journal, New Series vol. i. Dawson, Quart. Journ. Qeol. Sc- 1846. I? IHoJM l»\ws(iN -vi:(,i;r\iii,i: siimti iui;h i\ coai,. i\:\:\ 2. Anutlicr \»rv ubundiiiit loiin of tiMMU". iirt'sciiting to tho naked <'V<' tile siiiiic I; iriK fcrs witli tliut l.ist (Irs(iil)t'(l. coiiliiins throe, four, or five ntw.s of pores (ILsposcl i;lteriiiifely. Th" eells theinselveH in this viiriety arc hir^cr, and the pores ])roportlonally smaller. They nie UMiijtlly less distinetly areitlated, and . jietimeH l»y transverse union paxt into a kind of scaliirif(»rni Mr reticulated tissue. Tluy are associated, liki- the last, with scalaiifonii vessels as represented in fiKs. U (t to /'. In the narao eighty specimens refeired to uhovo, fifteen were of this variety ; and. while the first was more al)undant in the I'ictou coal, this was (jbtained chielly from tiiat of Sydney. •i. Other du(!ts or \vood-(!olls, of still lar};;cr size, have numerotis rows of pores spirally arranj^ed. Thev are comparativelv rare (see Hj,'. 11)*. 4, In a numher of specimonH, slender scalaritorm vessels like those above dcM ril)ed wri' found ulone, without the porous tissues. i"). The comparatively coarse scalaritorm vessels of Sfli/inarltt should !)(> included here as heloii},nn}; to Snjil lurid. They occur not infrefi'.icntly in the coal. (».;o of thorn is* represented in tiR. 10. to shoN* its rci dive size and i.j)pc!iriiiice. This kind of charcoal is moiv lax and course in its texture ihan those pr(>viouHly mentioned under this head. Tlie_ tissues ahove-dtscrihed constitute ()y far the lai'^er part of the mineral cimreoal, not only in the coals more particulirly refened to, but in all others belouf-iuf^' to the true coul-forn\ation that 1 have examined. We may now iiupiire to what families and <,fenera they may iiave belonged ; and this can be ascertained only by comparison with the structures known to occur in plants of the eoal-formation. Materials showinu,- at the same time une(piivoc,i;l external form and internal structure arc unfortuuati'ly rare, and sometimes liable to different botanical interpretations. 1 shall not enter into the con- troversies which have been raised on thes(; points, but shall state us shortly as possibh> such observations of my own as appear to confirm or modify the resiilts of previous incpiirors. Of the views which have been maintained in resi)ect to the mineral charcoal of n porous or diseigenms structure, three, f think, deserve liero especial mention. (loepi)ert apjiears to regard these structures as Arauearian, referring them mainly to the species AraucarUcn awhonaccusf. Hooker be- lieves them to have belonged to Ci/caifsi. Harkness is inclined to attribute them to Calaniifca^. If for '' Cycads" we read Si(/l/l(tri(c, Avhieh I believe were the (^cads of the Carboniferous period, the unioTi of all these views will represent the true state of the case. The gemis >Si(j>lJa.-ia, in the wide sense in which the term is usually employed, no doubt represents -a famibj of phiuts, tiiough it is at present impossible accurately to distinguish the generic or even the specific forms. The correctness of Brongniart's reference of the * Bailey has observed this structure in American anthracite, together with some of the others prc\ ioiish- uolircd. ^' Heiiiert and (jooppcrl. .Vl)liandlinig. etc. \ Memoirs of Geological Survey. § Edin. N. t'liil. Jonfiial. Xcw Sei ii's. vol. i. •2 V «;u I'KUCKKDlNliS OV TlIK CiKOI,OGICAL SOl'IKTY. [ Mtll'ch 9, fuiuily to (jlyniuospcnus allied to Cyciids I cannot for a moment doul)t ; but tlifv pvoscnted among themselves a wide range of organ- ization, on tlie one hand prosc'iiting apparent affinities to Conifers, and on the other to (,'ryi)togamons plants. According to Brongniart*, the essential structure of Sir/iUarid' includes a woody axis surround- ing the i)ith and consisting of two series of scalarii\)rm or porous t vessels, with medidlary rays and transverse hundlus of vessels pro- ceeding from the inner cylinder through the n> jduUary rays and the thick cellular bark to the leaves. I have two examples from the Joggins, illustrating at once this structure and the tissues found in the 'coal. The first is an erect stump, not presenting external mark- ings, but standing on a Stigmaria-underclay and having a central axis retaining structure. Fig. shows a longitudinal section of a part of the axis, of the natural size. The pith (a) exists in the form of transverse coaly plates, showing that it had a structure like that of the 8t€rnhmji(x, some of which at least have been ascertained to be the pith of ConifersJ. Immediately surrounding the pith is a narrow cyUnder of scalariform vessels, fig. G h and fig. 7 /. From this pass ijauus of similar tissue chagonally outward (fig. 6 c) through the outer cylindcr,which consists of the discigerous tissue, represented in difl'erent aspects in figs. 7 a, h, c. c. The minute markings of the same tissues as seen when prepai'cd by an acid are represented in fig. 7 '/. In this plant we have woody tissue which might readily be confounded with that of Conifers ; but it is associated, not with the sjjiral vessels of these plants, but with scflariform vessels, which, if alone, might be mistaken for those of Lycopodiacete or Cycadeae ; and the general arrangement of the tissues is that of the Sigillarioid plants. A glance is sufiicient to show the similarity of both the dis- cigerous and scalariform tissues to those of the mineral charcoal above described. Another example is furnished by the mineral char- coal filling the lower part of a large erect stump, ribbed in the man- ner of ^igni(()-io', — tht? same which aftbrded the bones of Deiuircrpe- ton Acadiamuii. These bones and certain foreign vegetable matters lay on a bed of mineral charcoal, all of one description and repre- senting the decayed wood which occupied the base of the stump. The structure of this wood, as it appears in slices, is represented in fig. 13 a. It is given in this form in my pa])er on the Joggins § ; and at that time I supposed it to bo a variety of interru])ted scalari- form tissue. AVhen subjected to the actum of uiti'ic acid, it displays the beautiful porous and areolatcd structures shown in figs. V.\ h, - further exani))les, these may suffice to show that the great I'ibbed Sig-illarioid trees of the erect forests of the Joggins are identical in the structure of their woody axes with much of the mineral charcoal of the coal. They also show that the wood of some of these Sigillarioid trees ajjproachcd very nearly to that of Conifers, from which, however, they differed in the possession of a cvlindcfr or bundles of scalariform tissue, and in the great thickness of their cel- lular bark as compared with that of their woody axes. Tlu" differ- ence in this last particular has, howevr, been much exaggerated. Such facts as I have been a ble to obtain give for the average relative diameters of the axis and of the whole stem the proportion of one to four or ii\c ; so that a Shjillaria of tlie very ordinary diameter of two feet would have an axis o to 6 inclies in diameter, and might atfoid much mineral charcoal*. 'J'he structures observed in the genus Calcnnlfea (Cotta) or Cala- morhndfO); (lirongn.), to wliich some at least of the Ca/aniites l)elong, are very similar to those above described, and include both discige- rous and scalarifoi'm tissues t. In tigs. H and \2 a to h I have re- presented the structures of a plant luiving the external markings of Cala mites, preserved as pyritous mineral chareoai in the Sydney main-seam. They are similar to those described by Cotta, Unger, Dawes, and (loepjiert. aTid cannot be distinguished from mu(;hof the discigerous mineral cliarcoal. They can scarcely be distinguished from those of SUjiUaria, to which, as Brongniart maintains, they were probably closely cdlied, differing principally in their verticillate leaves and thinner cellular bai'k J. It is unfortunate tliat we do not know how i-AY such structures prevailed among the Calamites ; but here, as in the case of /S^/f////r^)vV>', we probably have a gymnospermous family presenting several degrees of complexity of structure, but in all cases having Avoody structui'cs which have furnished important contribu- tions to the beds of coal. Tlie only doubt resting on this conclusion is the possibility that similarity of external form may cause us to as- sociate i)lants of cry])togamous aHinities with those which, on the evidence of structure, we have good ground to believe will, when their fructification is known, be found to occupy a higher place §. * Hooker, in Mem. Geol. Survey, states that the axis of a ^igiUaria^xe feet in diameter might be only two itielios in thickness. Aeoording to my observations, the pith-cylinder alone of sncli a trunk, independently of the woody matter sur- rounding it. sliould be of tliat thickness. i Dawes. Quart. Journ. Ged. Soc. \M\). Goeppert, ' Transition Flora '. Harkness, lac. rif. \ The vonf$ of Cdhrmifra were not unlike the rooflrfs of Sf/gmcr/n, and occur with tliem in the underclpys. isee paper on the Joggins seetion. Qiun-t. Journ GiH)l. Soc. ISivJ. § My observations merely show that plants having the surface-markings of CV//«/«/Vft'iha\('thestrueturedeseribefl. Goeppert also has obtaiiu'd similar results ('Transition Flora'). 1 do not by any means niainlain that all tlie plants in- cluded in the genera CaJnmodfndni,). ('(ildudfta. and CdJconiti^ sliow such struc- ture.s; on the contr.iry. J am I'ouviuecd that much conrusioii exists in tliesa (j:{G i'Ko(i:i:uiN-iis oi- Tin: okological socikty. [March 1». Tho tissiK^s of Coii'ifira nvo conipnnitivfly well known ; bnt they so dosfly roscniblo, when in tlie state of mineral eharooal. those of SiijUhtrUt and Cahiumlemlron that 1 cannot with certainty affirm their presence. Fig. U represents a discigerous structure whuli 1 have not found associated with scalariform vessels ; iuid this kind of negative evidence affords at least a possibility that some of the fra<^raents of mineral charcoal showing only discigerous cells are Conifi'rous. Several vears ago I stated my belief that this is the character of much of the mineral charcoal* of the Tictou coal ; but I ^\-as not at that time aware of the existence in the coal ot other discigerous tissues so like th.ise of Conifers ; and my specimens we're slices poHshed in the usual Avav, and not showing in perfection the minute markings, i'igs. 20 a and h represent two of tho best ot these si)ccimens. which no doubt may be Coniferousf. ^\x examinations of the mineral charcoal open up the furtlu'r question, whether we really know the precise line of separation be- tween Conifera; and ShjiUana'. 1 am even inclined to suspect that some of the described' species of Conifers of the coal may be tlu^ woodv axes of large Si^f'/o.r///o*t with large Sternbergia-piths,— though Prof. Williamson appears to have ascertained that in one of these last the medullary sheath is spiral, not scalariform, and though those which I have examined appear destitute of scalariform tissue and diverging vascular bands. In further illustration of these singular Conifers as connected with the coal-structures, I give in figs. 1 o and 1 the struc- tures of the wood of two Stenihenjia', which I have been able to resolve by the nitric acid process since the public^ation of my P»P«-''' ^^ these fossils. These plants had no trace of scalariform vessels ; and it is not uninstructive, with i-eference to the possible affinities of Slj/llla- r!7/-(.s(-r('t(/ in the Lat/ers of Conij)act Coal. The compact coal, constituting a far larger proportion of the mass than the "mineral charcoal" does, consists either of lustrous con- choidal Chernj- or I'itclt-coal, — of less lusti'ous Slate-coal, with tiat fracture, — or of coarse coal containing much earthy matter. All of these are arranged in thin, interrupted lamina) f. They consist of vegetable matter Avhich has not been altered by subaerial decay, but which has undergone the bitinninous putrefaction, and luis theieby been resolved into a nearly homogeneous mass, which still, however, retains traces of structure and of the forms of the individual flattened plants composing it. As these last are sonu'timc's mure distinct than the minute structures, and are nccessaiy for their comprehension, 1 sliall, under the folloM-ing heads, notice both as I have observed them in the conls in (piestion. 1. The lamina' of jjitch- or cherry-coal, when carefully traced over tlic surfaces of accumulation, are found to present the (mtline of flattened trunks. This is also true, to a certain extent, of the finer varieties of slate-coal ; but the coaise coal appeiirs to consist of ex- tensive lamina'" of disintegrated vegetable matter mixed with nuid. 2. When the coal (especially the more shaly varieties) is held obliquely imder a strong light, in the manner recommended by (jioeppert, the surfaces of the lamina^ present the forms of many well- * I have little confidence hi the establishment of genem or species on the mi- nute struetiu'es of fragments of wood ; but I believe that this evidenee is suflieient to distinguish from each otlier the vascidar ervptoganis and gymnosperms of the eoal-period, and tiiat in tlio hitter, if careful to avoid being misled by ditferent states of preservation of the same tissue, we may distinguisii in tliis way Sii/illurice, Cdhimifice, and Con/fen^. Having gained this ))oint. wliieh T liave tried to reach by the observations recorded in tiie text, we may safely go on lo subordinate di- stinctions. This T have not attempted, but it is a promising field of investiga- tion. J. \V. 1).. May iS.V.t. t Tlie ipuuitity of vohitilc matter is not material to our present inquiry, since the diH'erences of structure depending on the vegetable matter remain even in anthracite. t.i;w i'Uu(.i;i:i)iMis di' riii: iii;()i,n(;icAL socir.TV. March U, known coul-pliUitH, as SiiiiUar'td, Ht'KiinKria, I\uicih'>i ov Smfjemthia, Ltpido<(inih'()ti, L'lodi'Htlron, and rougli bark. ])erhai)s of Conifci'S. :i. When the coal is traced npward into tlie roof-shales, we often find the lamina^ of compact coal i'ei)resented by fiatteued coaly trunks and leaves, now rendered distinct by being separated by clay. 4. In these flattened trunks it is the outer cortical layer that alone constitutes the coal. This is veiy manifest when the upper and under bark arc se]nirated by a film of clay or of mineral char- coal, occupying the ])lace of the wood. In this condition the bark of a large ^S'njtUiina gives only one or two lines in thickness of coal ; St'Kjinnria, Li'jiidodctK/roii, and U lode ml ran give still less. In the shiiU's these flattened trunks are often so crushed togcthei- that it is diflicult to se])ai'ate them. In the coal they are, so to speak, fused into a liomogeneous mass. '). The ])henomena of erect forests explain, to some extent, the manner in which layers of compact coal and mineral charcoal may result from the accumulati(m of trnnks of trees in sitn. In the sections at the South .Toggins, the usual state of preservation of erect Sir/iUariiH is that of casts in sandstone, enclosed by a thin layer of bark converted into compact, caking, bituminous coal, whih; the re- mains of the woody matter nuiy be found in the bottom of the cast in the .state of mineral charcoal. In other cases the bark has fallen in, and all that remiii^is to indicate the pLice of a tree is a little pile of minci-al charcoal, with strips of bark citnverted into compact coal. Lasth-, a series of siich remains of stnmjjs, with flattened bark of prostrate trunks, may constitute a rudinu'utary bed of coal, many of which exist in the Toggius section. In short, a single trunk of Sii/iliaria in an erect ibrest presents an epitome of a coal-seam. Its roots represent the Stigmaria-underclay ; its bark the compact coal : its woody axis the mineral charcoal ; its fallen leaves, with renuiins of herbaceous ])lants growing in its sluide, mixed with a little eartb\ matter, the layers of coarse coal. The coiulition of the dnral)le onivv bark of erect trees concurs with the chemical theory of coal, in showing the es]iecial suitableness of this kind of tissue for the production of tlu^ purer compact coals. It is also probal)le that the comparative impermeability of bark to mineral infiltration is of impoi'tance in this res])ect. enabling this n.aterial to remain un- atfected by causes which have tilled those layers consisting of herbaceous materials and decayed wood, with earthy matter, pyrites, etc. (). The microsco])ic structure of the purer varieties of compact coal accords with that of the bark of SujilJuria. The compait coals are capable of affoiding very little true structure. Their cell-walls have l)een press<>d close t<)gether ; and ])seudo-cellu]ar structures have arisen from molecular action and the segi'cgation of bituminous matter*. Most of the striu'tures which have been figured by mi( ro- * Sfiinc lignites and fossil woods IVoiii ihc .Saskatchewan, recently placer! in my hands h\ Sir Win. l^ofjan. \cy\ well illustrate the mri-hdiiind ehaii^c eN|)i>ri- cmid holh In wood and (.cirlical ceils in tlie i)i'nci>s ot' ciMnersion inlo louipaet \sr)U. ItA^^Ml^ -vr,(.i;iAii[,i; MurcrruKs i.\ (ovr,. 639 Kcopists niv of lliis lust i-luimctor, or at the utnui.st urc cell-struc- tures masked by concretionary action. i)ressurc, and decay. Hutton, however, appears to have ascertained a truly cellular tissue in this kind of coal. Cioei)])ert also has figured parenchymatous and perhaps bast-tissiies obtained from its incineration. ]}y acting on it with nitric acid, 1 have found that the structures remaining both in the lustrous conii)act coals and in the bark of ShjiUaricp are ])arenchymatous cells and fibrous cells, probably bast-fibres. These tissues are usually badly preserved ; but in some cases 1 have found them as perfect as in fig. IS, which represents cellular tissue from the coal of Lingau, Cape lU'eton. 7. 1 by no means desire to maintain that all portions of the coal- seams not in the state of mineral charcoal ccmsist of cortical tissues. Quantities ttl' herbaceous plants, leaves. «tc. are also ])resent, espe- cially in the coarser coals ; and some small seams a^tpear to consist entirely of such nmterial. — for instance, of the leaves of Ncegf/erathia or Foacites. These materials, however, constitute relatively but a small part of the larger and more valuable seams. I have repr(>- sented in fig. 10 fragments of the only tissue preserved in the Poacites-coal, which appears to be the epidei-mis of those leaves, showing traces of u fine cellular tissue and rinuid perforations, perhaps remains of stomata. I would also observe that, though in the roof-shales and other associated beds it is usuaE" only the cor- tical layer of trees that appetirs as compact bituminous coal, yet F have found specimens which show that in the coal-seams themselves true woody tissues have sometimes been imbedded unchanged, and converted into structureless coal, forming, like the Coniferous trees converted into jet in more modern formations, thin bands of very pure bituminous material. The proportion of woody matter in this state differs in different coals, and is probably greatest in those which show the least mineral charcoal : but the alteration which it has undei'gone I'enders it almost imi)ossible to distinguish it from the flattened bark, which in all ordinary cases is much more abundant. 111. (femral Conclusions. 1 . With rcsj)ect to the plants which have contributed the vege- table nuitter of the coal, these are principally the Sl(jill(()'i(c and C(il(i'initea\ but especially the former. With these, hoAvever, are intermixed remains of most of the other plants of the period, con- tributing, though in an inferior degree, to the accimiulation of the mass. This conclusion is confirmed by facts derived i'voiw the asso- ciated beds, — as, for instance, the prevalence of Stit/inaria in the coal. The lignite lias tlu- % .'. are of jet, but retains sulUcient traces of structure to sliow that it consists oi vonifcrous wood ; and in specimens in wliich the presence of nuneral matter allows tliis strnelnre to be belter seen, the density of tlie material is seen to he occasioned, not h_v the dissolution of the textiUT hito a paste, but bv the conii)lele II Utcning of the cells. Thi» condition occurs also in the ancient bitinninous coal, and accounts for all those appearances really struc- tvu-al >ccu in slices of il. and for the diUcirnce? in these observable in horizontal and vertical slircs. G4(J I'UOCKKUIXOS Ol' Tin; (iK(»I.()(iKAL SOCli: IV. Marcli U, 'I imdcTcliiy.s, and of Slijillaria; and t'alamiteK in the roof-shales and erect forests. 2. The woody matter of the axes of >Sif/i/hiri(f and CaluinUece and of Coniferous trunks, as well as the sealariform tissues of the axes of the Lfpii/(>(/n>((ir(t' and IJ/di/cuf/rci', and the woody and vascular bundles of Ferns, appear principally in the state of mineral char- coal. The outer cortical envelope of these plants, together with such jjortions 0+' their wood ami of herbaceous plants and foliage as were submerged without subaerial decay, occur as comi)act coal of various degrees of purity, the cortical matter, owing to its greater resistance to a(|ueous infiltration, affording the piu'est coal. The relative amounts of all these substances found in the states of mineral charcoal and compact coal depend principally upon the greater or less prevalence of suluerial decay, occasioned by greater or less dryness of the swampy flats on which the coal accumulated. 3. The structure of the coal accords Avith the xivw that its materials were accumulated l)y growth, without any driftage of materials. The Si(jllliiri' and Calawites were not, as often supposed, composed wholly, or even principally, of lax and soft tissues, or necessarily short-lived. The former had, it is true, a very thick cellular inner bark ; but their dense woody axes, their thick and nearly imperish- able outer bark, and their scanty and rigid foliage would indicate no very rapid growth or decay. In the case of Si(/il/<(rlA\- woody Avedges in the axis, the transverse marks left by the successivi; stages of upward groAvth— all indicate that at least several years must ha\'e been required for the groAvth of stems of moderate^sizc. As the best means of illustrating these features of the gi'owth of >Si(/in((ria, I have given, in fig. 22, a restoration of a ])lant of this genus, with figures illustrative of its mode of growth, from si)ecimens in my OAvn possessiont. The enormous roots of these trees, and tlu^ conditions of the coal-swamps, must have exempted them from the danger of being overthrown by violence. They probably fell, in successive generations, from natural decay ; and. nuddiig every * Pi\\)vv on Foss^ils IVoiM A'uva Scotia. Quart. Jourii. Gcol, Hul: IS17. (• Siv al>o van'ous firruirs illustrating these point!?, in BronirniarlV -Vc'^iMaux Fossilc's." I I.S.V.). DAWsox — vi:(ii;r\ 111,1; sTi!i(.Tiiii:s in ccai. ♦ '.41 V* nllowanco for other matoriuls, wv may sal'cly assert that every foot of thiekuess of pure bituiuinous coal implies the (jiiiet growth and fall of at least fifty generations of Sir/lllaritv, and therefore an undisturl)ed condition of forest-growth enduring through many centuries. Further, there is evidence that an inimcinse amount of loose parenchymatous tissue, and even of wood, perished l)y decay ; and we do not know to what extent even the most durable tissues may have disappeared in this way ; so that in many coal-seams wo may have only a very small part of the vegetable matter produced. 5. Lastly, the residts stated in this paper refer to coal-beds of the middle coal-measures. A few facts which I have observed lead me to believe that, in tli!> thin seams of the lower coal-measures, re- mains of Niiyi/erathla and Lc^uihilcndron are more abundant than in those of the middle coal-measures*. In the upper coal-measures similar modifir;itions may be expected. These difterences have been to a cm'tain extent ascertained by (ioeppert for some of the eoal- I)eds of Silesia, and by Les(]uereux for those of Ohio f ; but the sub- ject is deserving of further investigation, more especially by the means proposed in this paper, and which I hope, shouhl time and oi)portuuity permit, to apply to the seventy-six successive coal-beds of the South Joggins. EXPLANATION OF PLATES XVII., XVIII., XIX., XX. Fig. 1. Scalariibnu vessel of f7w/'';«//v/;/ ; niiiipral chaivoul : ;i()(l (liiiindcrs. -*. ., ,, Lcpnl. X\\n rf S/(///f((r/t/ ; longitudinal section ; natural size. 7. Str-K'tures in the same, 300 diam. a, k c, d, f, longitudinal sections; c, transverse section. 8 & 0. Discigerous and scalariform gymnospernious tissues ; mineral charcoal ; .'!((() diam. 10. Scalariform vessel of Sfi<)ii!(iri(i \ 300 diam. 11 & \'l. Structures of Calaiaodindvon (? Cdldniifcs); mineral charcoal; ;j(M.)diam. i;{. Structures of axis of erect S/(/t7/(iri(i ; mineral charcoal ; .'300 diam. 14. Discigerous cells ; mineral charcoal ; ."lOO diam. !;")& If). Coniferous tissue enclosing »S7f remains of Lycopodiaceous pl'uits of tlie genus V.-^ilojdit/fdiK (Quart, tlourn. Gcol. Soc. No. tiO. p. 477-) )' Report of Survey of Ohio, iS-.8. pixvn. ei a I A w d. a* M .UniJ ■o '^ .o ■ UY:: Ae'i C We^Tt iiib WWest imp VECETAPLS TI;1SJES IN COAL f« rt. J W r de; & West '^_ i:i WWe?t mn: :i(-ETABLE TiSbUHS IN ZQM. ► fj f\ <:>! n iu IS 16 ?(' a IS ! \ ■il iCh .'WD AtL CrWait :'.:.- WTffst jta Vrrr-ix;- T- T ^- -^V. u-oart, Joi-a-n Ced ooc Vol W PI X},. i I :WDM aW9st!i4i WVVe;t ;t :mp SIGITJ.AFIA