NRLF B M 175 73fl m- a 8S8KSKW ll f> I \\ %i 1 v ' l l . V O •.>-!- v * • v » \ » • ^ ' \ ' '• ' 1 ' ' ' ' ' ' ' ' §®S*&^^^ ^l»^>»;Ss marks the felspar represented in fig. 4, PL I. For want of a better name, we stated that the laminae are fibrous or striated; but we were quite aware at the time that the intersecting striae or fibres are often separated, and seemingly tubif orm : thus, where a transverse section of such fibres is exposed they appear like circular dots — even more decidedly than as repre- sented in our figures. We draw attention to this peculiarity, as probably offering an explanation of the " empty tubuli." What we strongly suspect will prove to be the same are the cuts or lines forming the "first" or incipient stage in the development of chrysotile (pp. 9, 14, &c.) ; especially as we are unable to detect any difference between them and the striae, that are thread-like, of feldspar. Seemingly confirming our suspicion is the fact that the " striation "• of a feldspar in graphic granite has actually been taken for " vertical tubular structure " by Dr. Carpenter, as will hereafter be seen. Eozoon Canadense. Dr. J. W. Dawson. A letter in ' Nature/ 1874. June 11, vol. x. p. 103. " Eozoon " examined chiefly from a Foraminiferal Standpoint. 1874- Drs. W. King and T. H. Rowney. Ann. & Mag. Nat. Hist. ser. 4, vol. xiv. pp. 274-289. Final Note on Eozoon Canadense. Dr. W. B. Carpenter. 1874. Ann. & Mag. Nat. Hist. ser. 4, vol. xiv. pp. 371, 372, Dr. Carpenter, notwithstanding the many errors he has c Plate IX. : they are considered to represent another important " eozoonal" structure — the " canal system/' Similar configurations, consisting of saponite, and imbedded in calcitic interpolations, have occurred to us in Cornish ser- pentine J. Metaxite, another closely related mineral, if not an allomorph, similarly presents itself — that is, as multifoliaceous expansions. And we have made known that aggregations of crystalloids of malacolite (a white variety of augite), assuming the most beautiful arborescent or dendritic shapes, are not uncom- mon in the hemithrenes of New Jersey, Aker, and Ceylon ; while a similar rock in the Vosges has lately yielded them to our investigations. The presence of spheroids, lenticuloids, and plates of serpen- tine imbedded in calcite brings us into view of the frequent iuterlamellation, in ophites, of these two widely different minerals, especially in those characteristic of the Archaean group of rocks in Canada. The phenomenon is well displayed in what is called the " laminated variety of Eozoon"§ — the layers of calcite being regarded as the " intermediate" or "supplemental skeleton" || of this presumed fossil, and the layers of serpentine as casts of its * Intellectual Observer (Dr. Carpenter on "Eozoon"}, vol. vii. p. 294. f Quarterly Journ. Geol. Soc. xxii. p. 193. I Phil. Mag. ser. 5, vol. i. pi. 2. fig. 13. § See Plate I. fig. 1. || It is presumed that a layer of ca'lcite ("intermediate skeleton") was bounded on each surface by a "tubulated" lamina (" nummuline wall"), and that the latter enclosed the " chambers " occupied by the sarcodic divi- sions of ft Eozoon," STRUCTURES OF OPHITES AND RELATED ROCKS. 11 " chambers." Such layers of serpentine are admitted to graduate into the plates, lenticuloids, and spheroids above mentioned, and thus to constitute the " acervuline variety of Eozoon." Interlamellations of the kind are not confined to the Archaeans of Canada, as they, with other eozoonal features, have been made known by us to be present in the post-Archcean green marbles of Connemara and the Jurassic ophite of the Isle of Skye*. They have been observed by one of us in a hydrous dolomite in contact with an igneous rock near Predazzo ; and he was shown by Prof. Miiller a specimen in the museum at Basle, from Kalk- gebirg, Todte Alps, Davos, in the Engadine. A similar rock, we suspect, occurs somewhere in Southern Italy; for the fragments of coarse verd antique lying amongst the ruins of Pompeii and Herculaneum consist of rude alternations of calcite and serpen- tine ; and the " marble " columns, perforated by pholases, still standing erect in the ruined temple of Serapis, at Puzzuoli, have been formed out of a rock apparently of the same kind. The serpentinyte of the Lizard, Cornwall, has afforded us a great variety of bodies, consisting of serpentine, saponite, and other mineral silicates, and resembling corals, foraminifers, worm-tubes, as will be seen in our paper published in the ' Philosophical Magazine/ ser. 5, vol. i. pi. 2. Other simulations of organisms have lately occurred to us in the same rock : one, a plate (apparently of feldspar undergoing a change), is riddled in the most regular manner, so as to have the closest resem- blance to a leaflet of Retepora cellulosa. A number of silacid ophites (ophi-euphotides &c.) have quite a porphyritic structure — a common occurrence in those of Central Italy. We have, in the paper just referred to, shown that the serpentinyte of the Lizard is porphyritic, containing crushed crystals, originally of augite, but now possessing the cleavage and lustre of chlorite f. * Proc. Roy. Irish Acad. new ser. vol. i. (January, 1871). t Since this paper was published, the Rev. Prof. Bonney and Mr. W. Hudlestone have noticed these crystals, in the l Quarterly Journal of the Geological Society/ vol. xxxiii., the former stating that they are altered enstatite (pp. 921, 922), and the latter that " their chemical composition is similar to that of bastite j and they are probably the result of the hydration of a variety of enstatite " (p. 926). The latter statement may be taken as confirmatory of our earlier idea that the crystals are pseudomorphs of chlorite after augite. 12 KOCK-METAMORPHISM. CHAPTEE IV, ORIGIN OF CERTAIN MINERAL, STRUCTURAL, AND CHEMICAL CHARACTERS OF OPHITES AND RELATED ROCKS. THIN sections of peridote from Elfdalen, in Dalecarlia, exhibit this mineral in an amorphous or finely granular condition, and not unfrequently containing thin fibrous or striated undulating laminae, more or less separated and subparallel : the striae or fibres are at right angles to the surfaces of the laminae (Plate I. fig. 3). A similar structural character, seen on one of the two sets of eminent cleavage-planes, marks the feldspar exhibited in fig. 4, Plate I. From Harris (Hebrides) we have a remarkable quartzose feldspathic rock, called " graphic granite" (see Plate I. fig. 5): its feldspar, of a white colour, is in layers (a) transversely inter- sected by striated laminae (a), which alternate with others (b) devoid of structure, except occasionally a platy kind. The strise, which lie at right angles to the boundary-planes of the laminae, are of the kind called ' ' striping," characteristic of triclinic feld- spars'*. This graphic granite also contains layers of quartz (b) alter- nating with those of the feldspar, the latter being the thickest. The striated laminae contained in the feldspar layers meet those composed of quartz at a few degrees from a right angle : in the example figured one of the striated laminae abruptly terminates against a quartz layer. These three cases, with others that could be adduced, offer a striking similarity to the amorphous and fibrous interlamina- tions of the Colafirth ophite. The fibrosity, or call it striation, * We have studied in this connexion the striping of. feldspars, but without arriving at any satisfactory conclusion as to its nature, save that it may be related to striation and fibrosity. Looking at the fibrous cleavage charac- teristic of selenite, possibly we have in these different structures modifications of typical mineral cleavage : the stylolitic structure of rocks seems to be another form, ORIGIN OF CHARACTERS OF OPHITES AND RELATED ROCKS. 13 of peridote and feldspar may be safely assumed to be genetically identical with that of chrysotile, taking the incipient variety of the latter into consideration ; moreover, we hold it to be in all three a superinduced divisional structure. In short, the fibrous laminae of peridote and feldspar have as much right to bear a distinctive name as the fibrous allomorph of serpentine. Reverting to the alternation of two distinct minerals, viz. feldspar and quartz, in the graphic granite of Harris, the case affords a parallel to the interlamellation of the serpentine and calcite in the Archaean ophites, as already noticed. The same phenomenon is also instructively displayed in ared quartzose feldspathic rock from Siberia, of which a portion is represented under fig. 6, Plate I. The layers of quartz and feldspar exhibit such a definite alternation that in this respect the calcareo-ser- pentine interlamellation of the "eozoonal" ophi-calcite of Canada does not surpass it. The remarkable character last referred to, which first gave rise to the idea that it is due to organic forma- tion, does not, according to Dr. Dawson, occur in any rock of purely mineral origin ; the instances already adduced will show that this is an entirely erroneous statement*. Besides, we are now in a position to demonstrate more clearly our point ; Professor Heddle having kindly placed in our hands a remarkably beau- tiful specimen, from another locality, Tarbert, in Harris, which consists of white quartz and a silvery feldspar (? moonstone), alternating so definitely as to be identical in this respect with the interlamellation of serpentine and calcite in type specimens of "Eozoon Canadense " f . No wonder that this lamellar graphic granite has been pronounced to be of organic origin by believers in the eozoonal doctrine % \ But it may be urged that in these examples the alternation consists of two siliceous minerals. Granted. Still we are able to dispose of this objection. There is a large specimen from near Vigan, Department du Gard, under a glass shade in the centre of the Geological Museum of the Jardine des Plantes, and another one in a wall-case from the Pic d'Eridlitz, Pyrenees, both consisting of feldspar and calcite definitely interlamellated §. * See Introduction, A.D. 1879, t See Plate IX. fig. 3. t See Introduction, A.D. 1876. § M. Edouard Jannettaz very kindly showed me the specimens. The Vigan specimen, according to its label, Was procured by M. Cordier. — W. K. 14 ROCK-METAMORPHISM. And rarely is a good collection of Norway minerals without specimens exhibiting layers of crystalloidal hornblende alter- nating with others composed of calcite. Specimens of hemi- threne, collected by one of us near Dunglow, in Donegal, have the calcite sharply alternating with layers of small crystalloids of idocrase. Interlamellations of calcite and malacolite (ser- pentine is occasionally present), from Connemara, constitute a variety of what we have called malacolophyte. It will thus be seen that a lamellar alternation such as characterizes the " laminated variety of Eozoon " is not an uncommon rock-phenomenon; also that the interlamellation of a mineral carbonate and a mineral silicate, respectively answering to the " skeleton " and ' ' chamber-casts " of this " fossil," is to be observed in rocks every structure of which is entirely of mineral origin. Reverting to chrysotile, briefly noticed in the last Chapter, we may be permitted to give a more detailed description of its four stages of modification; for, in doing so, we shall be able to show them in their course of development. In the first stage, chrysotile consists of a layer of serpentine penetrated at right angles to its two surfaces by parallel filiform darkish-coloured lines or cuts ; and, though for the most part of uniform length, they are often individually broken into two or more short lengths (Plate IX. fig. 1 a). The cuts are more or less separated, with the intervening space consisting of ser- pentine in its ordinary condition — amorphous, subvitreous, and of a green colour. Occasionally this variety assumes a rude, irregular columnar structure, due, we suspect, to the cuts falling into contact, and ranging themselves into divisional planes, which meet one another at no definite angle : the planes are separated by widish structureless interspaces, which constitute the body of the columns (Plate IX. fig. 2 ax) . Strictly speaking, it cannot be said that chrysotile in either of these states is fibrous — rather that we have examples of chrysotile in its incipient stage of development. In the second, the cuts are indefinitely increased, so that they cause a layer to appear as if it consisted of a dense mass of fibres (resembling those of asbestos — an allomorph of horn- blende). The fibres vary in colour from silver-white to a golden hue, and in lustre from silky to metallic (Plate IX. ORIGIN OF CHARACTERS OF OPHITES AND RELATED ROCKS. 15 fig. 1 b) : it is necessary to mention that they have no definite form as in the case of crystals. In the third) the fibres are comparatively stout, appearing like cylindrical aciculse ; they closely adhere to one another, and are usually of a glistening white colour (Plate IX. fig. 1 c). In the fourth, the aciculse, still retaining their colour, are more or less separated, and the spaces between them are occupied with calcite. Immersed for a short time in a weak solution of hydrochloric acid, a layer of chrysotile in this stage has its calcite dissolved out, leaving the aciculae separated — thus be- coming pectinated (Plate IX. fig. 1 d). It is now necessary to mention that the aciculse under the two last forms have been taken for organic structures — casts of fine tubuli penetrating the calcite in immediate contact with and covering the spheroids and layers of serpentine. Hence they have been identified with the fine tubulation characteristic of the " nummuline wall " of the Nummuline group of Foraminifera. Specimens are common showing all the varieties of chry- sotile graduating into one another — satisfactorily demonstrating that they are simply modifications of one and the same divi- sional structure. But it will be necessary to show how the difference between the second and third stages in the modifications of chrysotile has been produced, also the greater difference which distin- guishes the third from the fourth. In respect to the first difference, the explanation is afforded by the fact, made known by Delesse, that in chrysotile occurring in the Vosges the fibres are swollen and changed to a white colour when exposed to the atmosphere* (thus made acicular) — also by what has already been stated (and to be more fully noticed presently) of the fibres in a specimen of chrysotile from Reich enstein, which are occasionally fused together into solid pillars : in certain cases the original fibrous structure has disappeared. As regards the second, we hold that it, too, is demonstratively explained by the specimen just referred to. Fig. 1, Plate II., represents a layer of chrysotile between two of serpentine. The specimen, for the purpose under consideration, was placed in a weak solution of hydrochloric acid : a represents chrysotile in its typical state ; b, the same passing into the state of closely- * Ann. des Mines, ser. 4, t. xviii. p. 328. 16 ROCK-METAMOEPHISM. adhering aciculae ; cs the latter separated by vacant interspaces, the separations having contained calcite before the specimen was decalcified * : underneath, the chrysotile has been entirely replaced by calcite, except in one place, as where the fibres re- main attached to the subjacent layer of serpentine. In another part of the same specimen the chrysotile retains its characteristic features and colour ; but some of the layers are brown and porcellaneous, and the fibrous structure is either faint or oblite- rated. The layers, besides having the fibres (b) changed into aciculse in close contact, have them forming solid pillars ; the fibres are also transformed into aciculse, individually separated by calcareous interpolations, and passing completely across a layer or only reaching about halfway — appearing, where the interpolated lime is dissolved out, like a fringe hanging from the adjacent surface, and in the same way as they are (c) in the instance just noticed. In all the above cases we have clear demonstrations that the typical ' ' nummuline wall " of ' ' Eozoon Canadense " is a pec- tinated form of chrysotile, and consequently a product of purely inorganic agencies. To continue, under polarized light serpentine differs impor- tantly from chrysotile. The former in its ordinary state shows no bright colours — merely pale yellow, passing, on rotating the analyzer , through light grey, dark or bluish grey, and returning to pale yellow; whereas the latter (beginning with parallel prisms) shows pale green, bright green, purple and blue, dull green, purple and blue (with crossed prisms), crimson, yellow, and green, returning to pale green, each change being variously tinted— the whole reminding one of the still more beautiful colouring got by polarizing peridote. Calcite, which is often associated with chrysotile, shows nothing but different shades of grey j and when intermixed with chrysotile its presence vitiates the bright colours of the latter mineral, turning them into pale yellow and dark grey. Fig. 2, Plate II., represents a portion of one of three slides received from Canada, specially selected for us, we suspect, by one eminently distinguished for his advocacy of "Eozoon Cana- dense" as proving the organic origin of the structures diagnosed * A similar case has been noticed by Prof. Heddle in amianthus from Unst, in which "there is generally more or less fibrous calcite between the filaments of the mineral" (' Mineralogical Magazine/ vol. ii. p. 33, March 1878). OPHITES AND RELATED ROCKS. 17 for this reputed fossil. The margins of the layers of serpentine, 0, are each unmistakably converted at the borders into chryso- tile. This allomorph, in some of the layers, is characterized with bright colours ; but in others, where a calcareous layer is in immediate contact, the fibres are more or less separated, and thus converted into a " uummuline wall," as shown by the dark- coloured lines between them. It will also be observed that the fibres have been abruptly bent where they are acicular, which, though evidently due to pressure, causes them to appear as if they were structures distinct from the chrysotile fibres : this abrupt bending, however, has not always taken place; for it often happens (as occurs in other specimens we have figured *) that such fibres pass into the aciculse without losing their con- tinuity of direction. In some places the fibres are undulating or curved, as has been observed in numerous instances that have passed under our notice. The serpentine layer on the right- hand side of the figure appears to have been in a fluidal condi- tion, and on consolidation to have become cracked : the cracks are brightly and variously coloured, indicating the presence of another substance (peridote) hereafter to be noticed. Fig. 4, Plate II., represents a portion of a slide from Canada, presented to one of us (T. H. R.) some years ago by Dr. Carpenter. The portion is part of a thick layer of serpentine, bounded on each side by a calcareous layer. The serpentine- layer is transversely intersected by separated and parallel laminae of chrysotile in its various modifications. Certain of the laminae have become openly divided; and the resultant opening filled in with calcite. In the instance represented at «, the divided lamina we hold to have originally consisted of chrysotile : strictly parallel with the others, its fibres not only graduate into closely- packed and separated aciculse, but they pass without break of continuity into the latter variety. Moreover the overlying lamina, b, is in the condition of incipient chrysotile, consisting of separated lines so finely cut as to be scarcely visible except by polarized light. Obviously the calcite which occupies the opening in the lowest lamina had been introduced ; and it is also inferable that the calcitic films between the separated aciculae are nothing more than infiltrations of the same mineral carbonate. Although the last is an instance which maybe safely accepted * Proc. Roy. Irish Acad. vol. x. pi, xli. tig. 2, d. C 18 ROCK-METAMOKPH1SM. as showing how the aciculse forming the " nummuline wall" of "Eozoon Canadense" have originated, we admit, from its posi- tion, and from apparently being no more than a mere crack across a " chamber cast/' that it cannot be recognized as contain- ing genuine examples of aciculse belonging to this ' ' wall/' It is therefore some feeling of satisfaction to us to find that we are enabled to bring forward a case against which no objections of this kind can be urged. A number of specimens, some of which have been added to the geological museum of Queen's College, Galway, were lately received by Mr. Damon, F.G.S., of Weymouth, from Principal Dawson, as indisputable examples of "Eozoon Canadense," col- lected in the type locality, Ottawa. Pig. 2, Plate IX., represents a portion of one of the specimens. It consists of interlamellations of serpentine and calcite in mutual parallelism, the former being often converted entirely, or partly into chrysotile. In many instances a layer of serpentine con- sists of separated but contiguous laminae of chrysotile parallel amongst themselves. This parallelism of arrangement between the calcite,, serpentine, and chrysotile is the general rule in the present specimens, as it is in others we have at different times brought under notice. Of the two layers, coloured green in the figure, the undermost one consists — the lower half of colloidal or amorphous serpentine, and the upper half of typical chrysotile : the uppermost layer is composed of what at first sight might be taken for ordinary serpentine, but which, on close examination with powers of 25 and 40 diameters, turns out to be in the state of incipient chrysotile, being traversed by separated fine filiform cuts, which seem to give rise to a columnar structure. The direction (which slightly deviates from the perpendicular) of the cuts and the columns, it is noteworthy, is the same as that taken by the fibres Of the underlying chrysotile. On the left half of the upper surface of the layer containing typical chrysotile may be clearly seen the fibres passing con- tinuously into a series of definite but closely adhering acicula?, and on the right half into similar acicula3 : here, however, they are distinctly separated by interspaces filled in with calcite. In eozoonal parlance the latter aciculse are " casts of tubuli " in undisturbed relation to the " intermediate skeleton " and its OPHITES AND RELATED ROCKS. 19 integral " nummuline wall/' just as would be the case when the latter was penetrated by the formative pseudopodial pro- cesses of the animal. We cannot conceive how any impartial investigator, having an acquaintance with mineralogy, and in face of the evidences placed before him, can resist the conclusion that this " num- muline wall" is the product of structural changes charac- teristic of chrysotile. By way of disposing of these evidences, the advocates of " Eozoon " have made a leap from Scylla into Charybdis. The " nummuline wall," it is argued, has been altered by crystallization and pseudomorphism, so that it, origi- nally lime, has been converted into chrysotile or its modifications. Nay, our " complicated theory of metamorphism " has actually been adopted by Dr. Dawson to explain this " change of calcite into serpentine" and its allomorph, chrysotile — which change it appears, he has seen in some specimens of " Eozoon"* \ As to the argument based on the idea that the " nummuline wall " has become accidentally associated with the chrysotile, the fact is of such common occurrence, as are also the concomitant parallelism of the layers showing the two modifications, and the continuity both laterally and lengthwise of their respective aciculas and fibres, as to completely destroy this and any other argument offered in support of eozoonism, based on these considerations. A few more remarks on this specimen. Mention has already been made of the serpentine often passing into the flocculent state. In the specimen which has supplied us with the portion last considered, flocculite is rather common in the layers of calcite, filling them here and there (as near the bottom on the left side of the figure), or bordering the serpentine enclosing these layers — occurring therein as simple or segmented clotules, which here and there graduate into configurations varying from the simplest rods to much-divided or branching shapes. It is necessary to mention that both the clotules and configurations enclose portions or cores of serpentine. The configurations have been taken for casts of tubes pene- trating the calcitic layers, and to represent the canal- system present in the calcareous skeleton of certain existing foramiuifers, But their characteristic irregularity of form, their gradation into the clotules, and their agreement in composition with the * See Introduction, A.I>. 1878. c2 20 ROCK-METAMORPHISM. latter completely prove that they are a genetically identical series, and of purely mineral origin, having no relation to organic structures. No field-mineralogist can look upon the flocculite without seeing at once that it is disintegrated serpen- tine undergoing waste or removal, such as may be seen on rocks of this substance where they are exposed to atmospheric action. Now, disintegrating forces (produced, doubtless, by chemical reactions) having set in, the flocculite of itself must necessarily become more or less a prey to dissolution. Hence we feel certain that the configurations, whether simple, or branching in their form, have been etched out of serpentine or flocculite by chemical processes dissolving or removing this substance : in short, they may be confidently pronounced to be nothing more than "figures de corrosion " of French mineralogists. The beautiful example of " eozoonal canal-system " given in the right corner of the bottom of our figure, though not exhibiting a complete series of the formative changes, is nevertheless of considerable interest as showing it to be an integral portion of a border of flocculite which is significantly penetrated by extensions or lobes of serpentine from the adjacent parent layer*. Configu- rations of the kind are generally white and opaque, with a core of green serpentine ; but frequently they are colourlessly trans- lucent, as is the case with two or three of the branches in this example. The process which converted the serpentine into its flocculent variety by entirely removing the latter would affect the plates, lenticuloids, and spheroids, causing their surfaces to be irregu- larly corroded into hollows and projecting lobes. These bodies are regarded as "casts" of the " chambers of eozoon;" but, holding to the view above given, we have no hesitation in rele- gating them, in respect of their origin, to the category of mineral structures. In some instances the flocculite has not been removed (see left side of the figure) , but remains as a layer between two other layers of serpentine. It will be seen hereafter that there is strong presumptive * The example does not belong to the place in which it is represented, but to one close by : it is truthfully represented in its relation to the adjacent border of floccidite. Another example, represented in a former paper (Proc. EOT. Irish Acad. vol. x. pi. xliii, fig. 7), is an intermediate modification. OPHITES AND RELATED ROCKS. 21 evidence that thermal waters aided the assumed chemical reactions. Having now disposed of all the serpentinous structures of " Eozoon Canadense " in strict conformity with the mutations known to characterize the minerals composing them, we shall next offer our explanation of the origin of the calcitic layers which form the " intermediate skeleton " of this reputed fossil. With respect to the interchanges between a mineral silicate and a mineral carbonate, often obtaining in ophite, we find that the calcite which frequently holds a place in the layers of chry- sotile is fibrous in certain instances, and retains the original structure of the latter mineral. Generally, however, the charac- teristic rhombohedral cleavage of calcite is developed. This replacement of chrysotile by a mineral carbonate retaining its typical fibrosity, and therefore possessing a fibrous structure similar to that of aragonite, has occurred to us in specimens of ophi-euphotide from the north of Italy, as already noticed in one of our previous memoirs*. We have also become acquainted with a similar fact occurring on the shore east of the Lizard, Cornwall, where serpentinyte, undergoing change, contains layers of different colours — white and red. Some are coarsely fibrous, others amorphous, a few rudely laminated : often the different kinds may be seen passing into one another. It is only of late that we have observed in the euphotide or " Crouza stone " of the Lizard another modification, consisting of dull white, also blue fibrous layers, with occasionally an imperfect platy structure : in many instances the white layers may be observed assuming a silvery lustre and changing into chrysotile. Moreover, specimens of the latter kind, after having been slightly acted upon by dilute hydrochloric acid, show here and there vacant spaces between the fibres, and other evidences of the removal of a mineral carbonate. The mineral silicate, malacolite or white augite, undergoes similar changes. A variety of ophite occurring in Connemara contains layers formed of crystalloids of this mineral. In most instances, besides being widely gashed, and the gashes filled up with calcite, the crystalloids are separated from one another by the same mineral carbonate ; and they exhibit their angles and * Proc. Roy. Irish Acad. vol. xx. pi. xliv. fig, 9. 22 ROCK-METAMORPHISM. edges more or less rounded off, evidently by corrosion. Decal- cification exposes these facts most instructively*. The origin the calcite must be obvious to any one who has studied pseudomorphism. The same phenomenon is displayed, though under somewhat modified circumstances, in the hemithrene ("calcaire saccha- roide/" Delesse) of certain localities in the Vosges. At St. Philippe, near Ste. Marie aux Mines, this rock is filled with crystalloids of malacolite and other mineral silicates, often almost to the exclusion of calcite. Confining ourselves for the present to the malacolite, its crystalloids, or aggregations of them, are more or less affected by corrosion or decretion, beginning with the rounding-off of their angles and edges; next, reducing the aggregations, it forms them into rude, irre- gular, geniculated shapes; and next etches them into some- what definite configurations — foliaceous, dendritic, plumose, radiate, and often beautifully arborescent. The configurations vary much in size, some being observable with a hand mag- nifier, others so minute as to be only made out by a good microscope. Fig. 1, Plate III., represents one of the aggre- gations, which has taken the shape of a branching configura- tion f: its component crystalloids, in a corroded condition, are well seen. Guided by a remark made by Delesse, we expressed our suspicion some time ago that the ' ' calcaire saccharoide " of the Vosges would yield on examination these and other structures J; but we had no idea that it was so rich in examples rivalling, and in no way surpassed in beauty and imitativeness, the configura- tions (" canal system viEozoon") which at that time had become known as occurring in Canadian ophite, — though since then we have published the fact, previously unknown, that precisely the * Heddle mentions what appear to be similar examples, occurring at Muir and Midstrath, in which the "lime is little more than granular malacolite with but little lime between the crystals " (see Trans. Roy. Soc, Edin. vol. xxviii. p. 461). t Zirkel has represented a portion of a crystal of mica (fig. 35, p. 87, ' Die mikrosk. Beschaffenheit d. Min. u. Gesteine ') which, through corrosion or decretion, has assumed a dendritic or branching form. This example shows very well how crystals and other mineral bodies have taken the remarkably imitative shapes often displayed by malacolite and serpentine. J Quart. Journ. Geol. Soc. vol. xxii. p. 188, footnote. The " nummuline wall " is also present, See Introduction, A.D. 1880. OPHITES AND RELATED ROCKS, 23 same bodies are present in hemithrenes from Ceylon, Aker (Sodermanland in Sweden), and New Jersey*. The Vosges is an additional locality, which we have only of late become acquainted with. The question naturally suggests itself as to how the separated aciculse in pectinated chrysolite, the arborescent configurations in flocculite and malacolite, and the plates, lenticuloids, and spheroids in serpentine have been produced. From the evidence, so far adduced, our view will have been to some extent antici- pated— that it is by chemical reagents involving the removal of serpentine or other mineral silicates, and their replacement by calcite or other mineral carbonates. Chemical changes of the kind, known as pseudomorphism, are not uncommon in the mineral kingdom. They may, for our purpose, be arranged under two heads — entire and partial, The first consists of cases in which all the original constituents of a mineral having been abstracted, are substituted by other substances ; the second consists of cases in which the removal of certain constituents of a mineral, and their replacement by other substances, have taken place. As an example of the first, crystals are found in Cornwall consisting of cassiterite or oxide of tin ; but, instead of repre- senting the form proper to this mineral (viz. a modification of a square prism), they occur under a false form — the one that cha- racterizes orthoclase, which is a silicate of alumina and potash: in this instance an entire change of substance has taken place. The second may be illustrated by selenite — a hydrous sulphate of lime, well known as crystallizing in right rhomboidal tabular crystals ; but occasionally such solids are found consisting of carbonate of lime, the basic constituent remaining, but the acid and water eliminated, both having been replaced by carbonic acid. Karstenite, an anhydrous sulphate of lime, when con- verted into selenite, as it often is, is also a partial pseudomorph ; though the change has been effected simply by the admission of water f- * Geological Magazine, vol. x. no. 1, January 1873. f Because a few cases have occurred of what appears to be one mineral enveloping another without change of crystalline form (e. g. prisms of anda- lusite ferruginated into a substance having the composition of staurolite), Dr. Sterry Hunt has " confidently affirmed that the obvious facts of envelop- ment," which have led Delesse to limit pseudomorphism, as advocated by £4 ROCK-METAMORPHISM. Serpentine, which has never been found in a crystalline form proper to itself, not unfrequently occurs in forms characteristic of other minerals., as peridote, augite, hornblende, chondrodite, phlogopite, garnet, diallage, spinel, feldspar, &c. In the case of a mineral so prone to assume false forms as serpentine is, it might be expected that some of the varieties and related species would also display the like protean character. As cases in point, loganite and picrosmine are pseudomorphs after horn- blende ; while crystallized rensselaerite and pyrallolite occur in the form of augite. The opinion that serpentine is in all cases a chemically changed or secondary product involves the idea that its sub- stance is a soluble compound. It is commonly stated, however, that silicate of magnesia, the substance in question minus H20, is insoluble— that it is one of "the most stable" compounds*. Hence there are some who assume that serpentine cannot be affected by ordinary chemical reagents. In denying the solubility of silicate of magnesia, it does not follow that serpentine is also insoluble : besides, in this con- nexion, there are some other considerations not to be over- looked. 1st. Serpentine is a hydrous silicate of magnesia, generally adulterated with alumina, protoxide of iron, or other acces- sories (see ante, p. 4), the presence of which make it liable to chemical reactions. 2nd f. It has existed under conditions of pressure and tempe- rature capable of materially augmenting the potency of the weakest dissolving agent that may have been in contact with its constituents. Blum, " are adequate to explain all the cases of association upon which this hypothesis of pseudomorphism by alteration has been based.'' Moreover, although there may occur occasionally serpentine surrounding a nucleus of peridote, such as led Scheerer to imagine that it " was a case of envelopment of two isomorphous species " (serpentine has no proper crystalline form of its own', therefore no other species can be isomorphous with it), it may be regarded as certain, taking the consensus of opinion entertained by the highest authorities, that the frequent entire replacement of the latter mineral by the former is solely the result of chemical changes. * Bischof's ( Chemical and Physical Geology/ vol. ii. p. 113. 1 We speak of the condition of serpentine before it became exposed at thg earth's surface, OPHITES AND RELATED ROCKS. 25 3rd. It has been buried at great depths, and presumably in a somewhat softened condition — thereby rendering it, especially where fibrously divided or intersected by cracks (as it frequently is), an easy prey to dissolving and decomposing agents. Moreover it is generally admitted by chemical geologists that thermal waters containing carbonic acid, or a carbonate in solution, are present in deep-seated rocks. It must also be considered that the terms soluble and insoluble are merely con- ventional, being applied to substances only known as such in the laboratory, and that they do not preclude the idea that substances capable of resisting powerful acids at the earth's surface maybe unable to resist the weakest when existing under the conditions stated. Notwithstanding the difficultly soluble character of silicate of magnesia, it has been " proved " by Bischof that an artificial preparation of the kind c ' when dissolved in water by carbonic acid " is decomposed : he thinks, however, that this ' f is owing to the silica being in the soluble modification." While " the natural silicate of magnesia (steatite, the silica of which is in the insoluble modification) , when finely powdered and suspended in water containing carbonic acid for twenty-four hours, did not show the slightest effervescence"*". But even absolute decomposition of serpentine was proved some thirty years ago by the Professors Rogers, who submitted several mineral silicates (the present one being of the number) to the analytic action of carbonated, and even simple water : the result in every case was a residue of magnesian and other carbonates, showing that decomposition had taken place f. There is one mineral, peridote, which is frequently converted into serpentine. If it were a pure silicate of magnesia, all that would be required to effect such change would be the hydration of the latter compound ; but as it contains a considerable quan- tity of protoxide of iron, this may have been peroxidized, or changed into a carbonate, in the one case by the addition of oxygen, in the other by carbonic acid contained in the penetrating water, a portion of the magnesia being removed at the same time. The process is so simple as to excite surprise that it has been unnoticed by the advocates of the envelopment doctrine. * Chem. and Phys. Geology, vol. i. p. 3, vol. iii. p. 164. t Edinburgh New Philosophical Journal, vol. xlv. pp. 163-108 (1848), 26 ROCK-METAMORPHISM. But, having in view the mineral changes which chrysotile and the associated serpentine undergo in the process of developing the separated aciculse and arborescent configurations of ' ( Eozoon," we are more especially concerned with the fact that these changes have been accompanied by the substitution of calcite for a hydrous silicate of magnesia, unassociated with calcium in any form of combination. A very simple modus operandi in this case would effect the substitution, nothing more being required than for water, holding carbonate of lime in solution, to penetrate into the linear openings of the chrysotile and the cracks of the serpen- tine : a portion of the silica and magnesia of these minerals would be removed in the soluble state, the carbonate of lime taking their place. According to this view, the calcite between the acicnke in a layer of pectinated chrysotile, also that in which the configurations are imbedded, has been conveyed thereto in solution from an independent or extraneous source — that is, from the calcium silicate common to minerals of metamorphic rocks. Feldspars, hornblendes, augites, and several other minerals, it is well known, are pseudomorphosed by the carbacidizing of their calcium silicate. Crystals of hornblende occur filled with calcite (Bischof, op. cit. vol. ii. p. 315); labradorite has been found having some of its own constituents replaced by the same mineral ; augite is frequently more or less calcitized ; it is the same with garnet, epidote, wollastonite, and others. As in most of these cases the pseudomorphs contain more lime than conld be yielded by the calcium silicate of the original mineral, it may be assumed that the solvent by which the latter were affected supplied the additional quantity. Even the minerals forming syenite, diorite, and other rocks, in situ, on becoming subject to carbacidizing processes are re- placed by calcite. We have diorite from Jersey: amidst its component horn- blende &c. there frequently occur patches of calcite associated with epidote and serpentine, Little or no doubt can be enter- tained as to the hornblende in this case having supplied the magnesia for the serpentine. It is therefore highly probable that the calcium silicate of the hornblende has been transmuted, through the introduction of carbonic acid, into the associated calcite, especially as there are no limestones nor other cal- careous rocks in the island. In one place near Galway, the OPHITES AND RELATED ROCKS. 27 glacial drift has been broken up, and its finest materials, clay and sand, washed off — leaving a mass of pebbles, cobbles, and large erratics, consisting of granite, syenite, and other rocks, piled up into an esker. This deposit has lately yielded to our examination several pebbles and blocks of syenite, in which are veinlets and isolated patches of calcite, obviously the result of changes effected in the component hornblende and oligoclase, — these minerals, where the calcite is present, being more or less corroded, and having a spongy structure. Minute patches of serpentine associated with the calcite are also occa- sionally seen. When the calcite has to some extent been removed by dilute acid, there are generally left siliceous bodies in the form of rude arborescent configurations rising out of the remaining calcitic matrix, — in one instance strikingly like those that are common in the cases elsewhere mentioned. Later investigations on the massive diorite, which has been ex- cavated during the past year (1880) in the construction of the new dock in Galway harbour, have yielded us a number of spe- cimens containing calcite in abundance, the secondary origin of which is indisputable. We have also found very recently in situ, near Salt Hill, Gal- way, a porphyritic feldsyte more or less serpentinous, at a spot where a quarry has been opened in a fruitless search for copper- ore. The serpentine is generally seen lining fissures, often superficially but occasionally to a depth of a few inches : it also occurs in detached pieces of rock lying about. A loose block a cubic foot in size, and altogether serpentinous, was taken out of an adjacent wall. The mass — greyish, brownish, and olive- green in colour, has in some parts quite a soapy feel, an oily lustre, and a coarse fibrous or slickenside-like structure, resem- bling in these respects baltimorite and pyrosclerite. These are examples of changes effected by chemical reactions in igneous rocks, certain of whose original mineral substances having undergone replacement by serpentine and calcite *. We are now, however, trenching on chemical changes that have taken place in rock-masses — a subject which properly belongs to the next Chapter. * Specimens of the examples referred to ore deposited in the geological museum of the Queen's College, Galway. 28 ROCK-METAMORPHISM. CHAPTER V, MINERALIZED AND METHYLOSED METAMORPHIC ROCKS, METAMORPHIC rocks may be divided into two groups — mine- ralized and methylosed, differing from each other, the one in having had the original substance of its members crystallized into minerals of various kinds, and the other in having had the same minerals altered or replaced by chemical reactions. The name pseudomorphosis, occasionally applied to the last group, is inappropriate, as the rocks to which it has been given possess no form to be imitated, and therefore no false form is involved. Influenced by this consideration, we have of late years employed the term methylosis in the case of metamorphic rocks which have had more or less of their minerals transmuted*. Ophites, which we include in the methylosed group, are so intimately related to the mineralized metamorphic rocks, that, in treating of the origin of the latter, the same subject in regard to the former forces itself on our consideration. Passing over the various views that have been held on the origin of the mineralized metamorphics, from the remarkable one held by Leibnitz in his < Protogsea' (1717) to the latest, as set forth by Dr. Sterry Hunt f, we propose to consider the latter, it having been contended for with a persistency and an array of argumentation that have won, if not their conviction, seem- ingly the favourable consideration of many geologists. * The term methylosis (/iera-, change, and v\rj, substance) was first pro- posed in my paper <{ On a Silo-carbacid rock from Ceylon," published in the * Geological Magazine,' vol. x., January 1873. The term metasomatosis (/zera- o-aj/iarwens), applied to the same class of rocks by Von Lasaulx and Knop, is of subsequent date, and had already been employed by myself in a memoir "On the Trimerellid» " (Quart, Jonrn. Geol. Soc." No. 118, p. 140, 1874), in which I am associated with Mr. T. Davidson. — W. K. t The principal views are given by Delesse in his memoirs on metamor- phism in the ( Annales des Mines/ s£r. 5, t. xii. 1857, &c. MINERALIZED AND METHYLOSED ROCKS. 29 Notwithstanding the weight of authority on the side of Vulcanism, Sterry Hunt maintains what he calls a " novel doctrine," but which seems to be similar to the one taught by Werner, and accepted to some extent by DelaBeche and others, — that the vast masses of ancient crystalline rocks known as " Azoic/' ' ' Fundamental/' " Laurentian," " Eozoic," and <( Archaean/' have been " directly deposited as chemical preci- pitates from the seas of the time " * ; to be particular, that the Canadian Archseans, comprising granitoid gneisses, syenites, chlorite-, talc-, mica-, and hornblende- schists, and ophites, have had their component minerals (steatite, serpentine, talc, chlorite, phlogopite, augite, hornblende, orthoclase, labradorite, quartz, epidote, and other species f) " formed, not by subsequent meta- morphism in deeply buried sediments, but by reactions " J, " by a crystallization and molecular rearrangement of chemically formed silicates, generated by chemical processes in waters at the earth's surface " §. As our reasons have been given elsewhere for decidedly rejecting this doctrine, it being altogether unsupported by ac- ceptable evidences ||, there is no necessity for us to do more on * Canadian Naturalist, n. s., vol. iii. p. 125 (1860). t As hemithrenes are Archaean rocks, culcite, inieniite, and some other mineral carbonates ought to be added to the list. | Quart. Journ. Geol. Soc. vol. xxi. p. 70 (1865). § Geological Survey of Canada, Report, 1866, p. 230. Sterry Hunt/ in the preface (p. 20) of the second edition, 1879 (the latest), of his Chemical and Geological Essays, expresses himself thus : — " The crystalline stratified rocks were originally deposited as, for the most part, chemically formed sediments or precipitates, in which the subsequent changes have been simply molecular, or at most confined to reactions, in certain cases, between the mingled elements of the sediments." l| See 'Proceedings of the Royal Irish Academy,' vol. x. p. 540. Sterry Hunt has adduced in his favour the existence of Tertiary sepiolyte in the Paris basin and at Vallecas near Madrid, " together with the formation, at the present time, of a hydrous silicate of alumina and magnesia, named neolite, a deposit from the waters in certain mines," and probably resulting from the " decomposition of the magnesian minerals hornblende, augite, and talc." But both cases may be safely set aside as totally inappo- site. The sepiolyte, instead of being a "direct chemical precipitate,1' has been shown by Dr. Sullivan, President of Queen's College, Cork, and Professor J. P. O'Reilly to be a secondary product, due to chemical alteration of the original deposit ('Notes on Spanish Geology,' p. 171, 1863) ; and as 30 ROCK-METAMORPHISM. this occasion than to briefly show that it is totally invalidated by Dr. Hunt's own dicta. Thus, quoting from a lately published exposition of it (in ' Nature/ No.460; Aug.22,1878), we proceed :— " Plutonists begin to understand that water cannot be excluded from rocky strata,, but is all-pervading, and that at great depths, kept by pressure in a liquid state at an elevated temperature, and having its solvent powers augmented by alkaline salts, it plays a most important part in metamorphism." Nevertheless, from the absence of any references to the matter, the minerals forming the Archaean stratified crystallines cannot, as must be understood from the quotations in the preceding page, have been formed in " L ..:. Thus the Glassillaun rock is an intimate mixture of mineral silicates and a mineral carbonate ; and as such it must be con- sidered to be a true hemithrene. Mr. Twining' s observations, kindly made at our desire, make the promontory 200 feet in width, and projecting 40 feet into the water at high tide. The hemithrene occupying the ditch- like channel becomes on the west side of the promontory a more dilated mass, much of which, except at the edge of the cliff, is obscured by a growth of seaweed, boulders, and beach- sand. An opportunity will occur hereafter, enabling us to account 1'or the presence of peridote in methylosed rocks; we may therefore confine ourselves to the serpentine and Calcite of the hemithrene in this instance. The serpentine is so associated with the tremolite as to make it evident that the former has resulted from chemical changes effected in the latter ; and there is nothing more certain than that the calcite stands in the same relation to the malacolite. That water .has accompanied the changes is proved by the presence of serpentine in the hemithrene; and the fact that HEM1THRENES AND OTHER CALCITIC ROCKS. 51 calcite occurs in the dyke may be taken as positive evidence that this mass has also to some extent undergone a chemical transmutation. In short, we feel it a safe conclusion that the change in both rocks is due to their having been penetrated by heated water containing a carbonate or carbonic acid in solution*. In addition to the important evidence which the hemithrene of Mont Saint-Philippe has afforded in connexion with the chemical changes effected in minerals, we have next to make known some facts from the same place which bear directly on the matter under consideration. The region of the Vosges, which embraces Wisembach, Chippal near Croix-aux-Mines, Laveline, Gemaingoutte, and Mt. St.- Philippe near Sainte Marie-aux-Mines (Haut-Rhin), consists principally of gneiss, with here and there intrusive masses of igneous rocks (syenites and dolerites) ; but in the places named there is a development of hemithrene. A large quarry of this rock (" calcaire saccharo'ide") is worked at St. -Philippe. It is a nearly rectangular excavation, about 60 or 70 paces long, 30 paces wide, and 15 feet in depth. The rock, of a highly crystalline character, is for the most part well strati- fied, as is also the associated gneiss : both dip in the same direc- tion (south-east), at an angle of about 20°. The hemithrene, light in colour, is more or less charged with pyrosclerite, chiefly in granules ; and the same mineral occurs, but rarely, in the immediately adjacent portions of the gneiss. The pyrosclerite and other mineral silicates, as previously notified, are irregularly intermixed with calcite ; but very often the different kinds (car- bonates and silicates) are disposed in layers : on account of the latter arrangement the beds have assumed a laminated structure. The lamination is often irregular, being variously undulated, and separating here and there through the interposition of com- pressed lumps which consist internally of a white granular or * Quart. Jourii. Geol. Soc. vol. xvii. p. Iviii. A short time ago Brcegger and Reusch observed at Hiasen, in Norway, " a vein of hornblende changed into a mass of calcspar" (Canadian Naturalist, n. a. vol. viii. p. 430). Both this case and the calcified doleritic dyke at Cleggan may be taken to prove that the well-known calcitic dyke, traversing nietaniorphic rocks near Auer- bach in Bergstrasse, first described by Von GEynhausen more than fifty years ago, was originally a silacid igneous mass, and since converted into hemi- threne : its accessory minerals (hornblende, treniolite, idocrase, wollastonite, epidote, and the like) favour this view. E2 52 &OCK-METAMORPHISM. amorphous mineral, considered by Delesse to be feldspathic and related to halleflinta : this part passes insensibly into a coat of pyrosclerite ; and the latter is surrounded by phlogopite. These lumps have consequently a concentric structure. The gneiss, dark in colour, consists of layers of opaque ortho- clase, translucent plagioclase, dark bronzy mica, and dark green hornblende : quartz is more or less irregularly intermixed with these minerals ; also, as accessories, garnet, pyrites, graphite, and sphene. Where the gneiss and hemithrene are in immediate contact, their respective minerals are intermixed ; nevertheless both rocks are well differentiated, and readily distinguished from one another. The hornblende unmistakably changes into pyrosclerite, as is also the case with the augite ; and the least reflection must make it clear that the two minerals plagio- clase and malacolite have contributed to produce the cal- cite : besides, according to Delesse, the dark bronzy mica of the gneiss has become magnesiated into the lighter-coloured phlogopite which characterizes the hemithrene. Moreover it is highly probable that the calcite, where it is the dominant con- stituent, has been increased by carbonate of lime brought in by water from extraneous sources. Figure 1, Plate VI., represents a section, about 13 feet in height, showing the gneiss and hemithrene in contact, as seen on the right-hand side of the quarry near its entrance. The beds of gneiss (coloured red in the figure) , it will be seen, come in between two masses of bedded hemithrene (coloured green) rudely jointed. There is no appearance of a fault bringing the two rocks into their present relative position; nor, considering the perfect continuity between their respective beddings (which, it is to be remarked, have a uniform dip), can it be said that the hemi- threne is in detached masses that have been let into a synclinal hollow and by this means preserved in their present position. The bedding of the hemithrene is in other parts of the quarry somewhat obscure ; and so is the limestone, especially where the interposed concretions are present. Delesse, who had favourable opportunities for examining the geology of the Vosges, where the hemithrene occurs, states that the " calcaire saccharoide est toujours completement enveloppe par le gneiss dans lequel il forme des flambeaux irreguliers ou lenticulaires tels que ceux qui ont ete signales dans la Scandinavie HEMITHRENES AND OTHER CALCITIC ROCKS. 53 par MM. Scheerer et Keilhau"*. He also mentions that at Laveline " le calcaire passe insensiblement au gneiss encaissant/ The conclusion we have formed respecting the origin of these tongue-shaped masses of hemithrene will be obvious to the reader. Undoubtedly there is a decided difference between them and the enclosing gneiss ; it cannot, however, have been caused by any mechanical break. Nor can the hemithrene be infolded masses due to a crumpling of the beds ; for the gneiss is plainly seen passing into the hemithrene both vertically and horizontally, the change of colour (dark in the former and light in the latter) affording the clearest evidence of the transition, while the bedding is perfectly even and only slightly disturbed. Even hand specimens may be collected with the two rocks in such union as to make it impossible to detect any thing like a mechanical division between them. With these evidences before us, and taking into consideration various others that have been adduced, including those revealed by the Cleggan dyke, proving rock-forming siliceous minerals to have had their calcium silicate changed into a carbonate, the conclusion becomes inevitable that the hemithrene of St. Philippe is a methylosed product after gneiss. M. Rozel held the opinion that the hemithrene masses of the Vosges are veins of igneous origin f — thus agreeing with Emmons (1842), Mather and Von Leonhard (1833), with respect to corresponding cases made known by each as occurring in other regions. Delesse, on the contrary, conceives that the Vosgesian masses are genetically contemporaneous with the associated gneiss. Our view is that they are due to heated water holding carbonic acid or carbonate of lime in solution, which has penetrated the gneiss through joints and other open- ings, and converted it, where so affected, into dyke-like masses of calcareous marble {. Having shown the convertibility of mineral silicates into calcite and calcitic rocks, the question that next turns up for consideration is whether a similar change has taken place over a large area. * Annales des Mines, 4C se"r. vol. xx. p. 181. t Bull. Soc. G6ol. de France, vol. iii. pp. 215-235. { It is said that the rock at Chippal resembles in its purity Pares marble. I was unfortunately prevented reaching this locality during my visit to the Vosges.— W. K. 54 ROCK-METAMORPHISM. It will not be denied, except perhaps by a few, that earthy sedimentary rocks have been mineralized over extensive regions into ordinary metamorphics ; there is therefore no reason why chemical changes or niethylosis may not have been effected on a regional scale. Bischof, according to the extract previously quoted from his ' Chemical and Physical Geology/ was evidently not averse to this view ; nor would Heddle, we strongly suspect, be opposed to it. Speaking of the euphotide at Portsoy, and the occurrence there of a " very siliceous limestone in immediate contact with it," Heddle proceeds — " Now the frequent association of thin beds of limestone with serpentine supplies very direct evidence of the conversion of hornblendic and augite rocks into serpen- tine. In that fact we have a ready answer to the question , ' What becomes of the carbonate of lime necessarily formed during such an alterative process as the above?' I will not say that limestone is always to be found in such association ; we do not always find limestones even where we have indubitable evidence that they once existed ; for here the very thing that makes can unmake or sweep away. The carbonate of lime thus fashioned out of the rock forms a belt beneath the residual serpentine, thicker or thinner in accordance with the original thickness of the stratum of transformed rock ; also thicker or thinner according to whether that rock was augitic or horn- blendic ; for the former can supply considerably more lime than the latter. This calcareous belt must lie beneath the parent rock, sealed against any great amount of further change, unless or until upheaval or denudation expose it to meteoric influences. Then water, flowing either downwards or upwards, may — nay, in time must — sweep it away in solution, leaving lime- sink or collapsed void to evidence its former existence. But if the limestones, so frequently associated with serpentines, are thus to be assigned to the decomposition of the rock which yielded these serpentines, we have a crucial test of the soundness of the theory of the change, in the inquiry as to whether unchanged gabbro, or other such rocks, occur in contact with lime. That it never does, I will not say : but, in glancing at a sketch geo- logical map which I have constructed of the district where these rocks occur, I find, as regards the great belt of diorite and diallagic rock which sweeps up Central Scotland, that where HEMITHRENES AND OTHER CALCITIC ROCKS. 55 either the limestone appears in contact with it, or a "wash-out discloses its former existence, there the rock is serpentine ; where it appears as unaltered rock there is no lime. "I find, moreover, that wherever the association can be observed, the lime invariably "is beneath the serpentine. So it is with the loch of Cliff lime and the serpentine of Unst, both of the lime and serpentine beds at Polmally, both of the lime and serpentine beds at Portsoy, at Limehillock, Tombreck, the Green Hill of Strathdon, and Beauty Hill ; and in enumerating these I have named all the most important masses in the country"*. We have only one objection to Professor Heddle's view. It would seem that he considers the calcareous matter, resulting from the changes he advocates, to be transferred to beneath the rock from which it was abstracted. But a difficulty strikes us as to what was in the place of the new " calcareous belt" before it was formed. This point seems to have been overlooked. We are quite ready to admit the important bearings of the fact stated by Heddle, that the lime " invariably is beneath the ser- pentine ; " we cannot but suspect, however, that the lime has taken the place of a mass of transmuted serpentine or malacolite. The notable shotted structure which frequently characterizes hemithrenes, and in many instances ophites, has an important bearing on the present question, inasmuch as this structure is due to the presence of crystalloids or portions of different minerals possessing a peculiar exterior. As far as we have been able to ascertain, Macculloch appears to have been the first to bring the matter before the notice of geologists f. According to his description of the Tiree marbles (one mass of which is "an irregular rock — a nodule of limestone, improperly called a bed, lying among the gneiss without stratification or continuity"), they consist of amor- phous or finely granular calcite in pink, white, and other colours, in which are imbedded crystalloids of sahlite, augite, and hornblende, termed in a general way coccolite, separately dispersed, or aggregated — in the latter state attaining the size of an orange or larger. Other minerals are present, as mica, ser- pentine, steatite, sphene, &c. The crystalloids are often either * Op. cit. pp. 540, 641. Bischof mentions the occurrence of beds com- posed of limestone beneath others formed of mineral silicates. See ' Physical and Chemical Geology,' vol. iii. pp. 304, 30o, 307, 308. t Western Islands of Scotland, vol. i. pp. 48-56 (1819). 56 ROCK-METAMORPHISM. partially rounded or entirely shapeless — a peculiarity which Mac- culloch conceived to have resulted "from an incipient solution." Nanmann next drew attention to the rounded form of the crystalloids of augite and other minerals in calcitic rocks, par- ticularly in that at Pargas, in Finland, and attributed it to partial fusion. Emmons, in 1842, referred to the rounded outlines of crystals of certain minerals imbedded in the crystalline limestones of Rossie, New York, and suggested that they were due to a partial fusion *. In 1863 Sterry Hunt briefly noticed the occurrence of rounded grains of various minerals in the Archaean rocks of Canadaf- Our attention having been drawn to the so-called ' c chamber- casts of Eozoon " in 1866, we identified them with the crystal- loids in the hemithrenes of Pargas, Tiree, and New Jersey, also with the lobulated grains of serpentine common in the ophite of Connemara. It was likewise suggested by us that they had been shaped by chemical reactions aided by heated water, which, having affected them superficially, had removed their substance and replaced it by the calcite in which they are now imbedded J. About the same time as ourselves Giimbel noticed similar rocks, also the peculiar form of the crystalloids contained therein; but he adopted the view, then pretty generally received, that the latter were of organic origin §. In his report, addressed to Sir William E. Logan, and pub- lished in 1866, Sterry Hunt, again referring to the subject ||, drew attention to the fact that crystalloids with rounded forms, besides occurring in the bedded hemithrenes of Canada, are also present in "calcareous veins " (some being 150 feet thick) intersecting the latter and the associated gneisses, dolerites, &c. As this fact will have to be noticed again, its further consideration may for the present be deferred. * Geology of the First District of New York, pp. 37-39. Delesse, we are aware, lias noticed rounded crystals ; but, unfortunately, our extract, stating the fact, from one of his memoirs has got mislaid. t Report, Geological Survey of Canada, 1863, p. 592. $ Quart. Journ. Geol. Soc. vol. xxii. pp. 198, 199, 209 (Jan. 1866). § Sitzungslberichte d. Miinchener Akad. d. Wissensch. (Jan. 1866). || We find no reference to the rounded crystalloids in S. Hunt's memoir " On the Mineralogy of certain Organic Remains from the Laurentian Rocks of Canada," published in the Quart. Journ, Geol, Soc, vol. xxi. (Nov. 1864), HEMITHRENES AND OTHER CALCITIC ROCKS. 57 In our paper of 1868 we also noticed the occurrence of crys- talloids of malacolite in ophite from Connemara, which are not only more or less externally eroded and separated by calcite, but have gaping or chink-like cleavage -partings filled with the same substance. The mineralogist, it was asserted, knows full well that originally the cleavage- par tings had individually their two planes in perfect contact ; hence there is no other explana- tion open to him than the one which admits that some solution, charged with carbonic acid or a carbonate, has gained access into the cleavage-partings of the malacolite, and, necessarily reacting on some of its constituents (? calcium silicate) , has generated the infilling of calcite. Reverting to our account of the different minerals in the Tiree marble, it was stated that we had found in this rock •'crystalloids of hornblende, others of sahlite, a few of quartz, and some apparently of serpentine ; while an occasional one appears half composed of hornblende and the other half of sahlite "*. Pretty much the same result has been lately obtained by Heddle, who mentions portions of sahlite in which " an inci- pient as well as a perfected passage into serpentine is seen^f- Since the year 1869 we have described the microscopic struc ture of hemithrenes from the Isle of Skye, New Jersey f, Aker, and Ceylon§. Zirkel, in 1870, noticed the " roundish, sharply- defined, serpentine grains " in Scandinavian hemithrenes, and assumed them to be pseudomorphs. In our papers it was shown that they contain not only rounded crystalloids of malacolite and white pargasite (or it may be wollastonite) , singly and in aggre- gations, but the same things decreted and fashioned into slender shapes, corresponding with the remarkable configurations already described as occurring in Canadian ophite. A specimen from New Jersey contains numerous configurations associated with spinel, a well-developed example being actually imbedded in the calcite occupying a fissure-like opening in a large octahedral crystal of this mineral|| . Specimens of the Skye rock, where it is ophitic^, * Proc. Roy. Irish Academy, July 12, 1869. t Op. cit. p. 459. f Proc. Roy. Irish Acad. n. s. vol. i. (1871). § Geological Magazine, vol. x. Jan. 1873. || Proc, Roy. Irish Acad. vol. x. p. 547 (Feb. 1870). 5[ This rock seems for the most part to be a carrarite j but the portions we have examined are in the state of ophite or hemithrene. 58 ROCK-METAMORPHISM. after being decalcified, showed separated grains or crystalloids of malacolite with a thin white crust enclosing their translucent sub- stance ; others were seen with a portion of the latter removed, but the crust remaining intact ; while close by were hollow spheroidal cases identical in composition with the crusts: before decalcification the cases were filled with calcite. A few examples of the cases occur entirely riddled, reminding one of the per- forated shell of a Polycystine *. In our remarks on these things it was contended, considering the crystalloids of malacolite show themselves in every stage of decretion, that in numerous instances they must have disappeared altogether ; also that it was equally to be inferred that the inter- stitial calcite separating the grains, even that forming the asso- ciated layers, had replaced a corresponding amount of malacolite. From the vai'ious evidences just given, added to those pre- viously adduced, all demonstrating that both minerals and rocks essentially silicates have been converted into calcite and cal- careous masses, we feel it impossible to form any other con- clusion, as regards the hemithrenes so far considered, than that they were originally silacicl metamorphics, also that the asso- ciated marbles, <( consisting of nearly pure carbonate of lime/' are examples in which the mineral silicates have been completely extracted by the same methylotic process. Professor Heddle, it may be said, takes a different view. Referring to the hemithrenes that have come under his own observation, such as occur in Tiree, Harris, Lewis, and some other localities in Scotland, and contending that the crystalloids they contain " are one and all pseudomorphs of preexisting crystals of augite, sparsely sprinkled throughout a great mass of lime, in an amount which is altogether quite trifling," he declares that " these trifling specks could never have been the origin of a lime stratum tens of feet in thickness "f. Whatever may have been the origin of the other hemithrenes referred to by Heddle, we certainly must contend, with respect to the Tiree marbles, that their rounded crystalloids are not only pseudomorphs but also the residue of what was once an augitic rock, and that the portion which has been destroyed, or rather removed, has undergone replacement by calcareous * Proc. Koy, Irish Acad, n. s. vol. i. p. 138 (1871;. f Op. cit. p, 542, HEMITHRENES AND OTHER CALCITIC ROCKS. 59 matter"*. Agreeing with Heddle that it was not the " sparsely sprinkled" pseudomorphosed crystals "which have been the origin of the lime stratum," we nevertheless ask why may not augitic rook-masses, by methylosis and decretion into the ee trifling specks," have given rise to it? Considering also that the marbles of Harris and Lewis are Archaean, like those of Tiree, we have no hesitation in placing them in the same category. Professor James D. Dana, as it seems to us., advocates the methylotic origin of hernithrenes ; but he takes a different view from ours as to the modus operandi of the process. In his description of rocks of the kind common in Westchester County, New York, he argues that they have been originally limestones or dolomites, which, through the action of hot silicic solutions, have become silicated, their mineral carbonates (calcareous and magnesian) being thus converted into mala- colite, tremolite, and other mineral silicates f. We have no objections to urge against changes of this kind having occa- sionally taken place. But in the rocks to which we refer the evidences are so palpable and prevalent of the calcite having replaced malacolite, that we feel assured, if Prof. Dana were to decalcify specimens of the Westchester rocks, he would at once see the force of our view in its application to his par- ticular instances. Another objection lies against the idea that hemithrenes are silicated limestones, inasmuch as, if applied to the Archaean deposits of the kind, it would make the latter to have been even more calcitic than they are at present ; and this would increase the difficulty in solving the problem (to be dis- cussed in a subsequent Chapter) as to the source which supplied the calcite. * The Tiree marble evidently contains much more augitic residue than is represented by the crystalloids, as there are frequently imbedded in it siliceous bodies of various sizes : the crystalloids are also often surrounded by a white, spongy, siliceous covering (corresponding to flocculite), to be seen after decalcification. Macculloch has noticed something of the kind (op. cit, vol. i. p. 53). t American Journal of Science, 3rd ser. vol. xx. p. 28, &c. The origin of these rocks will be further noticed in Chapter XIII. 60 ROCK-METAMORPHISM. CHAPTER XL ON THE ORIGIN OF THE MINERALS CHARACTERISTIC OF OPHITES AND RELATED ROCKS, THE MINERAL PERI- DOTE IN PARTICULAR. THE object of this Chapter is to show that the minerals referred to in the heading are the direct products of hydrothermal reactions in methylosed and volcanic rocks, also of pseudomor- phism in minerals characteristic of ordinary metamorphics and different kinds of igneous masses. If this object can be accom- plished, it will necessarily follow that all the minerals under notice are, in a certain sense, secondary products. Several of these minerals, such as calcite and the different serpentines, are generally admitted to have had this origin; there are others, however, which, as far as we can ascertain, have never been considered any thing else than original in the same sense as the quartz, feldspar, and hornblende of granites are held to be so. At first sight the occurrence of secondary minerals in volcanic rocks seems improbable. Still it is pretty generally considered that many of the minerals found in dolerites, trachytes, and ordinary lavas have been generated under circumstances, as to time and conditions, totally different from those under which the rocks in question were formed. Even granite, generally assumed to consist of original minerals, occasionally contains zeolites, which all must admit to be secondary products. The same must be asserted of calcite and serpentine, already notified as present in the diorite and other rocks near Galway and in Jersey. It is necessary to mention, in the next place, that numerous instances must occur of a rock that has undergone partial me- thylosis, having some of the minerals it contained in its previous or primary condition (whether such rock be igneous or ordi- nary metamorphic) retaining their original character, and others partially altered. Thus the hemithrene of St. Philippe carries hornblende and muscovite ; but as these minerals lie immediately adjacent to the contact rock, viz. gneiss, there can be no doubt that they have escaped the changes which converted the same species accompanying them into malacolite and phlogopite. ORIGIN OF MINERALS CHARACTERISTIC OF OPHITES ETC. 61 It has been clearly shown that, besides other minerals, mala- colite and serpentine — demonstratively secondary products — have themselves undergone chemical changes which, gradually removing certain of their constituents, especially the silacid, have terminated in calcitic or miemitic (dolomitic) replacements. With respect to several other minerals characteristic of ophites and hemithrenes, such as spinel, idocrase, apatite, sphene, &c., it may be assumed that, although original in these cases (that is, formed independently of any other mineral), they are neverthe- less of secondary origin, having resulted from the intervention of hydrothermal reactions. Another mineral of importance in the subject under conside- ration is peridote, also known under the name of olivine *. It runs into several varieties, as limbilite, chrysolite, glinkite, bol- tonite, olivinoid, hyalosiderite, &c. Analyses make it to consist of magnesia from 32 to 50 per cent., protoxide of iron from 6 to 30 per cent., and silica from 31 to 44 per cent. Accessory ingredients, as alumina, oxide of nickel, oxide of manganese, titanic acid f, and silicate of lime, are not unfrequently present. The magnesia is often about 13 per cent., generally 10 per cent., and occasionally 2 or 3 per cent, in excess of the silica. Silicate of lime is present in specimens of peridote from the lava of Fogo (Cape-Verd Isles) to the extent of nearly 6 per cent. Monti- cellite is a highly calciferous peridote, containing 34 per cent, of calcium silicate. The minerals nearest to peridote are humite (which contains, in addition to the essentials of peridote, between 2 and 3 per cent, or more of fluorine) and chondrodite, in which the fluorine is increased in some cases to more than 7 per cent. Leipervillite (bronzite according to Pisani) only differs from peridote in its magnesia reaching to 75 per cent. Forsterite, another related species, but containing only a few per cent, of protoxide of iron, carries us on to diaclasite, enstatite, and certain reputed diallages and hypersthenes : in these last the silica is in excess of the * D'Argenville's name peridote (date 1755) lias priority over that of Werner's olivine (1790) ; obviously, then, French mineralogists are right in adopting it. Pliny's name, chrysolite, it would appear, was not applied to the mineral now so called by mineralogists (see Dana's ' Mineralogy,' under "Olivine"). t A titaniferous peridote from Zermatt, according to M. A. Damour, contains 5-30 per cent, of titanic acid (Bull, de la Soc. Mineralogique, t. ii. pp. 15, 16). 62 KOCK-METAMOKPHISM. magnesia. Fayalite is usually classed with peridote, though containing only a small quantity of magnesia. Serpentine is another chemically related mineral, with the dif- ference principally in being hydrous. It frequently contains iron protoxide, by which its relation to peridote is further sustained. The frequent change of peridote into serpentine, which we have already noticed, has no doubt largely contributed to the prevailing idea that the former mineral is always an original product. Taking this view, some difficulty would be felt by conceiving that peridote itself could occur as a product of pseu- domorphism. Hence it is, in the case of a rock containing peridote associated with augite or hornblende, that the idea of the first having been pseudomorphically generated out of either of the latter two does not seem to have been entertained. Rocks containing peridote have been called peridolytes, the principal of which are Iherzolyte, dunyte, and picryte. Peridote is of somewhat common occurrence in many dole- rites, trachytes, and lavas. Its presence is well known in what may be an igneous rock, — the hypersthenyte near Elfdalen in Sweden*; and it is said to occur in the granite or syenite between the Nile and the Red Sea. Delesse discovered a ferru- ginous variety of peridote in cavities of the pegmatite or granite of the Mourne Mountains f. The related species, fayalite, has been found in ordinary metamorphics (those simply mineralized) at Tunaberg in Sweden, forming, with augite &c., a bed, called eulysyte, in gneiss. A true peridote dominates in a rock, called olivenyte, associated with talc-schist, at Uddevalla, Sweden ; a variety has been observed near Kyschtimsk, north of Miask, and near Synersk in the Ural in talcose rock : these cases show that the mineral is also present in methylosed metamor- phics ; while its frequent occurrence in serpentine, made known by Breithaupt, Bischof, Hunt, Otto Hahn, and others, leads to the same conclusion. As it is found in methylosed rocks, the presence of peridote in veins of calcite, presumably of secondary origin, traversing large blocks of talcose schist scattered over the southern moraine of the Findelen glacier near Zermatt, is not surprising to us, though it must be to those who subscribe * This rock is said to consist of hypersthene, labradorite, and peridote. t Reference to the fact is unknown to us ; but the occurrence of peri- dote in the Mourne granite is mentioned by Dr. Haughton in Quart. Journ. Geol. Soc. vol. xii. p. 191. ORIGIN OF MINERALS CHARACTERISTIC OF OPHITES ETC. 63 to the prevailing idea just stated. But it ought to be greatly embarrassing to the latter to find that it notably occurs (as bol- tonite) in the hemithrene of Massachusetts, also as fe roundish sharply defined grains/' more or less changed into serpentine, crowding the corresponding rock (" granular limestone >}) at Snarum, Aker, and Modum in Norway*. To add to its remarkable indifference as to the nature of the material selected by peridote for its matrix, the fact may be mentioned that this mineral, or some of its varieties, besides being often in association with other substances, exists imbedded in the nickeliferous iron of meteoric amygdaloids that have fallen in Siberia, Atacama, and elsewheref ; and it is a common ingre- dient in meteoric stones. We have shown that peridote is a variable mineral through the increase or decrease of the percentage of its magnesia and iron, also through the presence of accessory substances. Thus, let it be conceived, in the case of a rock like ophite, which has often resulted from hornblende and augite, that during the process of alteration the protoxide of iron belonging to either of these minerals were not removed, there seems to be no valid reason against the probability that this substance would become united, all things being favourable, with any available free sili- cate of magnesia, and thus form peridote or some variety of it. The two minerals peridote and serpentine can be readily differentiated by the polariscope when they are associated as in ophite. Guided by the peridote in the hypersthenyte of Elfdalen (which is optically in agreement with the same mineral from Expailly, in Auvergne, and several other localities), and allowing for variations due to circumstances such as have been mentioned, it may be stated that this mineral exhibits a display of the richest colours by rotating the prisms, viz. yellowish green, sap- green, pink -rose, ruby, blue, &c. — being a greater variety and of more brilliancy than those displayed by chrysotile. On the contrary, serpentine only shows pale yellow (under parallel prisms), changing into dark grey (when the prisms are crossed), and back again into pale yellow. The specimens of peridote which we are about to notice exhibit * Neues Jahrbuch fur Mineral ogie, 1870, p. 828. f As meteoric peridote is generally believed to be of extramundane origin, probably solar, obviously any endeavour to explain its origin may be properly avoided. 64 ROCK-METAMORPHISM. the beautiful green, ruby, and other colours that especially distinguish it. Rather often may be detected by polarized light septa and strings of this mineral loosely and confusedly interreti- culated with serpentine. In fig. 4, PL III., we have represented a short plate (" chamber-cast " of " Eozoon ") consisting of ser- pentine at one end and peridote with a strong oblique cleavage at the other. Fig. 5, PL III., exhibits peridote forming a portion of another " chamber-cast ;' and intersected by layers of calcite. The interreticulation of peridote and serpentine in the first of the above cases is strongly suggestive of its being due to segre- gative chemical action ; but in the two latter it would seem that the peridote is pseudomorphic after some mineral, partly retaining its crystalline form. The intersecting layers of calcite in the specimen under figure 5 may be accepted as strong presumptive evidence of chemical changes involving the pseudo- morphic development of this carbonate. The opinion is pretty general that peridote is in all cases an igneous product. A different view may be entertained by S terry Hunt, who, considering the doctrine of chemical precipitation which he advocates with respect to the origin of the Archsean hemithrenes, ought to regard peridote, at least that form of it which is known to occur in these rocks, as having been formed in the humid or wet way ; for it has come under his notice both in their bedded and " vein "-masses*. It has seemingly never struck any one that the frequent intermixture of peridote and serpentine (the latter in bedded and dyke-shaped masses) is strong evidence at least of its being due to chemical alteration. Breithaupt, who had knowledge of masses of serpentine containing peridote, was evidently influ- enced by the prevailing idea, already noticed, in making the suggestion that they are altered examples of the latter mineral —that is, of a rock related to what is now called a peridolyte. But as to a rock of this kind being a methylosed dolerite, diorite, or any thing else, the idea to Breithaupt was apparently, as it is to several others of the present day, entirely out of the question. The occurrence of peridote in talc-schists seems to have perplexed Bischof ; for he, too, advocated its igneous origin : but being a thorough hydrothermalist as regards the agencies which developed rock -metamorph ism, he was compelled to admit that the case "would seem to indicate that this mineral has * Chemical and Geological Essays, pp. 31 and 210. ORIGIN OF MINERALS CHARACTERISTIC OF OPHITES ETC. 65 been formed in the wet way"*. Bischof, we are strongly inclined to think, would have felt no hesitation in taking this view had he been acquainted with peridotiferous ophites that were sediments originally. Supported by numerous evidences which have come to light of late years through the study of ophites, we must declare ourselves on the side of the opinion that the presence of peridote in all such rocks is as much the result of hydrothermal and che- mical changes as is the serpentine with which it maybe associated. Besides, it is well known that lava contains water and steam for years after its eruption : jets of vapour are still given off from the lava of Jurolla, now nearly ninety years after its ejection. The occurrence of peridote in granites, though countenancing the idea of this mineral having been formed by igneous action and in the dry way, cannot be accepted as proving any thing of the kind ; for it is now well known that these rocks occasionally contain zeolites, serpentine^ chlorite, and other hydrous minerals, which must have been formed in the wet way. Hence hydro- thermal conditions, required by our theory for the production of secondary minerals, have not always been absent in granites. Not long since we pointed out instances of peridote in the amorphous or colloidal form, where it is interreticulated with serpentine. It may also be mentioned that the former mineral occurs in the crystalloid and crystalline conditions, as at Snarum. Beautiful crystals have been found in the lavas of Vesuvius, in forms proper to itself, and not pseudomorphic after those belonging to another mineral : this may be regarded as proving that peridote is not of secondary origin. Any argument of the kind, however, is invalidated by the fact that other minerals are in the same predicament. Epidote, calcite, and talc occur crystalline, each as a secondary product and in forms proper to itself, also as pseudomorphic after garnet &c. Again, the occurrence of crystals and crystalloids of peridote in true igneous dykes and beds of lava is held by some to prove that this mineral and the ejections containing it are of contem- poraneous origin, But a formidable difficulty meets this view at the outset. As generally considered, peridote is one of the most refractory bodies known ; and in agreement with this fact is the well- attested statement that peridote frequently occurs under circumstances showing that the heat of its matrix, when * Chemical and Physical Geology, vol. ii. p. 358. F 66 ROCK-METAMORPHISM. in a molten condition, was insufficient to fuse it. The opinion is therefore pretty general that the peridote in these cases has been thrown out of volcanoes in a solid condition (crystalline and amorphous), imbedded in the lava; in other terms, as ex- pressed by Bischof, " all these facts are indications of its exist- ence in a solid state in the melted lava, and are quite inconsis- tent with the opinion that it has originated from the lava/'' Moreover " the occurrence of peridote in a metamorphic rock renders it intelligible that lava penetrating through beds in such rocks might carry up with it lumps or crystals of this mineral " *. But we cannot admit that its presence in lava is to be explained in this way. The fact that it occurs abundantly in the old lavas of Vesuvius and rarely in those of historical eruptions may be held as supporting a contrary interpretation, in cases of this kind at least; for, admitting that this mineral has been derived from previously existing rock-masses lying at great depths, there is no reason against the subterranean sources yielding it at the present day. Before closing this portion of our present argument we shall cite a passage from Bischof giving his view as to the origin of other minerals (augite, hornblende, leucite, &c.) common in dolerites and lavas, though it may not be unnecessary to remind the reader beforehand that, in agreement with passages already cited, our author must have inconsistently excluded peridote. "It is evident that, after the solidification and cooling of lava, crystals can no longer be formed* by fusion. Therefore, when we find that the older lava contains crystals which either do not occur at all in the more recent lava, or which are at least much larger and better developed than in the latter, it is certain that these crystals have been formed in some other way than by fusion, and there is no other way than the wet way in which they can have been formed : . . . . there is no reason why this mode of for- mation should be considered impossible in the older lava, which has for long periods been exposed to the action of water. " .... "We must therefore refrain from regarding crystalline minerals which occur in volcanic masses as products of fusion " f. In view of the various points thus far brought forward, it seems most unreasonable to exclude peridote from the list of products of alteration. Therefore, until better evidences than have hitherto * Chemical and Physical Geology, vol. ii. p. 358. t Ibid. vol. ii. pp. 94, 95. ORIGIN OF MINERALS CHARACTERISTIC OF OPHITES ETC. 67 been adduced to the contrary are forthcoming, we shall continue to maintain that peridbte, whatever kind of rock may be its matrix, or whether occurring in crystalline forms peculiar to itself or as amorphous masses, is as much a secondary product as other secondary minerals which may be its accompaniments. Weighing the evidences and considerations hitherto adduced, it may be taken as clear that, although genetically a secondary mineral, peridote is of independent origin in the crystalline and crystalloidal examples last under notice. Our next object is to show that peridote, or a mineral sub- stance assumed to be the same, has originated through pseudo- morphism after other minerals, and that therefore in cases of this kind it cannot be of independent origin. That peridote occurs frequently changed into other minerals, as serpentine, is a well-established fact; but the proposition just laid down does not seem to have been entertained by mineralogists : it is one, however, which cannot be said to be gratuitous ; for other minerals are known to be similarly poly- genetic. As hexagonal crystals, quartz is in a form proper to itself and of independent origin ; but as the same mineral sub- stance is found in forms peculiar to calcite, such forms are pseudomorphs, and cannot have originated independently. Our illustration has its parallel in peridote. Again, as we purpose to show that peridote is pseudomor- phous after hornblende and augite, we may be allowed before- hand to ask the question — As serpentine frequently occurs pseudomorphosed after hornblende and augite, why cannot the latter species be pseudomorphosed into peridote, which is so closely related to . serpentine ? We shall endeavour to answer the question in the affirmative by the consideration of evi- dences pertaining to the crystalline form, cleavage, and polarization respectively characteristic of the three minerals concerned. Peridote belongs to the trimetric crystalline system ; and it necessarily has for its primary a right rectangular prism, which in this particular mineral rarely occurs otherwise than under somewhat complex prismatic and basal modifications. Its simplest prismatic modifications, by removal of the lateral edges to the complete effacement of all the primary faces, would be rhombic, with acute and obtuse angles respectively 68 ROCK-METAMORPHISM. 86° and 94°; but it seldom, if ever, occurs in this form. In the annexed woodcut, fig. 1, the outside dotted lines represent a cross section of the primary rectangular prism of peridote ; the thick continuous lines within represent the secondary rhombic modification ; the broken lines represent the cleavage, Usually the prism has eight, ten, or twelve lateral faces ; a cross section, being thus many-sided, has a somewhat rounded or rather ellip- tical outline. Cross sections of peridote are figured in ZirkeVs work with six sides'*. The cleavage of peridote consists of three dissimilar sets, two being parallel with the prismatic faces of the rectangular pri- mary, and one conformable with the basal faces ; consequently their mutual intersections are at right angles to one another (see inner broken lines in fig. 1), as may be observed in the cleavage-solids which occasionally occur in masses of this mineral common at Unkel and elsewhere. Turning to augite and hornblende, both belong to the mono- clinic crystalline system ; their respective crystals, which often occur as six- or eight-sided prisms, are usually less modified than those of peridote; and moreover, on account of their inclination, they are further differentiated from crystals of the last mineral. The simplest modification of the primary of hornblende, when it obliterates, as stated of peridote, all the original faces, gives rise to an oblique rhombic prism, whose acute and obtuse angles are respectively 55° 30' and 124° 30' (see fig. 3) . The corresponding modification of augite yields angles 87° 5' and 92° 55', as in fig. 2. It will thus be seen that the cross section of a rhombic prism of augite differs extremely little in its angular measurements from a similar section of peridote, and that both are nearly a square. The decidedly rhombic form of the cross section of hornblende need not be mentioned in this comparison. The basal modifications in each of these three minerals render it difficult to determine the form of their respective longitudinal sections, depending, as the question does, as to whether such section be a true one or not, on its parallelism with the vertical crystalline axis. * Die mikroskopische Berschaffenheit der Mineralien und Gesteine,pp. 99, 216. None of Zirkel's figures represents a true cross section of peridote, as they are too rhombic j they must be somewhat oblique to the vertical axis instead of at a right angle. Bock-3fe(amorphis>n.] K. 1. [To face p. G8. Pig. 2. Fig. 3. Fig. 1. Diagrammatic transverse sections of a prismatic crystal of peridote. Obtuse angles 94° j acute angles 86°. The outer dotted lines represent the form of its primary j the inner continuous lines angular measurements of its simplest modification ; the broken lines its rectangular cleavage. The prism is upright. Fig. 2. Diagrammatic transverse sections of a prismatic crystal of augite. Obtuse angles 92° 55' ; acute angles 87° 5'. Fig. 3. Diagrammatic transverse sections of a prismatic crystal of hornblende. Obtuse angles 124° 30'; acute angles 55° 30'. The augite and hornblende prisms (Figs. 2 & 3) are inclined (mono- clinic). Their inclination to be assumed as from bottom to top of the page. When their angles are truncated down to the white lines, the resulting section is that of the usual prism. The outer dotted lines represent the form of the primary ; the con- tinuous lines a secondary form, or that of the cleavage-solid j the broken lines the cleavage, which is rhombic. ORIGIN OF MINERALS CHARACTERISTIC OF OPHITES ETC. 69 The cleavage of hornblende and augite is prismatic and basal : the two prismatic sets are parallel with the faces of the secondary rhombic prism ; while the basal set, which is parallel with the basal faces of the inclined primaries, necessarily intersects the prismatic cleavage-planes obliquely *. Obviously, whatever difficulty attaches to crystalline form, cleavage is of great assistance in enabling the investigator to distinguish peridote from either augite or hornblende ; for while there are only right-angled sets of cleavage in the first, there are nothing but oblique sets in the two last. Still there are difficulties to be encountered in connexion with this point, depending on whether or not the section which may be under examination is in its correct, or approximately correct, position of intersection relatively to the vertical axis of the prism : but their consideration may be advantageously passed over as in- volving niceties of calculation unnecessary in this work; and we feel ourselves bound to argue out our conclusions suitably for geological students in general. Let us next examine the sections of crystals of " peridote/' as represented by two of our fellow labourers, to see how far they are in conformity with the cleavage-characters of this mineral. Unfortunately the figures of " peridote " given by Allport " f, although representing each a cross section, do not exhibit any cleavage. Professor HulPs figures are not in this predica- ment J. Allport, who has carefully examined a number of British dolerites, has represented in his figures 25 and 27 cross sections of what he takes to be peridote ; but, from their form, it may be equally assumed that they represent augite. He has also given another section, four-sided, in fig. 26; but exception may be taken to it as being too simple in form for peridote, and having its angles approaching too closely to those characteristic of the acute rhombic modification peculiar to hornblende in its simplest secondary form, and as equally represented by its clea- * The basal cleavage of these minerals, it will be understood, could not be represented in figures of cross sections. t Quart. Journ. Geol. Soc. vol. xxx. pi. 34. figs. 25-27. \ Trans. Royal Irish Academy, vol. xxv. pi. xi. figs. 25-30, u Report on the Chemical, Mineralogical, and Microscopic Characters of the Lavas of Vesuvius from 1631 to 1868," by the Rev, Prof. Haughton and Prof. Hull, 70 ROCK-METAMORPHISM. vage-solid : compare Allport's figure with our fig. 2. We there- fore cannot accept this case, any more than the others, as an example of peridote, unless in the condition of a pseudomorph. We have been obligingly favoured by Mr. JohnjYoung, of the Hunterian Museum of Glasgow, with a numberV speci- mens of trap rocks from the Clyde district— the ^same rocks which supplied Mr. Allport with several of his specimens. From the specimens which Mr. Young sent us we have had prepared several microscopic sections ; but in none do we observe any thing opposing the above conclusion. Many of the crystals contained in them (showing peridotic colours) exhibit irregular rhomboidal sections, which, we consider, prove that the longi- tudinal or vertical axis of the sections is inclined ; moreover several of them exhibit prismatic rhombic cleavage. The mineral we must therefore consider to have been originally augite or hornblende *. In our present investigations we have been materially assisted by Professor Hull, who, though aware that we were disposed to take a view different from his, has kindly allowed us to exa- mine the sections represented in the " Report " with which his name is connected. In the figures to which allusion is now made, attention has been paid not only to angular measurements, but to cleavage-peculiarities. We have not been able to form any satisfactory opinion on any of the sections referred to (being unsymmetrical forms), except the one under fig. 20 ; fortunately this is sufficiently decisive for our purpose. The section given in fig. 7, Plate I., is from a drawing made by ourselves, which, it will be seen, closely agrees with Dr. Hull's representation, except that the cleavage is shown a little more in detail. It is stated that this is a " section of olivine crystal from the lava of 1794." The mineral or chemical nature of the section we do not dispute ; for it exhibits under polarized light the colours characteristic of peridote, excepting that they are slightly * A crystal of augite described by Sandberger lias, disseminated through its entire mass, a substance which he considered to be peridote (Bischof, op. cit. vol. ii. p. 307). We should like to draw Mr. Allport's attention to this case. Moreover the presence of calcite associated with serpentine in Mr. Allport's examples is a noteworthy evidence in favour of our view as to the origin of the " calcareous skeleton" and the calcitic interpolations in the "nummuline wall" of " Eowon Canademe" ORIGIN OF MINERALS CHARACTERISTIC OF OPHITES ETC, 71 duller, and their variation is somewhat different *. These differ- ences, however, are no more than what may be observed in most minerals assumed to be peridote. It may also be mentioned that, considering the approximate agreement between peridote and augite (woocut fig. 2) in their cross sections, Dr. Hull was justified in assuming the crystal to be peridote : for the same reason we see no impropriety in taking it to be augite ; but the sectional difference they exhibit is not to be overlooked. Still, although so much can be said in favour of the identifica- tion made by Hull, it is extremely doubtful to us that it is a correct one, inasmuch as the prismatic cleavage, so well displayed in the crystal under notice, instead of being right-angled in its intersections or parallel with the sides of a rectangular prism, as required for peridote, is rhombic, precisely like that of augite. We are therefore inclined to the conviction that this example was originally a crystal of augite, which has undergone a chemical change into a substance resembling that of peridote, or, as we prefer to say, which has become peridotized. The presence of peridote in certain rocks may be satisfactorily explained in accordance with the views that have been advanced ; for, admitting that crystals of hornblende and augite can be pseudomorphosed into peridote, there is no reason why a crys- talline rock, whether xerothermal, methylosed, or volcanic, which contains these minerals, may not become more or less peridotized. It is in this light that we regard dunyte, picryte, Iherzolyte, ossipyte, olivenyte, and others of their class, also many of the doleritic dykes described by Allport f. We are therefore opposed to accepting the presence of peridote in any one of the above-named examples as evidence that a rock of the kind is in its original condition, but rather as proving that it is more or less a methylotic product. * When the prisms are parallel the colour is pale yellow, at 45° pale purple, at 90° (cross prisms) deep purple, at 135° pale pink, returning to pale yellow: there is no brilliant sap-green, nor ruby. The structure is granular. t The remarkable doleritic rock at Carmoney Hill, co. Antrim (the matrix of the new mineral, hullite, Hardman), is so charged with peridote, as first made known by Prof. Hull (Proc. Royal Irish Acad. 2nd ser. vol. iii. (Science), pp. 166, 167), that we are disposed to consider it another ex- ample of the kind. 72 ROCK-METAMORPHISM. CHAPTER XII, ON THE ORIGIN OF THE ARCHAEAN " CRYSTALLINE LIMESTONES " OF CANADA. THE Canadian Archseans are for the most part mineralized metamorphics, the methylosed members (usually silo-carbacid) being a subordinate group. In general highly crystalline, the former consist of bedded masses of granitoid gneisses,, quartzites, diorites, dolerites, various crystalline schists, &c., and the latter of calci-hornblendic gneisses, ophites, " crystalline limestones," and other related kinds. The " crystalline limestones " having been proved to be chemically, mineralogically, and structurally identical with hemithrenes, we shall assume them to be rocks of the latter kind. The " crystalline limestones " or hemithrenes, which are often interstratified with the mineralized metamorphics, vary much in their mineral composition : calcite, or miemite (either of which is usually present) generally serves as a matrix for the mineral silicates — chondrodite, augite, hornblende, phlogo- pite, orthoclase, labradorite, serpentine, quartz, idocrase, &c.; which, with apatite, graphite, and other non-silicates, occur as crystals, or as irregularly-shaped grains (crystalloids) varying much in size, " either alone or variously associated, and some- times in such quantities as to make up a large proportion of the rock, to the exclusion of carbonate of lime •/' so that beds are often seen to graduate completely into diorites, gneisses, and other silacid metamorphics *. The beds are often greatly contorted ; and their component layers are frequently and independently crumpled in the most extraordinary manner. At the Ragged Chute on the Mada- waska (Canada) there is a bed, according to Logan, " three feet * Report of the Geological Survey of Canada for 1866, p. 185; and Chemical and Geological Essays, p. 206, 73 thick, which consists of alternating layers of limestone and gneiss singularly corrugated, lying between masses of evenly laminated hornblendic gneiss"*. The latter, from its structure, may be regarded as due to sedimentation • but the corrugation of the enclosed " layers of limestone and gneiss " is totally inconsistent with its being of similar origin. Obviously the phenomenon is an example of segregated lamination, whose corrugation is the effect of internal movements, in individual beds, produced by chemical forces, The latest estimate of the thickness of the entire group of Archaean rocks has been made by Mr. Henry G . Vennor, who, from very careful and laborious observations carried on during a number of years, is entitled to the highest confidence, and " from whose views " Mr. Alfred R. C. Selwyn, Director of the Canadian Survey, says " he has no reason to dissent." " It would appear that the whole volume is not less than between 50,000 and 60,000 feet ; and this estimate does not include the great fundamental unstratified or obscurely stratified gneisso- syenitic series, but commences only with the first strata of clearly stratified gneiss." As to the great fundamental series, it is at present impossible to suggest even its approximate thick- ness ; but, forming the backbone of Eastern Ontario, as well as thousands of square miles in the region to the northward of the Ottawa river, it ' ' is apparently a distinct formation, though the separation between it and the first strata of the overlying stra- tified gneiss is not always clear "f- It is well known that the late Sir William Logan designated the great metamorphic group under consideration by the name Laurentian, also adding thereto, with a distinctive title, another metamorphic group, on the whole mineralogically different, which he regarded as an unconformably overlying one. But the inves- tigations of Vennor have thrown some doubt on the alleged unconformity ; the latter geologist is nevertheless disposed to * Geology of Canada, 1863, p. 27. Dr. Bigsby, speaking of the bands of Canadian crystalline limestone, states " they are tortuous, and often, by bend- ing round, sharply return by a parallel course to within a short distance from their visible point of departure. Corrugated seams of gneiss are sometimes enclosed in the limestones" (Geological Magazine, vol. i. p. 156). t Report of the Geological Survey of Canada for 1876 and 1877, pp. 280, 299; 300. 74 ROCK-METAMORPHISM. adopt a twofold division, but to make the separation on a dif- ferent horizon to that adopted by Sir William. If we have not misunderstood the proposed subdivisions, the lowest one, which includes most of the Laurentian system of Logan, is essentially composed of silacid rocks, only a thin zone of crystalline limestone being included in it ; the next one embraces, in addition to the silacid members, a massive zone of hemithrenes (" crystalline limestones ") , in which are included stratified dolerites, ophites, and other related rocks. Accord- ing to Vennor, who had favourable opportunities for taking measurements, the zone has " an average thickness of 5600 feet"* Mr. Selwyn proposes to make a system for the last group of rocks, retaining for it Logan's name, Huronian f, but to add thereto some other groups (certain members being slightly altered). Respecting the correctness of this proposal there may be some disagreement among American geologists. It has been remarked by Sir William E. Logan that "even during the Laurentian period the same chemical and mecha- nical processes which have ever since been at work disintegrating and reconstructing the earth's crust were in operation as now. In the conglomerates of the Huronian series there are enclosed boulders derived from the Laurentian, which seem to show that the parent rock was altered to its present crystalline condition before the deposit of the newer formation, while interstratified with the Laurentian limestones there are beds' of conglomerate, the pebbles of which are themselves rolled fragments of still older laminated sand- rock; and the formation of these beds leads us still further into the past" {. Cordially agreeing with these remarks, we are prevented ac- cepting Sterry Hunt's belief that in " pre-Cambrian times there prevailed chemical activities dependent upon greater subter- ranean temperature, different atmospheric conditions, and abun- dance of thermal water, and that under these circumstances * Ibid. p. 264. This is Mr. Vennor's estimate for the " Hastings limestone zone." His estimate for what is considered to be a corresponding zone in Lanark township is not " less than from 5600 to 6000 feet thick " (Report, 1874-75, p. 143). . t Canadian Naturalist, vol. ix. p. 30. J Quart, Journ. Geol. Soc, vol. xxi. pp. 46, 47. " CRYSTALLINE LIMESTONES " OF CANADA. 75 were deposited the materials of the great crystalline rocks "*, — that is, as chemical precipitates. There are no more reasons for ascribing the 60,000 feet of Laurentian sands, argills, and other deposits (that is, in their original condition as sediments) to Dr. Hunt's " chemical acti- vities " than there are for attributing the 30,000 feet of similar deposits forming the Longmynds to the same agencies. At the Dublin meeting of the British Association, 1878, Dr. vSorby referred to the presence of certain crystalline substances, in the "red clay" and other deposits, which the ' Challenger5 expedition brought up from the bottom of the Pacific and Atlantic Oceans, in such a way as to make it appear that he is ngt opposed to the notion that they are the products of chemical precipitation. The nature of these crystalline substances and the deposits enclosing them is engaging the attention of Renard and Murray. The full result of their investigations has not yet been published ; but some of their conclusions are stated in the British-Association Report of the Sheffield Meeting in 1879, pp. 340, 341. There is nothing in their statements to coun- tenance the idea that any thing in or belonging to the red clay, or the deposit itself, is a chemical precipitate from ocean water, but on the contrary, that it is for the most part a volcanic ash, which through decomposition or disintegration has become con- verted into red clay. The crystalline substances it contains, viz. plagioclase, augite, peridot e, sanidine, magnetite, zeolites, sideromelane, &c., are more probably non-disintegrated portions of mineral aggregations contained in the volcanic ash, than the result of chemical reactions. Prof. A. Geikie (seemingly referring to some of the above), in stating that " these silicates (there may be several of them) have certainly been formed directly on the sea-bottom/ 'f commits himself to what we expect will turn out to be an erroneous conclusion. Dismissing the silacid metamorphics for the present, Sterry Hunt's theory of the origin of the Canadian Archaeans in their totality necessarily makes the interbedded " crystalline lime- stones " a " chemical deposit ; and there is no doubt that a part of these limestones, like those of more recent formations, have been directly precipitated by chemical reactions from the waters * Nature, August 22, 1878, p. 444. t Article " Geology," in ' Encyclopaedia Britannica/ vol. x. p. 288. 76 BOCK-METAMORPHISM. of the ocean. " to " react! ons, which are still going on in the ocean's waters, and which have, in past times, given rise directly to limestone strata, in which the occurrence of shells and corals is only accidental" *. That the latter deposits are in the main of organic origin is doubtless the opinion of most geologists ; but a difficulty in connexion with this point has arisen as regards the Archaean hemithrenes. To those who believe in "Eozoon," with the exception of Sterry Hunt, the organic origin of the Canadian hemithrenes, it may be assumed, is a settled matter. But it must be admitted, even by those to whom we refer, that there are others who totally dissent from this belief, and who, besides, reject the theory of chemical precipitation. Obviously, then, the onus lies with them to suggest or propose a different solution of the problem. It would be rash on our part to deny that organisms of any kind were in existence during the Archaean periods ; but our researches having afforded no proper evidence in support of the affirmative, we are induced to make the attempt to solve the question, as to the origin of the Canadian hemithrenes, other- wise than by the doctrine either of organic intervention, or chemical precipitation. Reverting for a moment to the Laurentian silacid rocks, our view as to their origin is based on a full recognition of existing operations pertaining to physical geography : it involves the intervention of mechanical and chemical agencies in effecting the disintegration of rocks already in existen.ee ; also the removal and dispersion of the materials derived therefrom, and their consequent deposition in other and distant areas. It will therefore not be unreasonable to regard the Ontarion funda- * ' Geology of Canada/ 1866, pp. 200, 201. The author further declares "the often repeated assertion that organic life has built up all the great limestone formations, is based upon a fallacy ; for animals have no power to generate carbonate of lime." It may therefore be presumed that mollusks, Crustacea, &c. have no power to convert the sulphate of lime, or chloride of calcium (the principal lime constituents of the ocean), into the carbonate of lime of their shells. But as we hold the contrary opinion, we may be allowed to maintain that the shells and corals contained in the limestone are the essential contributors to its formation, and consequently, that " organic life has built up all the great limestone formations." "CRYSTALLINE LIMESTONES OF CANADA. 77 mental " unstratified gneisso-syenitic series " as the rocks which yielded the silicates of alumina, magnesia, and lime, the alka- line silicates, and the oxides of iron, that gave rise to the sands, muds, and agglomerates of the overlying Laurentians. While the assumed mechanical actions were going on the carbonic acid of the water and atmosphere would doubtless act on the calcic, magnesian, and alkaline silicates ; next would necessarily follow the conversion of the latter into soluble carbonates. Besides other resultants, bicarbonate of lime would be gene- rated, and forthwith conveyed by rivers into lakes and seas. Having never denied the existence of plants or animals during the Archaean periods, we see no reason why some of the lime salt, as above generated, may not have contributed to the for- mation of organic skeletons ; but this yet remains to be proved. We must also admit the probability that in shallow seas and lagoons, also along shores, carbonate of lime would be thrown clown by the evaporation of their water ; and it may be equally admitted that the water under certain conditions would other- wise part with it. In this way may have been formed deposits of limestone, but comparatively insignificant in quantity, while carbonate of lime may have become intermixed with the com- ponents of other shallow-water deposits. The calcareous base of the conglomerates near the Coulonge river, Co. Ottawa, strikes us as being probably a littoral formation. But with respect to the 6000 feet of " crystalline limestones " — the silo-carbacid members of the Laurentians— we must hold to a different explanation of their origin. Without availing ourselves of heat emanating from contiguous masses of molten rocks, we shall simply assume that before the Laurentian deposits were thrown out of their original position, a considerable portion (being buried at great depths and under enormous pressure, and retaining their original water) became affected by the elevated temperatures prevailing at such depths; and thus they passed into the mineralized condition character- istic of ordinary rnetamorphic rocks. Afterwards, through the introduction of extraneous water containing a carbacid solution, these mineralized metamorphics became chemically changed into hernithrenes and other related rocks. In support of this view we might appeal to evidences, already described, of the most conclusive kind — those yielded by the 78 ROCK-METAMORPHISM. Mont St. Philippe and the Glassillaun dyke at Cleggan ; but we have others at hand. It has already been stated that S terry Hunt has shown that the Laurentian rocks of Canada are intersected by two classes of veins, distinguished as " granitic " and " calcareous/'' " Most of their characteristic minerals are common to the two classes ; and it is easy to trace a gradual change from the typical granitic veins to those in which carbonate of lime is the predominant mineral " * ; hence " it is impossible to draw any definite line between " them. The fact is, the one passes by " insensible degrees" into the other through the gradual decrease of mineral silicates and the increase of carbonate of lime. The same authority has also shown that there are certain Lau- rentian beds, "generally granitoid or gneissoid in structure, which may be described as pyroxenites, from the prevailing mineral/1 pyroxene (augite) : this mineral is te sometimes nearly pure, and at other times mingled with mica, or with quartz, and ortho- clase, and often associated with hornblende, epidote, &c." The beds are occasionally of great thickness, and are " then often interstratified with beds of granitoid orthoclase gneiss, into which the quartzo-feldspathic pyroxenites pass, by a gradual disappearance of the pyroxene." " These peculiar strata, which contain at the same time the minerals of the associated gneiss and of the limestones/'' also " gradually pass, by an admixture of carbonate of lime, into the adjacent crystalline limestones" t —those whose origin is under consideration : it is noteworthy that, besides serpentine, they, too, contain the mineral species of the granitoid gneiss and the pyroxenites. Thus the opposite extremes of the Lauren tian series of rocks, from the essen- tially silacid to the highly calcareous members, are linked together by intermediate gradations. Another remarkable point remains to be noticed. It is stated by Sterry Hunt that ' ' in their mineral character the calcareous veins so closely resemble the stratified limestone that the dif* ferent geographical relations of the two alone enable us in some examples to distinguish between them," and that " the observer will often find it difficult to determine whether a detached mass; or an imperfectly displayed outcrop of crystalline limestone^ belongs to a bed or a vein" J. * Geology of Canada, 18G6 (Dr. Sterry Hunt's Report, p. 191> t Ibid. p. 185. t Ibid. p. 188. 79 In a former Chapter mention was made of many of the mine- rals, both in the limestone beds and the calcareous veins, having more or less rounded or corroded surfaces. It would appear that this phenomenon is more common in the beds than in the veins. Kef erring to the corroded minerals belonging to the veins, Sterry Hunt avers that " this rounding of the angles of certain crystals appears to me to be nothing more than a result of the solvent action of heated watery solutions/' But he holds a different view respecting these bodies in the crystalline lime- stones : they " sometimes occur in small distinct crystals, but more generally in rounded irregular grains, which present a marked contrast to the same minerals occurring in the veins. This rounded form of the minerals in the beds of limestone, is to be carefully distinguished from the rounding of the crystals in the veins. In the latter case the rounding is by no means constant, and is confined to a few species, while in the limestone beds it will be found that a rounded form characterizes alike apatite, and quartz, and such silicates as pyroxene (augite), hornblende, serpentine, and chondrodite " *. In accordance with his view, maintaining the organic origin of " Eozoon Canadense" he declares that the " rounded form 3> which cha- racterizes such silicates, occurring in the beds, has been ( ' demon* strated, in many cases at least, to be due to no such subsequent action, but has been given by the calcareous organic structure, in whose chambers these silicates were originally deposited "f. It will not be expected that we can accept this so-called de- monstration after having shown, with" other evidences to which the merest allusion is only necessary at the present moment, that the " chamber-casts " of c< Eozoon Canadense" when they are not bordered by the " nummuline wall/' are usually coated with flocculite, resulting from the disintegration and waste of their component serpentine, just as the same part in its typical state is the product of corresponding changes in chrysotile — and more especially since "we now know," as stated by Sterry Hunt, " that water, aided in some cases by heat, pressure, and the presence of certain widely distributed substances, such as carbonic acid, alkaline carbonates, and sulphides, will dissolve the most insoluble bodies/' And con- sidering that watery solutions of the kind are ever present, * Geology of Canada, 1866, p. 190. t Ibid. p. 191. 80 ROCK-METAMORPH1SM. and all-pervading, in deep-seated rocks, as repeatedly contended for by the author, the fact may be taken as positive evidence that the ' ' granitic veins 3> and the peculiar pyroxenic beds were alike penetrated by such solutions, whose solvent powers were exerted on the mineral silicates, whether belonging to the beds or the veins. As to the " marked contrast " which it is endeavoured to make out between the "rounded irregular grains " of the former and the " rounded crystals " of the latter, it seems extremely doubtful that the difference is more than one of degree : it is not, however, beyond an explanation on our view. As the veins are clearly posterior to the beds, the latter have an important factor in their favour: time would enable the corroding agents to act more decidedly and more generally on the " grains " in the beds than on the " crystals " in the veins. Moreover the "contrast" may also be due to a difference in solvent power between the penetrating water in the two cases. Again, the action of solvents, by rounding the " grains " or " crystals," involves removal and displacement, whether it takes place in beds or veins ; and as there are no openings surround- ing either the grains or "crystals," but on the contrary an environment of calcite, obviously the latter is the replacement substance. And it may be equally affirmed that all the calcite which occupies the interspaces between the grains and crystals is in the same predicament, as in certain pseudomorphs whose mineral silicates have been replaced by a mineral carbonate. Applying this reasoning to the solution of the problem in regard to the origin of the Canadian bedded " crystalline lime- stones" or hemithrenes, it will be understood that we have simply to enlarge the field of solvent action. We may take a region occupied by contorted granitoid, labradoritic, horn* blendic, and other gneisses, permeated by thermal water charged with a carbacid solution — the water having gained admission into the rocks either directly or indirectly from an overlying ocean, along zones of outcrops, jointing, or porous beds, the direction of such zones corresponding to the strike of the rocks : by this means the gneisses would become regionally converted into hemithrenes, the quartzo-feldspathic diorites, with their admixtures of calcite and serpentine, being "beds of passage between the two rocks." " CRYSTALLINE LIMESTONES " OF CANADA. 81 We may now summarize the evidences and considerations which have been brought forward by way of substantiating our view as to the origin of the Archaean " crystalline limestones " of Canada — that they were silacid metamorphics which have become chemically and otherwise changed or methylosed into calcitic masses (hemithrenes) , it having been shown : — That the calcite of the calcareous structures of " Eozoon " in the ophite of the Canadian Laurentians is a replacement of serpentine and other mineral silicates ; That the mineral silicates, augite, hornblende, &c. (character- istic of hemithrenes), are often pseudomorphosed into calcite ; That only under extremely limited conditions are chemically precipitated limestones formed ; That no reliable evidences have yet been adduced proving the existence, during the Archaean periods, of lime-elaborating organisms (the only ether kind of agency admissible on our part), through whose intervention the "crystalline lime^ stones " could have been formed ; That the peculiar convoluted or tortuous lamination which distinguishes many of the " crystalline limestones " is only explicable, in the absence of cases to the contrary, on the idea of its being a superinduced phenomenon ; That the " crystalline limestones," also the associated ophites, are most abundant where they are interbedded with silacid rocks (" pyroxenytes ") , whose essential minerals contain a large percentage of calcium (and magnesium) silicate ; That the gneissose rocks of St. -Philippe (Vosges) and of Glas- sillaun (Cleggan Bay) have been converted into hemithrenes by chemical action ; finally, That the great Archaean limestone formation of Canada consists of silacid gneisses which " pass insensibly " into hemithrenes through decrease of mineral silicates and increase of carbonate of lime, — that this formation is inter- sected by "granitic" and (e calcareous veins," between which " it is easy to trace a gradual change," — that the " pyroxenytes " and beds of limestone, also the " granitic " and " calcareous veins," include in a general sense identical crystalline siliceous minerals, with their surfaces more or less corroded by the dissolving action of heated water con- 82 BOCK-METAMORPHISM. taining carbacid solutions, ever present in deeply-seated rocks, — and that, under the influences stated, the " granitic veins " and the bedded " pyroxenytes " have gradually undergone changes, which eventually converted them re- spectively into " calcareous veins " and beds of " crystalline limestone." If any extensive beds of really pure limestone, resembling carrarite, are included in the great Archsean class of rocks, we should, rather than ascribe them to simple mineralization as we do the marble named, prefer the suggestion that they have been deprived of mineral silicates through the latter having been washed out by dissolving waters. HARITY OF POST-ARCHAEAN LIMESTONES. 83 CHAPTER XIII. WHY ARE LIMESTONES COMPARATIVELY RARE IN THE FORMATIONS IMMEDIATELY SUCCEEDING THE ARCH^EANS ? IT has long been a matter of surprise that the Cambrians (the great series of rocks from the base of the Longmynds to the top of the Tremadocs), exceeding by far 40,000 feet in thickness, contain very few limestones; and the surprise is heightened when the paucity of the latter formations is compared with the vast masses of calcareous members belonging to the Archseans. We are referring to the Cambrians as they occur in Wales, Scotland, and Ireland, where their calcareous matter, comparatively a mere fractional constituent, is generally in a diffused state, or forming only thin layers. Even by including Hicks's Dimetian series (which, however, he regards as Archaean) , the ' ' limestone beds " it contains at Porthlisky would form no valid exception to our statement, particularly as there are strong grounds for the opinion that they, and the ophite associated with them, do not retain their original composition*. * Dr. Hicks has kindly favoured us with two small specimens taken from the " impure limestone bands " (from 1 to 3 feet in thickness) of Porth- lisky, and belonging to his Dimetian series. This is separated by uncon* formity from his Pebidian series. He is inclined to consider both as Archaean, the Pebidians being overlain unconformably by unaltered Harlech grits. We find the specimens to be principally composed of white augite or malacolite, in short well-cleaved crystalloids confusedly aggregated. After decalcifi- cation they present themselves separated by interstices, cavities, and con- tinuous passages, which before had been filled with calcite. In many instances the crystalloids of malacolite are translucent, and their angles are sharp ; but often they are opaque, rounded, and incrusted with white floccu- lent matter. Eveiy thing observed in connexion with the malacolite and calcite convinces us that the former is in course of replacement by the latter. The calcite is more abundant where the crystalloids of malacolite are eroded 84 ROCK-METAMORPHISM. It is not until the Lower Silurians are reached that limestones are found to occur to any extent ; and these are either very impure, as in the Llandeilo flags, or they form thin beds, e. g. those in the Bala limestone of the Caradocs. The Durnes lime- stone in Sutherlandshire (a partly mineralized rock, older than our Llandeilo flags, and probably the equivalent of the Stiper- stones, the bottom of the Lower Silurians) and the Coniston limestones may be regarded as improvements on their Welsb equivalents. During the Upper Silurian period a marked change took place : limestones were developed on the grandest scale ; and since then the formation of the same class of rocks to the like extent has continued throughout every succeeding period. On the continent of Europe a similar scarcity of limestones characterizes the Cambrians, while a fair increase in their amount marks the Lower Silurians. In North America the Cambrian limestones offer no very marked exception to contemporaneous formations in the British Isles or on the Continent. Of the Acadian and Potsdam groups, which have been bracketed with the Welsh Llongmynds, Harlech Grits, and some other Lower Cambrians, the first contains no recorded non- crystalline limestones*; while the second only comprises some of inconsiderable thickness and occupying but small areas : such are the red sandy dolomites and other calciferous formations of Troy (N. B.), North-western Vermont, the Straits of Belle Isle (Newfoundland) , and a few more places. and covered with flocculite (which substance can only be their residue) than where they are sharp and translucent. The crystalloids of malacolite in a few cases were observed to be so far decreted as to assume rude branching forms. In one instance of this kind the form closely resembled typical imi- tative configurations. Intermixed with the calcite and malacolite we found a pale greenish-yellow granular substance resembling serpentine, bundles of prismatic epidote or actinolite, magnetite in octahedral crystals, a triclinic feldspar showing striping, and galena. This case strikingly resembles the one at Cleggan. If these specimens are characteristic of the limestone beds at Porthlisky, there can be no question about the latter being methylosed products. * For reasons which will be understood after a perusal of the last para- graph of the present Chapter, we confine our remarks on the question under consideration to unaltered limestones. RARITY OF POST-ARCHAEAN LIMESTONES. 85 Iii the Canadian group, presumed to correspond with our Upper Cambrians, we have evidence that a further increase of limestones took place, while the Chazy limestone (considered to be the youngest member of this group) and the Trenton lime- stones of the Lower Silurian system afford ample evidence of an abundant increase of the calcareous element. With reference to the large increase of limestones in the last- named formations, the fact must be taken as showing that in North America the increase took place earlier than in the Euro- pean areas. Cconnected with this is the remarkable fact, already stated, that the earliest Palaeozoic formations (those constituting the two Cambrians) contain few fossils with ordinary calcareous skeletons. Besides some others (fucoids) of no importance to the question at issue, the Cambrians yield the remains of Protozoans, Crustaceans, Coelenterates, Mollusks, and Echinoderms, occa- sionally in abundance, and some of them (Paradoxides) of gigantic proportions ; instead, however, of the skeleton of these organisms being ordinarily thick and composed of lime, as is general in certain of their classes respectively, they are (admit- ting a few apparent exceptions) thin, and for the most part horny, with comparatively a small quantity of phosphate of lime, and a much smaller of the carbonate — a circumstance which may be taken as favouring the idea that the Cambrian stages of organic development were not much beyond that of the larval evolution of the invertebrates whose remains have been noticed*. The absence of calcareous fossils, and the rarity of limestones in the Cambrians, it may therefore be assumed, are correlative phenomena. It would seem that the seas of the very earliest Palaeozoic periods were poorly charged with calcic constituents, and that they were thinly tenanted by lime-elaborating organ- isms. Was the latter consequent on the former? The Cambrian rocks, whether occurring in North America, on the continent of Europe, or in the British Islands, consist of * Mollusks and other invertebrates in their larval condition have horny shells. It seems extremely doubtful that there was much calcic matter of any kind in the Cambrian trilobites. Stony corals have not yet been found in the Cambrians. Archaocyathus Atlanticus (presumed to be a sponge), from the Potsdam of the Straits of Belle Isle, Newfoundland, may be calcareous, Stems of crinoids occur in the Potsdams, 86 ROCK-METAMORPHISM. materials precisely such as would result from the mechanical degradation of the gneisses, syenites, and other silacid members of the Archaeans. The abundance of argillytes, sandstones, and other mechanically produced deposits, the rare accompaniment therewith of limestones, and the contemporaneous presence of non- calcareous fossils admit of no difficult explanation ; but when viewed in connexion with the occurrence of 6000 feet of crystalline limestones in the preceding Archseans, the problem becomes inexplicable. It must strike every geologist that, if such an enormous thick- ness of calcareous rocks was available during the Lower Cambrian period (that is, when these sandstones and argillytes were in course of derivation, through disintegration and denuda- tion, from the Archaean gneisses &c.), we ought to expect that the great lime-bearing series referred to yielded, through organic intervention, a considerable amount of limestones. But where are they ? Certainly not in the diffused quantities, or the " belts" and "bands," that are known, particularly as it is questionable that the lime in these cases is much in excess of the amount which could have been generated by the contemporary action of the carbonic acid, then pervading the atmosphere and different waters, on the calcium silicate in the labradoritic, hornblendic, and augitic debris produced by the disintegration and denudation of the Archseans during the Cambrian periods. With respect to the period when the Archsean crystalline limestones were completely elaborated (for our hypothesis in- volves slow processes requiring immensity of time), we can offer no decided opinion. It would appear improbable that their denudation, operating throughout a vast chronological term that embraced the two Cambrian periods, and producing miles in thickness of debris, would give rise to no more than the small quantity of limestones that were deposited during these periods. On the other hand, however, there seems to be great probability that the crystalline limestones constituted the great factors which so vastly increased the number of organisms with calcareous skeletons, and the consequent calcareous deposits, during the Silurian periods. To account for the paucity of early post- Archaean limestones, we offer the suggestion that, during the Lower Cambrian period there was no great series of crystalline limestones included RARITY OF POST-ARCH^AN LIMESTONES. 87 among the Archaean rocks, that the hemithrenes and ophites constituting this series were only in process of elaboration when the Lower Cambrians were in course of formation. It would be of importance to learn if any specimens of hemi^ threne or ophite occur in the Archaean conglomerate of the Coulonge river and other places. The circumstance would have to be accepted as proving that the methylosis which had de- veloped the crystalline limestones had set in antecedent to the formation of the conglomerate. The explanation involved in the above suggestion embraces all the facts of a problem which is unresolvable by either of the doctrines we are opposed to, as regards the origin of the Archaean hemithrenes and ophites; for if one were true, its advocates would be able to point to something less vulnerable than sparingly-developed limestones of the Cambrians ; or if the other were a fact, its supporters would be able to refer to other than the little better than corneous fossils characteristic of the same rocks. Again, why chemical calcareous precipitates should cease, or why the presumed " Eozoon Canadense " should precede others only furnished with skeletons of a larval type, while it never puts in an appearance subsequent to the Archaean periods, except in the like methylosed and crystalline rocks, as the ophites or hemithrenes of Connemara, Ceylon, Aker, Mt. St.-Philippe, the Isle of Skye, and other places (most of them considered to be Postarchaean, the last-named one being Jurassic), are also questions which require to be satisfactorily answered. In the meantime we must continue to put faith in our sug- gestion that the rarity of limestones in the Lower Cambrian system is due to the absence of preexisting calcareous rocks from which they could be derived — that the great series of Archaean crystalline limestones, which would represent such preexisting rocks, was only imperfectly elaborated during the Lower Cambrian period. This we beg to be accepted as our answer to the question, Why are limestones comparatively rare in the formations immediately succeeding the Archaeans ? The preceding observations, we wish it to be understood, refer to Cambrian calcareous deposits that have undergone neither mineral nor chemical changes : we are thus particular because it is well known there are certain metamorphosed groups which com- prise crystalline limestones and ophites, in Canada, the State of 88 ROCK-METAMORPHISM. New York, and New Brunswick, considered by many geologists to be of post- Archaean age, as the altered Quebecs, Emmons's " pri- mary limestones/' &c. A corresponding formation (the lime- stone of Essex and adjacent counties) is considered by Pro- fessor James Hall to belong to a " period subsequent to the deposition, metamorphism, and disturbance of the rocks of" authentic Laurentian age," and to apparently hold a place in the series between the Laurentian and the " Potsdam periods ; but whether of Huronian age or otherwise " he does " not pre- tend to say ; and it may even prove of later date than this"*. Contending for the existence of true Archseans or Lauren- tians in the highland region of Northern New York, and that these rocks are unconformably overlain by a younger and well-developed series of gneisses, mica- and hornblende-schists (including crystalline limestones and ophites) in Westchester and other adjoining counties, Prof. Dana has of late endeavoured to prove, by their relations to certain fossiliferous deposits occur- ring in neighbouring places, that the younger metamorphics are of Upper Cambrian and Lower Silurian (" Calcif erous," " Quebec," " Chazy," and "Trenton") agesf. Should this prove to be correct, there will be no need of trying to make out that the crystalline limestones they contain were ever other- wise than calcareous ; for as such they may have been simply mineralized. Nevertheless it does not seem improbable that some of the Westchester crystalline limestones may be methy- lotic products, particularly as there is no reason to exclude from the series containing them two or more older groups, the Acadian and Potsdam, whose remarkable deficiency of cal- careous matter in their unaltered condition would render it highly probable, should they in their metamorphosed state contain well- developed crystalline limestones and ophites, that metamorphosing agencies had generated such limestones. * ' American Journal of Science and Arts/ 3rd series, vol. xii. p. 300. t American Journal of Science [and Arts,' 3rd series, vols. xix. and xx., "Geological Relations of the Limestone Belts of Westchester Co., N. Y." The reader is referred, in this connexion, to Selwyn in ' Canadian Naturalist,' new series, vol. ix. pp. 17-31. MINERALIZED CRYSTALLINE LIMESTONES. 89 CHAPTER XIV. SOME CRYSTALLINE LIMESTONES AEE SIMPLY MINERALIZED. IT will be recollected that we have made an exception to cer- tain crystallized calcareous rocks related to hemithrenes being of purely methylotic origin. It seems probable that this excep- tion may apply to the marbles of Dalnein in Strathdon, Glen Elg, Glen Tilt, and the neighbouring localities : still we cannot altogether reject the idea that there are nomethylosed examples amongst these marbles, even should they be of post- Archaean age, just as it may be doubted that all the Westchester crystalline limestones are simply mineralized products. The fact men- tioned by Dana that " green hornblende in minute rounded crystals is occasionally found disseminated through the lime- stone of Westchester/'' and the similar well-known fact of the Scotch marbles being shotted with decreted crystalloids of augite &c., involve the action of corroding or dissolving solutions — the prime agents in methylosis — which marbles, according toHeddle, contain trifling specks or remains of augite, and are of post- Archaean age ; and for this reason doubts may be entertained of their belonging to the group treated of in Chapter X. But we see no improbability of an impure limestone belonging to the Cambrian, Silurian, or later systems *, through mineraliza- tion, changing into a rock which it would be difficult to distin- guish from a hemithrene f. Therefore, until further light has been gained respecting the geological position of the marbles of Glen Elg and other places referred to by Heddle, and generally considered to be Upper Cambrian or Lower Silurian in age, we consider it safest to admit that some of them may have been calcareous in their original condition, especially as the Durness limestones are proved to be early Palaeozoic by their fossils. * Some of the Scandinavian " primary limestones," described by B. Gotta (Zeitsclirift der deutsch. geol. Gesellschaft, 1852, Band iv. pp. 47-53), are undoubtedly of this class ; but we strongly suspect that many others he has noticed (Aker &c.) are methylotic. t Excluding its ophitic portions, much of the Isle-of-Skye " white marble " may be an example of mineralized limestone, 90 EOGK-METAMORPHISM. Many of the bedded crystalline calcitic rocks of Connemara appear to be simply mineralized metamorphics. There are calcareo-siliceous bands, more or less crystalline, commencing at Lough Bonn, a little north of Oughterard, and skirting, apparently continuously, the mail-road on to Ballinahinch, thence across the mountains to Letterfrac*. There are good reasons for believing that they were also originally impure limestones. Still we entertain a suspicion that the " crystalline limestones " which have been noticed, by Mr. R. Glascott Symes, occurring in what appear to be corresponding metamorphic schists in Mayo, will turn out to be methylosed hernithrenes. They are ' ' seldom following the foliation of the beds, but occur along lines of great fault." " From such evidence," according to Mr. Symes, "it may be assumed that this crystallized limestone has been formed by infiltration or percolation of bicarbonate of lime, from the once overlying Carboniferous rocks, into joints or cracks in the now metamorphic series." Prof. Hull, however, expresses his aversion to this view, and considers these limestones " to belong to the group of strata in which they are found, just as similar limestones do in the West Galway district "f. It is highly probable that the pure marbles (carrarite) of the Apuan Alps are simply mineralized. But it rarely happens that the Irish cases to which we have referred are unassociated with methylosed rocks — that is, beds chemically changed into dolo- mites, ophites, and even into true hemithrenes. Moreover the limestones in the north-east of Donegal are undoubtedly no more than mineralized, and probably of Lower Silurian age ; while the calcitic rocks occurring further west, filled with ido- crase and other mineral silicates, and interbedded or intimately associated with granites &c., may be of methylotic origin and represent a much earlier geological period. * We are not sufficiently acquainted with the stratigraphy of the calca- reous marbles and ophites of the Barna-Oran district, east of Ballinahinch, to .offer a decided opinion as to whether the former are methylosed or mine- ralized— though, from their being associated with the latter, the probability seems strong to us that they were originally silacid rocks. t Explan, Mem. of Sheets 41, 53, and 64, Geological Survey of Ireland, p. 12. Mr. Symes mentions other cases of the kind occurring in Mayo — beds of micaceous limestone in mica-schist, in some places traversing the latter at right angles, at others following the direction of its folia. See Expl. Mem, of Sheet 75, p. 12, and Expl, Mem. Sheets 63 and 74, p. 11, METHYLOTIC ORIGIN OF DOLOMITES. 91 CHAPTER XV. DOLOMITES AND DOLOMITIC ROCKS HAVE UNDERGONE METHYLOSIS. DOLOMITES, in our opinion, are in all cases products of methy- losis; but the phenomena have been effected in two different ways — one by substitution of their original basic and acidic compo- nents, as in the dolomitic hemithrenes of the Canadian Archseans, and the other by replacement, more or less, of only their original basic carbonate of lime. Although the rocks we propose to con- sider in the present Chapter are of sedimentary origin, it must be understood that we do not deny the probability of certain igneous and' silacid masses (veins and dykes in the Canadian Archaeans), through methylosis, having been dolomitized. But the rocks we are more directly engaged with are the mag- nesian limestones which, chiefly belonging to various systems of the primary and secondary groups, are spread over extensive areas of Europe and North America, and beneath or associated with which nothiug is present except sedimentary deposits in an unaltered condition. Our typical example is the Permian magnesian limestone of the north of England. Various opinions have been propounded as to the origin of the magnesian carbonate present in this limestone and others analogous to it. There are some, as Apjohn, T. Richardson, Hunt, and Ramsay, who believe that this compound is an original constituent; while others, as Virlet, Scouler, Haid- inger, Von Morlet, Bischof, Johnston, Sorby, and Hardman, make it to be a superadded ingredient : the last opinion is methylotic. But let the different opinions of both classes be examined, and it will be found that all are at variance as to the modus operandi which has developed the magnesian con- stituent. Von Buch's celebrated theory of dolomitization (applied, however, to rocks we have excluded) advocates that the lime- stones forming the dolomite mountains of the Tyrol have 92 ROCK-METAMORPHISM. been impregnated with magnesia sublimed from the adjacent diorites. Reverting to the magnesian limestones to which we have restricted ourselves, our view, as already made known in one of our memoirs*, is methylotic ; but as it differs in some important particulars from others that have been published, we proceed to give a description of it. Our objection to the opposite hypo- thesis involving simultaneous precipitation of carbonates of mag- nesia and lime is based on the absence of any recent deposits of the kind. The magnesian limestones of the north of England are plen- tifully fossiliferous ; and from the general facies of their orga- nisms, also their wide distribution, it seems far from improba- ble that they have been formed in deepish water of a wide-spread oceanf. There are beds, as the Durham marl-slate and the Manchester marls (the one underlying and the other overlying the magnesian limestones), which are largely characterized by carbonate of magnesia. To explain the methylosis of these different rocks the follow- ing hypothesis is suggested. It is based on the assumption, which has gained many adherents of late years, that during the Triassic period, particularly towards its close, the widely-spread ocean which had thrown down the Permian deposits became reduced to mere inland seas resembling the Caspian and Lake Aral. Our hypothesis assumes that, under these altered circum- stances, the water of the Triassic seas, besides much of it dis- appearing through evaporation, sank into the subjacent rocks, and that the salts of this water, under the force of chemical re- actions generated by new conditions, became variously combined with the materials of these rocks. The abundance of chlorides of sodium, magnesium, and calcium, and o£ sulphates of magnesia and lime in sea-water readily explains the origin of the rock-salt and gypsum in the lower members (red shales of Cheshire and Lancashire) of the Keuper J. We are more especially concerned, however, with the other deposits, resulting from the magnesium * Quart. Journ. Geol. Soc. vol. xxii. pp. 211, 212. t See ' Monograph of the Permian Fossils of England,' Palseontographical Society, Introduction, pp. xvii, xviii ; also Appendix in present work. I The thick beds of rock-salt at Nantwich, Cheshire, appear to be of METHYLOTIC ORIGIN OF DOLOMITES. 93 chloride and sulphate. As both are highly soluble compounds, and have no particular affinity for the basic substances of ordi- nary Triassic rocks (chiefly sandstones), they would penetrate the latter without being precipitated or producing reactions. But on coming into contact with the underlying calcareous marls and limestones of the Permian system, mutual decomposition would ensue. A portion of the lime of these rocks would be taken up by the chlorine or sulphuric acid (originally in com- bination with the magnesium of the Triassic sea-water) and carried off in solution ; while the base that had been in union with these acids would combine with the liberated carbonic acid to form carbonate of magnesia *. The latter compound, by virtue of its affinities, would become combined or incorporated with the remainder of the base of the marl or limestone t ; and thus these rocks would become dolo- mitic or converted into dolomite J. Another kind of methyl osis, however, supervened in some localities. The Permian magnesian limestones in the north of England are for the most part compact, earthy, or micro- crystalline; but in the neighbourhood of Sunderland they are singularly characterized by imbedded coralloidal, globose, mamillated, and other forms, whose secondary or superinduced origin, though first proved by Sedgwick, cannot be said to be generally admitted. Consisting for the most part of carbonate of lime in a crystalloidal condition, and containing only a few per cent, of carbonate of magnesia, it is a fair inference that these varied forms are a local development, caused by some demag- nesiating agent which penetrated the depositional partings and subaerial origin, in which case they may have resulted from the evaporation of the salt water of lagoons. * It may be admitted that carbonic acid would also be present in the water, and thus assist in the reaction. t Owing to the varying percentage of carbonate of magnesia in these rocks, it is not an undisputed point that this compound is in chemical combination with the carbonate of lime. t The paste of the Bristol dolomitic conglomerate may have been derived from a mechanically abraded Permian magnesian limestone j but the pebbles contained in it have clearly been detached from adjacent beds of Carbo- niferous limestone &c. The drift of Galway is essentially a calcareous deposit derived, by glacial degradation, from the carboniferous limestone of the district, and has consequently been mechanically formed. 94 ROCK-METAMORPHISM. well-developed joints of the rock1*; for it is a fact that they (coralloidal forms) are present on the grandest scale in immediate connexion with these divisional structures, shooting off from their planes, and appearing as if they were globose nullipores, stromatoporal sponges, branching corals, and other organisms f. This view of the demagnesiation of the coralloidal limestone seems to be perfectly warranted, since it is paralleled in the known cases of calcite occurring as a pseudomorph after miemite. According to our hypothesis of the dolomitization of the north-of-England Permian rocks, the magnesiated water has descended from overlying or subaerial sea-basins. But possibly in other cases, as in Ireland and the north of England, where local examples are common of ordinary Carboniferous limestone sud- denly becoming changed into dolomite, the water may have as- cended, through jointings and deposition-partings, from subter- ranean sources. We entertain no doubt, however, that the water was originally derived from the sea. There remains to be noticed another methylotic hypothesis, which has been suggested by ApjohnJ and Grandjean. Mr. Hardman adopts it in explanation of the Irish local examples referred to in the last paragraph §. He assumes that the lime- stone in these cases originally contained carbonate of magnesia, derived from corals and other organisms, a few per cent, (rarely 7) having been found in some living species. In limestones of the kind, when penetrated by water charged with carbonic acid, this solvent, it is suggested, would act on their calcareous * There are beds containing fossils at " Byers's Quarry," on the coast a few miles north of Sunderland, which have been demagnesiated, but without development of coralloidal structures ; they, however, are highly crystalline. t At the time of writing my Monograph of the Permian Fossils (see Introduction, pp. xiv-xxi) I was of opinion that the Permian limestones of Durham were originally dolomitic, but that in certain localities, as near Sunderland, they had undergone methylosis, the remarkable configurations being the result. A fine collection of these configurations, formed by myself during a period of several years, is contained in the Geological Museum of the Queen's College, Galway. An inspection of them will fully bear out the description of their similarity to certain organisms. — W. K. f " Analysis of some Irish Dolomites," Journal of the Geological Society of Dublin, vol. i. pp. 368-381 (1838). Apjohn, however, is more in favour of the rocks he noticed having been originally dolomitic* § Proc, R, Irish Academy, no. 7, pp. 705-730 (1876). METHYLOTIC ORIGIN OF DOLOMITES. 95 portion, carrying it off as a bicarbonate in solution without re- moving the magnesian constituent. According to Dr. Apjohn, " it is not improbable that this removal of the carbonate of lime may proceed until what remains behind contains the two car- bonates in the ratio of atom to atom/'' or in the proportion they are presumed to be in dolomite. This decalcifying process, it is conceivable, ought in many cases to be consummated by the removal of all the calcareous matter, and the consequent con- version of what was once a dolomitic limestone into a residual mass of magnesite. But it is extremely doubtful that any thing of the kind on a commensurate scale is known. BOCK-M ETAMORPHISM , CHAPTEE XVI. THE CHRONO-GEOLOGICAL KANGE OF OPHITES AND RE- LATED ROCKS, AND THE AGE OF THEIR METHYLOSIS. As, with certain exceptions already noticed, methylosis is con- secutive to mineralization, it may be taken as granted that, where both phenomena have prevailed, the former is of subse- quent elaboration to the latter. Our remarks on this subject, however, must be brief, as it is beset with no common difficulties, especially, as will have been seen, in respect to the oldest or Archaean rocks. Indeed^ amongst later systems insurmountable difficulties are encoun- tered, as in the case of the ophites and hemithrenes of Ireland and Scotland, which render it impossible to determine, with any but the most distant approach to certainty, the geological age in which these rocks passed through their last phase of meta- morphism. And to refer to post-Archaeans of the kind in North America, enough has been mentioned to show that not only the period of their methylosis, but the age of their deposition is at present involved in the greatest obscurity. Another difficulty attaches to rocks of a later age. In North- ern Italy, methylosis has been developed in formations ranging from an early period to the Jurassic ; and there are strong grounds for believing that in this region the formations of one system have undergone a change of the kind before those of a later system had been deposited. The only certainty in con- nexion with these rocks is that the latest change is of post- Jurassic age. The ophitic marble of the Isle of Skye, which is Liassic, it seems highly probable, has been mineralized and methylosed in much later ages ; possibly the latest is synchronous with the volcanic upbursts that have ravaged the western borders of CHRONO-GEOLOGICAL EANGE OF OPHITES. 97 Scotland, and which, according to Prof. Judd, took place during an early Tertiary period. In Eastern Europe serpentine masses are common in associa- tion with Upper Secondary formations. The euphotides and other igneous rocks of Central Italy, there can be no doubt, have broken through Cretaceous ' ' alberese }} limestone and Eocene sandstones and schists during a late Tertiary period, thereby producing the beautiful ophicalcite (oficalce), ophieuphotide, and gabbro verde for which Liguria is so renowned. It would even appear, from the discoveries of Achiardi, that the methylosis of rocks into serpentine is in process of elabora- tion at the present day in Tuscany. That the same process is still in operation amongst deep-seated rocks permeated by heated mineral waters, may be inferred with perfect confidence. APPENDIX. 99 APPENDIX. THALASSA AND XEKA IN THE PERMIAN PEBIOD. By Professor WILLIAM KING, Sc.D. &c. IN my ' Monograph of the Permian Fossils [of England 3* (1850) I expressed the opinion, based on the general character of their respective groups of fossils, that the formations of the Permian system had been deposited, some in shallow seas, and others in a deep ocean. In using these expressions it was meant to limit the application of the last one to depths known to long-line fishermen, and ranging from 50 to 100 fathoms or moref; for when this opinion was stated very little had been published respecting the existence of living things in the abysses of the great ocean. Further, in consideration o£ the fact that the Permian formations occupy widely separated areas, I inferred that the ocean (or seas) in which they were deposited was of considerable extent, encompassing, at least, a great portion of the European quarter of the globe. Investigations on Permian geology made during the last thirty years have in no way invalidated these conclusions ; yet Dr. Bamsay, in his f Physical Geology and Geography of Great Britain/ and certain of his memoirs, has been induced to interpret the evidences I relied on in a totally opposite sense. From the <( dwarfed aspect " of the Permian shells, and their " poverty in number," it is Dr. Ramsay's " belief" that the waters they lived in "were of an inland unhealthy nature/* comparable to those of " brackish lakes " like the Caspian. Before discussing these points, it is necessary for me to mention that I have lately (June 18, 1880) advanced the hypo- thesis that the regions over which the different rock-systems formulated by geologists are spread have each undergone, * Introduction, pp. xiii, xiv. t Several years ago I named a characteristic form of JSuccinum undatum^ var. pdagicum, on account of its living at depths ranging from 55 to 80 fathoms (Ann. & Mag. Nat. Hist, 1846, vol. xviii. p. 249). H2 100 ROCK-METAMORPHISM. during the period of formation of any one system, a cycle of vertical movements, upward and downward, which alternated with those of another cycle in an adjacent region*. Thus, as- suming a given region to be affected by an upward movement, it would be in alternating correlation with an adjacent region, where the opposite or downward movement would be going on. These movements would necessarily cause a region to be occu- pied by land, say in the beginning and ending of a given systemal period f, and by water in the middle division of the same chronological term. It will also be understood that the maximum of the down- ward movement of any one period would submerge a region to its greatest depth beneath the level of the sea; while the maximum of the upward movement would place it, as land, at its greatest altitude. Taking the vertical movements of a single cycle, the region affected by them would present a succession of land- and sea- features representing their different stages of development. Thus, in a region which has attained its maximum elevation, and which it may be assumed is in the 1st stage, land interspersed with lakes, rivers, and bordered by estuaries would predominate, producing lacustrine and estuarine marls, clays, and sands ; also other deposits, often terrestrial or conglomeratic, all more or less charged with the remains of a fauna and flora bespeaking the prevalence of widely-spread terrestrial conditions. In the 2nd stage, the movement being downward, and the sea necessarily encroaching on the lowlands, marine, littoral, and shallow-water deposits — argillaceous, sandy, marly, and calcareous — would be thrown down. In the 3rd stage — that in which the downward movement reached its maximum, and when the sea (Thalassa) * An abstract of a paper containing this hypothesis appears in the ( Pro- ceedings of the Koyal Irish Academy/ ser. 2, vol. iii. (Science), No. 5, Dec. 1880. A region is assumed to be of continental extent. See Supplementary Note A. t By a " systemal period " I mean a division of geological time, during which a rock-system was in process of formation : it may be said to corre- spond to a single cycle of vertical movements, Taking the masses of deposit constituting a rock-system, the mutations of its life-features, and the suc- cessive physical phenomena it witnessed, the conviction is strong in me that a systemal period is so vast as to be utterly beyond approximately calculating. APPENDIX. dominated over the region — pelagic deposits, eminently calca- reous and containing the remains of a deep-sea fauna, would be widely spread. Next, in the 4>th stage, opposite vertical move- ments supervening, the previously developed shallow-water phenomena would be repeated. Further elevation would bring on the last and 5th stage, in which, again, " the dry land (Xera) appeared " (&$6r) 77 grjpd. Septuagint, Gen. i. 9) . The following table will illustrate this hypothesis in its main points, at the same time affording evidence of its soundness by showing that one of the type rock-systems is constituted in accordance with its principles * : — Stage. Lacustrine, estuarine, anc terrestrial deposits . . . . Stage. Estuarine and shallow-sea sediments 3/Y? Stage. Deepish-water and pelagic formations 2nd Stage. Shallow-sea and estuarine deposits 1st Stage. Estuarine, lacustrine, and terrestrial dejections . . Permian conglomerates, &c. m Grits. Coal-measures of Durham and other places. Kilkenny coal-beds. Yorkshire Ganister coal series. Northumberland Millstone-grits. Malbay flags. Yoredale rocks. Scar Limestones, in many places alter- nating with shales. Tweedian beds. Lower Limestone shales. Lower coal-beds (? estuarine). Knocktopher Conglomerates. Devonian. If facts could be reconciled with Dr. Ramsay's belief, I should have no hesitation in discarding the Permian group of rocks as a system, and transferring it to the Carboniferous. Feeling * A few notes are necessary in this place. The recognized rock-systems afford ample evidence that the various sedimentary developments involved in this hypothesis have often deviated in a marked manner from regularity ; they have been irregularly recurrent, and have varied in horizontal and vertical extent, in time-limitation, and in force, as evidenced by the sudden and repeated changes (from freshwater to marine conditions) in rocks of a formation, the presence of a well-developed formation over a certain area and its absence in another and contiguous area, the dissimilarity between syn- 103 ROCK-METAMORPHISM. satisfied, however, that the group contains formations referable in the main to the different stages of a cycle of vertical move- ments, I cannot but regard it as of systemal rank, though I admit that it cannot be equalled in comprehensiveness with certain other rock-systems. The following table exhibits a series of formations representing the different stages, as successively developed in the course of a cycle of vertical movements which took place during the Permian period : — Lancashire Triassic Bunter sandstones. 5th Staff e. St.-Bees (Whitehaven) red sandstones. 4:th Stage. Manchester marls and thin-bedded limestones. Yorkshire and Barrowmouth (Cumberland) Schizodtis-limstones and gypseous marls. Sunderland coralloidal limestones. Hartlepool and Marsden lime- stones. Ardtrea (co. Tyrone) magnesian limestone. 3rd Stage. Humbleton (.Durham) fossiliferous limestones. 2nd Stage. Midderidge (Durham) compact limestones. Durham marl-slate. \st Stage. Pontefract sandstones. Solway red-sandstones. Westoe (Durham) >%^«rm-sandstones. Coal-measures of the Carboniferous system. chronous formations in separated areas of one and the same region, and the vast difference in thickness of the rocks severally constituting the different systems. Rock-systems are often deficient in formations representing the 1st and 5th stages : such are usually tripartite, as the Triassic ; though in some instances of the kind the missing formations have doubtless been removed by denuda- tion, which, as will be readily understood, must be energetic during these two subaerial stages. The Carboniferous system may be taken as an example in which all the stages, especially the 1st and 5th, are well represented. The Triassic system, which on the continent comprises three formations cor- responding with the 2nd, 3rd, and 4th stages, is in the British Isles devoid of any representative of the middle one (3rd) of these stages. On the other hand, this stage is well developed in the Cretaceous system of the south of England and other countries; but the 5th stage is poorly represented in most regions, except, seemingly, in the Rocky Mountains and the more western ranges, where occur intercalated formations containing Ammonites Baculites, &c., and Tertiary plants, which indicate the missing time-link between the Cretaceous and Eocene periods. (See Prof, J. Stevenson American Philosophical Society, June 18, 1875.) APPENDIX. ; 103 To accept Dr. Ramsay's belief, it would be necessary to over-, look altogether the middle (or 3rd) and much of the 2nd forma* tions of the Permian system, Reverting to Dr. Ramsay's grounds for assigning a brackish- lacustrine origin to the Permian rocks in general, I propose, in the next place, to notice the one founded on the " poverty in number of the fossils " they contain. Conclusions on this point are not to be influenced by consi. derations arising from the study of individual characteristics, but of general facts. Evidently Dr. Ramsay has been in- fluenced in his belief by the fact that brackish lakes, com- pared with sea-basins, are sparsely inhabited by molluscs \ but he overlooks another fact of importance, that depths exceeding 100 fathoms are not so prolific in animals of the kind as shallower bottoms. " Poverty in number " as a feature of deep-sea life affords a more satisfactory explanation of the point in question than the corresponding feature in its brackish^ water relations. The same argument attaches to the t( dwarfed aspect " of the Permian, invertebrates ; for the term " dwarfed " is equally applicable to a deep-sea fauna, which usually consists of small and delicate species. Such forms as Productus horridus, with its long projecting hinge-spines, and Fenestella retiformis, with fragile fronds six to eight inches in spread, and several other tender organisms could only live at considerable depths, where still water prevailed. But it cannot be said that the Permian fossils are particularly dwarfed; the Bryozoon just referred to is a strong fact against any statement of the kind. Spirifer alatus does not compare unfavourably with most of its Carboni- ferous congeneric species; and Camarophoria multiplicata is a match for the largest of its allies, C. Kingii (Davidson), of the Carboniferous limestones. Several other Permian fossils could be mentioned in this comparison. It is quite true that, so far, the Permian rocks have not yielded any thing equal to Productus giganteus and P. ponderosus and other heavy fossils, especially corals ; but these may be safely consigned to shallow seas. It is necessary to mention that the Permian palliobranchs were denizens of deep water ; for in fossiliferous beds which bear all the marks of having been formed in shallow water these shells are absent. Certain beds of coralloidal limestone : .\\ k ^EOCK-METAMOEPHISM. on the sliore at Sunderland which came under my notice have their surfaces distinctly rippled j but they contain no fossils of the kind — merely bivalves (Schizodus &c.) . These beds,, however, do not belong to the deep-water stage; for they overlie the limestone (fossiliferous par excellence) which crops out at Hum- bleton and other inland localities. Other beds occurring in Lancashire, Yorkshire, and Cumberland, which possess similar palseontological characteristics, are referable to the same stage. Another fact of importance, which also seems to have been overlooked by Dr. Ramsay, is that the Permian invertebrates, which he assumes to have lived in " brackish water " of an " inland unhealthy nature," are, as a rule, well-developed species, and bear the stamp of having inhabited an open sea. How can it be conceived that palliobranchs, only known as denizens of the sea, could exist in brackish-water lakes ? As to why an exuberant variety of invertebrate marine life is not a characteristic of the Permians, it must be borne in mind that information on the distribution of these rocks is still very limited. In the face of certain deposits in North America and Central Asia, doubtfully referred to the Carboniferous, or Permian Triassic system, it seems preferable to wait until they have received fuller attention. In the Austrian Alps and in Northern India there occur fossils (Ammonitida, Goniatidae, &c.) of types indicating that the rocks containing them may be of Permian age. Now, as my theory of regional cyclical vertical movements admits of different conditions prevailing at the same time in separated regions, it is not at all improbable that the marine formations referred to may belong to the pelagic stage, and nevertheless be synchronous with those which in Western Europe gave birth to the shallow- water and estuarine (2nd or 4th) de- posits of the Permian system. I would suggest the application of this argument by way of explaining the difference between the rock-systems of Europe and their presumed equivalents in extra-European regions. Dr. Ramsay has introduced into his remarks on the Physical Geography of the Permian Period * another matter, which he * Op. cit. pp, 149, 150. APPENDIX. 105 evidently thinks supports his belief, but which, as will be learnt from a perusal of Chapter XV., I am totally opposed to. To quote his words : — " I repeat that the Permian magnesian limestone was not, as used to be suppose^ formed in the sea, but in an inland salt lake, under such circumstances that carbonates of lime and magnesia were deposited simultaneously, probably by con- centration of solutions due to evaporation. In an open sea lime and magnesia only exist in solution in very small quantities ; and limestone rocks there are formed (as in coral reefs) by organic agency." If, as seems to be meant by the last sentence, lime and magnesia are unlikely to be precipitated in an open sea (a view I quite agree with — not, however, because salts formed of these bases are in small quantities, which statement is a slip of some kind), it is to be apprehended that there is little chance of their being deposited in an inland salt lake, consider- ing that Dr. Ramsay has not been able to point out a single instance of the kind. " In some of the lower strata of the magnesian limestone, when weathered, it is observable that they consist of many curious thin layers, bent into a number of very small con- volutions, approximately fitting into each other, like sheets of paper crumpled together. These dolomitic layers convey the impression that they are somewhat tufaceous in character, as if the layers, which are unfossiliferous, had been deposited from solutions. In other parts of the district, along the coast of Durham, large tracts of the limestone consist of vast numbers of ball-shaped agglutinated masses, large and small ; and I have observed in limestone caverns, in pools of water surcharged with bicarbonate of lime, that sometimes precipitation takes place on a small scale, producing similar nodular bodies. It is notable also that, when broken in two, many of the balls are seen to have a radiated structure ; that is to say, from the centre rudely crystalline -looking bodies, all united, radiate to the cir- cumference. In other places we find numerous bodies radiating in a series of rays that gradually widen from the centre, and are unconnected at their outer ends, which reminds the spectator of radiating corals. There is, however, nothing organic about them ; and I do not doubt that they owe their growth to some kind of crystalline action going on at the time that the limestone was being formed." 106 ROCK-METAMOEPHISM. The facts on which the above description is based I must regard as affording no more support to Dr. Ramsay's belief than the fossils that have been under consideration. It appears to me that the coralloidal and other bodies described in the foregoing extract, besides having been examined with insuffi- cient deliberation, have not been considered in connexion with the evidence adduced by Sedgwick and myself, proving them to be of superinduced origin. I pointed out upwards of thirty years ago their relation to well- developed rock-jointing*, a fact which may be unhesitatingly accepted as disproving the idea " that they owe their growth to some kind of crystalline action going on at the time the limestone was Jbeing formed/' And with respect to the presumed identity between the ef ball- shaped agglutinated masses" of the Permian limestones and the " nodular bodies precipitated in pools of water of caverns," the two phenomena are not to be compared, — the one (taking ordi- nary stalagmitic deposits into consideration) being a simple product, and the other a complex development — the " balls " of the " masses " essentially consisting of carbonate of lime, and their matrix of carbonates of lime and magnesia. In conclusion I would respectfully urge on Dr. Ramsay to consider the various points that have been adduced against his belief; at the same time I must express myself as indulging in the hope that the next edition of his valuable work will contain views on the physical geography of the Permian period more in harmony with the general evidences of the case. * See Supplementary Note B, SUPPLEMENTARY NOTE A. 107 &UPPLEMENTARY NOTE A, IN the abstract referred to in a previous page I first broached the hypothesis of cyclical vertical movements on a regional scale in connexion with ef Rock -jointing in its relation to Phe- nomena in Physical Geography and Physical Geology/' and as the subjects it treats of are of general importance, and exciting much attention just now, I may be excused introducing them into the present note. In my Report on Jointing and Slaty Cleavage, which appears in the ' Transactions of the Royal Irish Academy/ vol. xxv. (1875), reasons are given in favour of the doctrine that jointing is a physical phenomenon, which, constituting lines or zones of weakness in the earth's crust, has permitted subterranean dis- turbances, often accompanied by igneous upbursts, to follow the courses of these zones*. With respect to slaty cleavage, in consideration of various evidences adduced in the Report, I have for several years past maintained that it is the result of pressure consequent on subterranean disturbances exerted against planes of jointing (Bangor slates), or depositional partings (Delabole slates), — thus bringing them into approximate or immediate contact. Since the Report appeared, Daubree has given a description of some experiments which he regards as proving the mechanical origin of jointing. By means of torsional pressure applied to plates of ice, he has developed, in the latter, groups of " approx- imately parallel" lines of fracture, crossed by other lines nearly at right angles. But I cannot regard these experiments otherwise than as simply illustrating- the well-known truth that similar effects are produced by dissimilar causes ; and it may be strongly contested that ice, crystalline in its origin — in which it is doubtful that structural planes, original or superinduced, are ever absent — is a suitable substance on which to experiment by way * This doctrine suggested itself to me on my becoming acquainted with an early speculation in dynamical geology of Phillips's. See op. cit. pp. 637 and 638, and Phillips's Report Brit. Assoc. 1834, p. 657. 108 ROCK-METAMORPHISM. of illustrating a phenomenon abundantly characteristic of me- chanically deposited rocks. I do not deny that pressure, under certain conditions, has produced parallel fractures in rocks; and examples could be cited of a mechanically developed parallel divisional structure such as some geologists have taken for jointing. Still, as jointing is common in its indisputably typical or normal condi- tion, without a tittle of evidence of mechanical pressure having been in any way concerned in its development, obviously a physical causation must be ascribed for the phenomenon. As stated in my Report, there are miles and miles of Carboniferous limestone, in nearly horizontal beds, forming the bare surface of many parts of counties Clare and Galway, in which jointing is wonderfully developed ; but nowhere is it accompanied by evidences of crush or stratal disturbance involving a mechanical causation. On the contrary, the phenomenon so closely simu- lates mineral cleavage in fineness, and it is so completely divested of all indications of supervened compression, that the view of its mechanical origin, advocated by Daubree and others, must be regarded as completely at fault. M. Daubree, however, by way of adducing some evidence of the required pressure in rocks, has brought forward a case notified by Harkness and examined by myself*. It is an instance in Carboniferous limestone near Cork, which not only comprises typical jointing, but shows indisputable evidence of the rock having undergone powerful compression. This case, however, is altogether valueless with reference to the mechanical hypothesis; for it can readily be proved that the jointing (which is meridional) has been developed after the rock had become compressed. Clearly the divisional structure in this instance is altogether independent of compression. The fact is, the Cork limestone (as well as its associated rocks) has been flexured into east-and-west rolls, corresponding in direction with an old equatorial jointing, traces of which are still to be seen in the roughened and dislocated divisional planes everywhere present. On the contrary, the jointing referred to by Daubree cuts cleanly through the disturbed rocks, north and south, and holds on for miles in the same way as the meridional jointing of Ireland generally. * Op. cit, pp, 638 & 639, SUPPLEMENTARY NOTE A. 109 In the matter of slaty cleavage, Sorby (who, it is well known, ably supported the view originally brought out by Daniel Sharpe many years before my Report was published) has of late returned to the subject. In his Anniversary Address, as President of the Geological Society, he makes known some microscopic observations in connexion with the development of slaty cleavage. It would appear " that, though in some cases " his " original explanation of very perfect cleavage may be true " (that is, due to pressure alone, aad independent of any pre- existing divisional structure), he now recognizes "fine lamination in the plane of deposition" and " very close joints," as noteworthy agents, and affording an ee explanation which removes a very seri- ous difficulty in completely explaining the mechanical origin of slaty cleavage in rocks which have yielded to pressure as imper- fectly plastic substances "*. Do not these observations manifest a decided leaning in the direction of the theory which, for several years, I have been advocating ? — viz. that slaty cleavage, instead of being simply the result of pressure, is the outcome of pressure exerted against preexisting divisional planes of jointing and bedding. Reverting to the disturbances and accompanying igneous upbursts which have followed zones of jointing, the least reflection will make it clear that they would powerfully affect rocks possessed of this divisional structure, assumably com- pressing them at right angles to the course of an axis of disturbance, and bringing the joint-planes into immediate con- tact, thus developing slaty cleavage. The same agencies, besides producing enormous dislocations or faults, must have, by their transgressive action, flexured rocks into mountain chains and intervening valleys, also into parallelism with an axis of dis- turbance. Although agencies of the kind must have often obliterated jointing, I assume that cleavage-planes represent it — also that the strike of these planes indicates the course or direction pur- sued by the obliterated jointing, and consequently the system to which this divisional structure belonged. There are two systems of the kind — one meridional, and the other equatorial — depending on the maximum frequency of the joints within certain points of the compass. The first, especially, * Quart. Journ. Geol. Soc, vol. xxxvi, pp. 72-74 (1880). 110 ROCK-METAMORPHISM. is divisible into two sections — east-of -north, and west-of-north ; a third section, not so strongly developed, remains to be added, which, running north and south (that is, between the others), may be called medio-meridional. The equatorial system, in general not so well developed, may also be divided into two or more sections. As the earth's continents and great peninsulas (India &c.) have their chief coast-lines run- ning in directions corresponding to the two principal meridional sections (which is also the case with the main trends of the islands of the Pacific and other oceans), my contention is that their east coast-line belongs to, and has been aligned by, the east-of-north meridional jointing — also that their west coast-line stands in corresponding relation to the west-of-north section. Thus I look upon the coast-lines of continents as a correlated phenomenon, taking these features to be defined by the edges of the great submarine plateaux which, stretching out for 200 miles or more, abruptly terminate in a succession of bench-like terraces, suddenly descending into the abysses of the oceans. Respecting another prominent feature of our continents, I offer the suggestion that the east-of-north and the west-of-north sec- tions of meridional jointing have primarily marked out the sides of the triangle under the form of which these great land-masses are for the most part presented ; while the base of the triangle is ascribed to equatorial jointing. But as it is not yet clear to me why the base of the triangle faces the north and its apex points to the south, I am inclined to think the solution lies in the fact that the greatest elevated land-masses characterize the northern hemisphere and equatorial regions — a disposition which would cause a greater width of elevated land to lie within the basal area of the triangle than at the apex. This last point requires to be considered in connexion with the theory originally advanced by Dana"*, that the earth's con- tinents have always been continents, or, as I prefer to put it, that the present continents, in the main, have been from the earliest geological periods greatly elevated regions separated by enormously deep depressions. Dana, however, contends that the great land-features of the globe have been produced by regional up-bendings and down-bendings of its crust, the former having given rise to continental masses, and the latter to vast ocean-basins ; whereas, although accepting the pre- * { American Journal of Science/ 1846, and ' Manual of Geology.' SUPPLEMENTARY NOTE A. Ill Cambrian antiquity of the main surface-features of the earth's crust, it is my opinion that our continental coast-lines are in correlation with enormous faults, which have thrown down the rocks on one side of a dislocation thousands of feet below their corresponding masses on the other side, and, furthermore, that the general direction of any given continental coast-line has been determined by some system or section of jointing. In contending that jointing, slaty cleavage, great lines of faulting, continental coast-lines, and mountain- chains are cor- relative phenomena, I feel myself powerfully sustained, not only by the parallelism between the United- States coast and the Appalachian ridges, but equally by the corresponding parallelism of the enormous faults (some with a downthrow of thousands .of feet) which characterize this mountain- system. One of the faults is known to stretch from Quebec to New Jersey ! The disturbances which developed the " great feature-lines " of our globe seem to have been in operation in pre-Cambrian periods. Evidences have been discovered in the Wahsatch range (Rocky Mountains), by Clarence King, of an elevated mass, defined in one tract by a nearly vertical cliff, which, with an altitude of 30,000 feet, was presumably in existence before the earliest palseozoic deposits were formed ; and there are the strongest evidences for the conclusion that, before the adjacent Cambrian rocks were deposited, the Archseans of North America had been violently flexured, and thrown up into ridges — belong- ing to two divergent systems conformable with the zones of weakness which determined the east-of-north and the west-of- north outlines of this great continent ; the ridges forming in its central region a mountain-mass in pre-Cambrian times. The agent which gave an east-of-north trend to the west coast of Europe similarly affected much of the north-west coast of Africa : seemingly it struck obliquely across the equatorial section of the Atlantic, reappearing at Cape St. Roque, and proceeding onward along the mountainous sea-board of Brazil to the La Plata. Enormous as undoubtedly is this extent of coast-line formation, it is surpassed by what is presented on the west coast of the two Americas and the east coast of Asia : obviously the former, with its parallel mountain-ranges, is in genetic relation with the west-of-north and median sections of meridional jointing, and the latter with the east-of-north section. Attention may next be directed to the great inland ranges 112 ROCK-METAMORPHISM. constituting the Alps and the Himalayas. Both mountain- masses have been more or less affected by forces exerted in direc- tions corresponding with the two principal sections of meridional jointing; but in both cases the phenomenon has been greatly swayed by movements presumably acting under the influence of the equatorial system. The disturbances which developed High Asia — a vast conti- nental mass within a continent stretching from India to the tundras of the Taimyr peninsula in Asiatic Siberia — have ridged it up transversely into mountain- chains, with intervening desert platforms. Its southern extremity is formed by the Hi- malayan and other ridges, whose general level (20,000 feet or more in altitude) culminates in still loftier peaks, some not far short of 30,000 feet. All the transverse ridges, in a great (or the middle) portion of their length, have an east-and-west course. The development of High Asia is a vastly complex phe- nomenon, presumably resulting from disturbances which have been directed along different zones of weakness. The zone in correlation with equatorial jointing seems to have been the medium through which the transverse ridges and their respective igneous axis were upheaved ; while those referable to the two principal sections of meridional jointing may have similarly influenced the terminations of the ridges on both sides of this huge plateau, especially in the region east of it, where mountain-ranges, coast-lines, and off-lying islands all coincide in their strike with the east-of-north meridional jointing. The lofty parallel ridges east of Burmah, in being medio-meridional, are so far in conformity with the last-mentioned features. This brief notice will scarcely permit of any reference being made to the equatorial extensions from the Pamir through Western Asia, &c. The question next suggests itself, arising from a consideration of all the phenomena that have been noticed — If the great pre- Cambrian plateaux have always existed as masses, having an ele- vation far above the bottom of the great intervening depressions (ocean-basins), how have they become covered up with marine sediments thousands of feet in thickness, and representing suc- cessive geological periods ? In this connexion it may be argued that elevations of rock-masses are of two kinds ; — one due to stratal disturbances, which for the most part have been exerted SUPPLEMENTARY NOTE A. 113 horizontally, or approximately so, and the other to vertical movements extending over wide geographical areas. More than thirty years ago my attention was called to the latter class of movements by the beautifully developed series of terraces in the Burren of Clare, 'reaching to the height of nearly 1200 feet. This particular instance I have ascribed to a slow upheaval of the district above the sea, the surface of each terrace representing the bottom of a coast- shore — a plane of marine denudation — and an intermittent stoppage in the upheaval*. An examination which I made in 1870 of the terraces of Lochaber resulted in my becoming convinced that they are ancient sea-margins : 1495 feet is the height usually stated of these terraces ; but I detected on the flanks of Ben Nevis, and of the opposite mountains, the like features, which must reach to an altitude of between 2000 and 3000 feet. Besides the raised shell-beaches standing at a com- paratively low level on the coasts of Norway, terraces have been lately observed and described by Daykins, which occur on the Dovrefjeld, at heights of from 2000 to 3100 feet. Darwin's account of the remarkable examples that occur in Patagonia, up to the height of 1300 feet, leaves no doubt on my mind that they have been formed by the action of the sea. Hector has described vast terraces on both the eastern and Pacific slopes of the Rocky Mountains, stretching from the Athabasca river to Mexico, and rising one above another to heights ranging from 3500 to 4500 feet above the level of the sea. Well marked parallel terraces are striking features in other parts of North America. A series of " horizontal benches/' twenty in number, deeply cut into the mountain-slopes, and situated at heights between 1100 and 2580 feet above the sea-level, extend over an area of 10,000 square miles both east and west of the Alleghanies of Pennsyl- vania> Virginia, and Maryland. As properly remarked by Prof. Stevenson, who has lately described them, "they can be no other than sea-beaches marking stages in the withdrawal of the ocean" f. The late Daniel Sharpe made known the occurrence of lines of erosion on the inner and outer flanks of the Swiss Alps, at about 4800, 7500, and 9000 feet above the sea. And, to finish what could be made a much longer list, Rudolph Griesbach has described terraces in Natal lying at heights * See < The Geologist,' vol. vi. pp. 172, 173 (1863). t American Philosophical Society, August 15, 1879. 114 ROCK-METAMORPHISM. of about 1000/2300, and 5000 feet: it would also appear that these correspond with the main plateaus of Cape-colony. In short, it may be safely stated that marine terraces are to be seen in every region of the globe. In the deep valleys of the lofty southern buttress (Gangri range) of Thibet, terraces ascend to the height of 16,000 feet ; but as these may have been formed along the shores of elevated lakes, such as are now in Ladak and adjacent countries, it would be unsafe to classify them with the marine representatives that have been noticed. It may nevertheless be maintained that a number of geological evidences afforded by the area last noticed, combined with the proofs already brought forward, establish the conclusion that vertical movements of vast regional extent have affected not only High Asia but the entirety of the earth's surface — elevating con- tinents, including mountains and plateaus, at the same time uplifting the bed of the intervening oceans thousands of feet above their present level relatively to that of the sea, or plunging them as deeply in the opposite direction. Without denying that the level of the sea may have undergone great fluctuations at intervals during past geological time, caused by seonic flows and ebbs of the ocean, and that such changes may have participated, to an extent far beyond what physicists and hydrographers are at present disposed to admit, in developing phenomena which, for convenience' sake, I have collectively ascribed to vertical movements of the earth's crust — or without offering any opinion respecting the hypotheses suggested by Babbage, Herschel, and others as to the cause of phenomena of elevation and subsidence — it does not appear improbable that cyclical or periodically recurrent vertical movements, each one representing a vast chronological term, have by slow degrees alternately elevated and depressed opposite areas, corresponding in extent with a continental or even a hemispherical division of the globe. To illustrate this view, let it be assumed that one of our con- tinents, having attained its maximum elevation, is next to undergo subsidence* During this elevated period the land" surfaces of moderate height would be in what may be termed the first stage of depositional action, viz. the formation of subaerial, freshwater, and estuarine deposits ; in the second stage the same SUPPLEMENTARY NOTE A, 115 areas would be subjected to marine actions, producing littoral and deepish-water conglomeratic, arenaceous, argillaceous, and cal- careous beds ; in the third stage, in which maximum depth had been reached, they would be under pelagic conditions, developing limestones, argillytes, and siliceous rocks. Next, elevation having again come on, the fourth stage would be a repetition of the second, yielding comparatively shallow-water marine de- posits ; and this would terminate by passing into the fifth stage, which corresponds to the first one. Thus would our continents, notwithstanding their being at present at an average height of a few thousand feet above the sea-level, become overlaid in every systemal period with vast deposits of all kinds — those of any given stage representing one of the formations (assumably five) which constitute a geological rock-system, and, moreover, the whole agreeing with the formations of a system in their suc- cessive order of superposition. A few points may be briefly added. It is not assumed that all such vertical movements have proceeded in the invariable course and to the extent, vertical or areal, above illustrated, or that they were unaccompanied by minor ups and downs. The region opposite to the one given as an illustration would be undergoing a cycle of counter vertical movements. As to the deposits which were thrown down over the abysses of the oceans (Atlantic and Pacific) when the continents were under pelagic conditions, it is admitted they involve some questions difficult to answer, whether considered in connexion with Dana's hypo- thesis, or the one just stated. Obviously great recurrent climatal changes would result from these elevations and depressions — severe glacial conditions accompanying the one, and the replacement of the latter by genial ameliorations arising from the other. The consideration of these points gives rise to the question, often debated, How has it happened that after the Pliocene period climatal conditions prevailed which converted a great portion of Europe and North America (there are grounds for excepting Northern Asia, into ice-covered regions, and that during the Miocene (or probably some portion of the Pliocene) period areas lying within from 10° to 20° of the North Pole have enjoyed, as it appears to some geologists, a climate approaching in genialness that of the south of Europe at the present day ? 116 ROCK-METAMORPHISM. Influenced by objections to all the hypotheses that have been offered in explanation of these climatal interchanges, I find a much more tenable one in that which attributes them mainly to the aforementioned vertical movements. Confining myself to the climatal conditions which charac- terized Grinnell Land, Spitzbergen, and other Arctic areas during the Miocene period, as indicated by their plant-remains, I make the suggestion that these and adjacent areas stood at a somewhat lower level relatively to the sea than at present, and formed an archipelago, freely permitting currents with a tem- perature slightly more elevated than that of the gulf-stream where it now strikes the west coast of Ireland to bathe the coasts of its islands. Climatal amenities now prevail in Arctic Scandinavia, the Kara Sea, and on the western border of the Taimyr peninsula in Asiatic Russia : the last place, the most northern continental land of the globe, now supports an exuberant forest vegetation in a much higher parallel than anywhere else within the Arctic circle, and only about 16° or 17Q short of the North Pole — the fact being seemingly due to the presence of warm water, carried by ocean-currents and by rivers (as the Yenissei) from the south. Boreal Siberia, in direct communication with southern lakes, inland seas, and ocean- streams charged with warm water, and in the condition of an archipelago, — why may not its great forest belt be extended up to Spitzbergen and Franz -Joseph Land — to parallels corresponding with those in Grinnell Land, which formerly supported the growth of a vegetation approximately similar in some respects to that now characteristic of Northern Italy and the Southern States of North America ? As to the long winter-night of darkness and the long summer- day of sunlight, I feel satisfied from adducible evidences that, other things being favourable, such conditions would rather favour than impede vegetable growth. Areas favourably situated as to shelter, meteoric influences, soil, proximity to warm currents, &c., and especially where a thick covering of snow prevailed during the severe months of winter, would, in my opinion, become genial oases supporting an exceptional vegetation : such I look upon were the places in Grinnell Land and Spitzbergen where, during the Miocene or early portion of the Pliocene period, flourished guelder roses, water- SUPPLEMENTARY NOTE A. 117 lilies, sequoias, swamp-cypresses (Taxodium), &c. of extinct species, and varieties of known species now living in temperate latitudes, and doubtless acclimated to arctic conditions. The adaptive constitution of plants has not been sufficiently considered in connexion with this question. I shall conclude with a few brief remarks on linear igneous disturbances. Admitting the existence of a number which may be included in the equatorial system of jointing, disturbances of the kind are for the most part limited to the meridional zones of weakness. As is well known, a most important series of volcanoes characterizes the western borders of the two Americas ; and a similar series lies off the east coasts of Asia, belonging, in my opinion, one to the west-of-north and the other to the east-of-north section of meridional jointing : both series become united in Behring's Sea. Other writers, by connecting the equa- torial series of volcanoes north of Australia with the above two, have constructed a " circle of fire ;" but with far too limited a range. The two meridional series (by pursuing a direct course, so as to embrace the Cocos Islands, St. Paul's, Kerguelen's Land, Enderby's Land, thence curving to Trinity Land, passing on to the South Shetlands, and through Fuegia into the Patagonian Andes) form but one a great volcanic girdle, which may be said to stretch without interruption round the world, traversing the Arctic regions a few degrees east of the North Pole, and inter- secting the Antarctic circle at a corresponding distance west of the South Pole — thus dividing the crust of the globe along its greatest zones of weakness into two nearly equal halves, and at the same time separating its superficies into a water- and a land- hemisphere. As to reflections which may naturally arise in connexion with the last subject, I avow myself to be, scientifically, too much of a teleoptimist, too extravagant a timist, and too little of a catastrophist to entertain any that involve serious or disquieting apprehensions. SUPPLEMENTARY NOTE B, ON THE CRYSTALLINE BODIES OF THE SUNDERLAND PERMIAN MAGNESIAN LIMESTONE. Fig. 1. Fig. 2. FIG. 1 represents a section of Permian limestone, exposed in a railway-cutting between Sunderland and Ryhope, as displayed when I made a sketch of it about the year 1846. The beds exhibit well-developed jointing, running in two directions. The front or face of the section is a joint-plane (as is also the one further, in) belonging apparently to the equatorial system ; the oblique lines, also the end-surface parallel with them, belong to the west-of-north meridional set. The horizontal lines and the corresponding surface at top represent bedding. The interest which attaches to this section is the fact, well displayed, that the coralloidal bodies spring from the planes of both bedding SUPPLEMENTARY NOTE B. 119 and jointing. I have repeatedly observed the same thing at other places in the neighbourhood of Sunderland; on one occasion a coralloid with a stem as thick as a man's arm, came under my notice. It is also a common occurrence, where semi- globular bodies are developed, for them to be integrally con- nected with the surface-portion (usually upper) of a bed. Fig. 2 represents a portion (about a foot square) taken from a bed at Building Hill. The oblique lines represent west-of- north joints, and the horizontal bounding lines bedding-planes, The specimen, with several others of the kind, was collected about the year 1839, on an occasion when the quarrymen had exposed a singularly beautiful development of such forms. In this example, as in the section fig. 1, the coralloids branch off from the planes of bedding and jointing ; but one of its features is worth special notice : the coralloids, according to a note I made at the time, are best developed where branching from the west bounding-plane of the meridional joints. These examples are amply sufficient to disprove the idea that they were formed at the same time as that in which the limestone was deposited j also that, howsoever they may have originated, the agent which produced them must have penetrated the partings of both jointing and bedding. It cannot be too strongly impressed on the mind of the reader that the coralloids, also the other configuration to which Dr. Ramsay has referred, are more or less crystalline internally, and consist of carbonate of lime, adding a few per cent, of carbonate of magnesia ; while their matrix, or the body of the rock, is structureless and essentially dolomite. 120 ROCK-METAMOEPHISM. SUPPLEMENTAKY NOTE C. CEKTAIN LIMESTONES ARE OF MECHANICAL ORIGIN. IT is well known that Dr. Sterry Hunt, in advocating the purely chemical origin of limestones, places himself in op- position to the opinion generally prevailing on this subject, viz. that while some rocks of the kind are chemical products, such as freshwater and marine travertines, by far the greater portion are of organic origin, that is, the skeletal exuviae of shells, corals, foraminifers, and other animals. Excluding the methylosed crystalline limestones, I quite agree with this view ; but I have now to bring under notice another class of calcareous deposits, whose origin, it is assumable, was altogether different ; though they may have been primarily of organic, chemical, or methylotic development, The calcareous nature of the erratic drift, so well developed near the " Citie " of Galway (to which reference has been pre- viously made, footnote, p. 93) , can only be due to this deposit having been derived from the Carboniferous limestone of the surrounding district by the abrading action of glaciers. From what has lately come under my notice I see no reason why certain calcareous deposits, obviously of littoral origin, cannot be the debris of mechanically abraded limestones. Lately, availing myself of an occasion when Lough Corrib was in a muddy condition, caused by heavy rains, I put aside a couple of quarts of the water to stand for a few days* Testing the sediment which had settled at the bottom of the vessel with hydrochloric acid, a brisk effervescence took place, denoting it to contain a notable quantity of carbonate of lime. The remaining clear water, on being tested with oxalate of ammonia, exhibited decided evidence of its containing bicarbonate of lime in solution. These facts prove that the water of rivers in limestone districts contains calcareous matter both mechanically SUPPLEMENTARY NOTE C. 121 suspended and in a chemically dissolved state ; so that, while the latter may be carried out into the open sea, to be appro- priated by shells, corals, &c. for their skeletons, and the debris of these converted into limestones, the former may be mechani- cally deposited in estuaries and along shores. I have often thought over the fact that many limestones are so greatly deficient of calcareous fossils as to render their organic origin doubtful : the lithographic limestones of Bavaria are cases in point, also the Permian marl-slates of Durham and Germany. I am therefore now strongly inclined to assume that these and other calcareous deposits are of mechanical origin. When writing the footnote above referred to, I thought it not improbable that the dolomitic -conglomerate of the Bristol district was a Triassic glacial deposit, its paste having been derived from a Permian magnesian limestone now entirely removed. But from information I have liberally received from Prof. Sollas (whose short note in the f Geological Magazine ' of February last led me to put a few questions to him) I feel per- suaded his opinion is correct, that both the paste and the pebbles it contains have been dolomitized since they were accumulated, and that the paste had not been derived from a Permian limestone. Prof. Sollas and Mr. Margetson, I understand, are preparing for publication an account of the Bristol rocks. DESCRIPTION OF THE PLATES. PLATE I. Fig. 1. Chrysotile interlamellated with serpentine. Colafirth, Shetland. Fig. 2. Lamellated ophite (" Eozoon Canadense"), consisting of layers of serpentine and calcite in parallelism with one another. Canada. Fig. 3. Peridote (olivine) enclosing fibrous or striated laminae. Elfdalen in Dalecarlia. Fig. 4. Feldspar enclosing fibrous or striated laminae. Fig. 5. Graphic granite, consisting of layers of feldspar (a) alternating with others composed of quartz (6). The former are transversely intersected by striated fibrous and striped laminae. Harris, Hebrides. — Specimens of the kind have been taken for a fossil related to " Eozoon Canadense " ! Fig. 6. Specimen in which there is a definite and parallel alternation of reddish quartz and red feldspar, rivalling that of the lamellated variety of "Eozoon Canadense." Astracan. Fig. 7. Transverse section of a crystal, presumed, from its cleavage, to be augite ; an optical examination, however, shows that its substance is peri- dotized. Vesuvius. PLATE II. Fig. 1. Chrysotile changing into' "the acicular varieties ; also the same replaced by calcite (a) : the variety c answers to the " proper wall," and the calcite to the " intermediate skeleton " of "Eozoon Canadense" Reichenstein. Fig. 2. Portion of specimen of t( Eozoon Canadense" Layer of chrysotile («) in parallel alternation with serpentine (b) on one side and calcite (c) on the other. In the latter case the fibres at the upper surface of the layer of chry- sotile are broken and pressed obliquely out of position, thereby forming a thin lamina ; these fibres are more or less separated by films of calcite in- tegrally connected with the overlying calcitic layer. Thus the lamina corre- sponds to the " proper wall," and the latter part to the "intermediate skeleton." (N.B. The layers of chrysotile and serpentine are in such position to each other as to prove their genetic correlation : it is also noteworthy that on the margin where the chrysotile is in immediate contact with the serpentine there is no fibrous lamina with its fibres separated by calcitic interpolations.) Canada. L 124 DESCRIPTION OF THE PLATES. PLATE II. (continued). Fig. 3. Portion of a specimen of " Eozoon Canadense" showing a fissure or crack (several more are present in the specimen, see ' Proc. Roy. Irish Acad. vol. x. pi. xli. fig. 4) obliquely intersecting a lobulated layer of serpentine ("chamber-cast;" other layers of the kind are similarly intersected). This •fissure, as in other cases, is characterized by " intermediate skeleton " («), bounded at top and bottom by " proper wall " (d). A layer of chrysotile (c), in its incipient stage of development (see PI: IX. fig. 1, a), lies above and in parallelism with the fissure, proving the genetic correlation of the two parts. Observe that the fibres of the chrysotile in the fissure pass continuously lengthwise into the separated aciculee. PLATE in. Fig. 1. Decalcified specimen, magnified, of saccharoid marble (hemithrene), containing crystalloids of malacolite (a) decreted or etched into branching configurations (a*\ identical with "canal system" of "Eozoon;" also crystal- loids of pyrosclerite (6), some of which have their surfaces coated with aciculae. Mt. St. Philippe, near Marie aux Mines, Vosges. Figs. 2 & 3. Crystalloids, highly magnified, of pyrosclerite invested with a fibrous lamina (6), portions of which are pectinated (d), as in the variety of chrysotile corresponding with the " proper wall" of u Eozoon.1' Mt. St. Philippe, Vosges. Fig. 4. Section of a crystal of (P)peridote (polarized), associated with ser- pentine. " Eozoon Canadense," Canada. Fig. 5. Section of a crystal of (?) peridote (polarized), divided by laminae of calcite corresponding with the cleavage-divisions. The calcite is obviously a replacement product (pseudomorphism). Same as section fig. 4. Canada. PLATE IV. Ground-plan of a trap-dyke (a) intersecting gneiss (b) in a cove (A) at Glassillaun (B), on the north shore of Cleggan Bay, Connemara. The gneiss is converted into hemithrene (c) in places on both sides of, and adjacent to, the dyke. The dyke is calcitized. PLATE V. Natural vertical section of the last case (letters the same), showing gneiss converted into hemithrene by action of trap-dyke. PLATE VI. Fig. 1. Vertical section (being face of a quarry at Mt. St. Philippe) of gneiss (red colour) intersected by irregular masses of hemithrene (green). DESCRIPTION OF THE PLATES. 125 Given as affording evidences that the gneiss has been methylosed into the hemithrene (pp. 61, 52). The specimens represented under figs. 1, 2, and 3, PL III., are from this quarry. Fig. 2. " Eozoon Canadense." A layer of chrysotile in its different stages of development : — a, incipient stage ; b, typical chrysotile ; c, closely acicular (" velvet pile ") condition of " proper wall." Canada. Fig. 3. " Eozoon Canadense.''1 " Proper wall " in two separate laminae, and in different stages of modification. One of the laminae is pectinated, and thus assumes the form typical of the u proper wall." Canada. PLATE VII. Specimen of a serpentine rock, pseudomorphic after tremolite (natural size). Cannaver, Lough Corrib, Connaught. PLATE VIII. Specimen of tremolite (natural size), for comparison with that represented in Plate VII., the crystalline structure of both being identical. St. Gothard. PLATE IX. (Frontispiece.) Fig. 1. Diagrammatic sketch, showing serpentine (a) changing into chry- sotile (b) 5 also the latter in its various modifications or stages of development. Under a the chrysotile is in the (Jirst) incipient stage, being striated ser- pentine (that is, the variety marked with separated thread-like lines or cuts) ; 5, (second) fibrous stage, which is typical chrysotile ; c, (third) close-acicular stage, in which the fibres are changed into definite aciculae ; d, (fourth) pecti- nated stage, in which the aciculse are separated by films of calcite — thus con- stituting the " proper wall " of " Eozoon Canadense^ Fig. 2. Portion, highly magnified, of an accredited specimen of " Eozoon Canadense" showing the different features of the presumed fossil originating from chemical and structural changes in the minerals entering into its composition (pp. 18, 19). The layer a consists of ordinary serpentine changed into the striated variety or incipient chrysotile (a) (other layers consist entirely of structureless serpentine). At b is chrysotile in its typical condition. This layer at top, under c, is changed into the close-acicular variety, and under d into the separate-acicular or pectinated variety, the aciculae in which are separated by calcite : it is thus converted into the tf proper wall " of " Eozoon Canadense" The aciculae in the last stage are considered to be " casts of tubuli," such as characterize the nummuline layer (" proper wall ") of certain recent foraminifers. The letter c denotes flocculite, a white (floc- culent) variety of serpentine, which usually occurs as layers, or clotules : it is often decreted or etched out by solvent action into arborescent configurations (c*), forming the "canal system " of " Eozoon." (N.B. The last'feature is not in its original place : it occurs in another part of the specimen ; but its relations to other eozoonal features are correctly represented.) The layers d (" in- 126 DESCRIPTION OF THE PLATES. PLATE IX. (continued). termediate skeleton ") are in calcite, resulting directly from a chemical change in flocculite, also indirectly from serpentine : it forms the " inter- mediate skeleton" of " Eozoon." (N.B. The films of calcite between the aciculse of the pectinated variety of chrysotile are a similar pseudomorphic replacement.) Canada. Fig. 3. Portion (natural size) of a specimen of a lamellar graphic granite, consisting of layers of feldspar (? moonstone) definitely alternating in paral- lelism with others of quartz. The layers of feldspar are obliquely intersected by fibrous or striated laminae. Tarbert, Harris. — Specimens of the kind are considered by eozoonists to be of organic origin. The striae (f< striation ") were taken by Dr. Carpenter to be a " vertical tubular structure," corresponding with that of the u proper wall " of " Eozoon Canadense " ! Pl.l. ?AUGITE. (Transverse section) Changed into PERIDOTE. Yesuvms Pinx .W.?C Hanhart luh 2 <$c 3 *Eozoon pi.ni Pinx .W.K. Hanhart lith , PI VI Ko.nu.-irt in GNEISS (RED) CHANGED INTO HEMITHRENE (GREEN) MONT S7 PHILIPPE.. 70S GES. Serpentine pseudomorphic after Tremolite. Lough Ccrnb, Galwa.y. PI. VII! TREMOLITE. ST GOTHARD Hanhart imp »r f r- r*V * f *' f „' ' * f*r c / r ""r r ' r' r r* T * »* * * » » B « ^ rr INDEX. A. Abysses of the ocean, deposits in, 75, 99, 115. Acadian rocks, 84, 88. Acclimatization of plants, 117. Achiardi, A. d', on argillaceous schist changed into serpentine, 44, 97. Agalmatolytes, 2, 43. Aker (Finland), hemithrenes &c. of, xvi, 23, 40, 57, 63, 87. Alabaster, a methylosed product, 43. Alberese (Cretaceous limestone) ser- pentinized by euphotide, 97. Alkaline salts in rocks, 30, 31. 1 action of, 35, 79. Alleghanies, terraces of the, 113. Allomorphs of serpentine, 6, 19. Allport, Samuel, on peridote &c., 34, 69, 70, 71. Alps, ranges of the, 111. , terraces of the, 113. Aluminate of magnesia. See Spinel. Aluminous serpentine minerals, 5. America, mountain-ranges of, 111, 113. , terraces of North, 113, 116. , volcanoes of, 117. Amianthus with fibres of calcite, 16. Amity (New Jersey), hemithrene of, xxiv, xxviii. Hi, 10, 23, 56, 57. Ammonites in metamorphics of Switzerland, 32. Ammonitidse in Austria and Northern India, 104. Andalusite ferruginated into stauro- lite, 23. Andes. See Patagonian. Anhydrous magnesio-siliceous • mi- nerals, 7. minerals of metamorphic rocks, 35. ' Annales des Mines,' 28, 35. Antarctic Circle intersected by the earth's volcanic girdle, 117. Anthophyllite, 7. Antigorite, ferruginous serpentine, 5. Aphrodite a siliceous serpentine, 5, 44. Apjohn, Dr. J.,on origin of dolomite, 91, 94, 95. Appalachian ridges, 111. Aquosity of rocks, 35. Aragonite(?) replacing chrysotile, 21. Archaean " crystalline limestones," 72-82. ophites. See Serpentine rocks. rocks, 29, 32, 33, 34, 56, 72-83, 86, 87, 88, 91, 96, 111. Archceocyatlius atlanticus, 85. 1 Archiv fur Mineralogie,' 32. Arctic regions under climatal ameni- ties, 116. intersected by the earth's volcanic girdle, 117. Ardtrea magnesian limestone, Per- mian, 102. Argile magne"sienne, 2, 29, 44. Argillaceous limestone converted into rottenstone, 43. Argillytes, 34, 43, 44, 86. Arkoses, 31. Asbestiform serpentine. See Chry- sotile. M 128 INDEX. Asbestos an allomorpli of horn- blende, 14. Asia, High, mountain-ranges of, 112. Asiatic Russia, climatal amenities of, 116. Atacama, meteoric amygdaloids of, 63. Athabasca river, terraces of, 113. Atlantic Ocean, 115. See Abysses. Auerbach, calcitic dyke of, 51. Augite, 11, 24, 26, 29, 37, 58, 63, 66, 67, 68, 70, 71, 72, 79, 8}, 89. Augitic rocks. See Dolerites. Austrian Alps, AmmonitidaB of, 104. Auvergne, peridote in lava of, 63. Axes of stratal disturbance in rela- tion to jointing &c., 109. Azoic rocks, 29. B. Babbage, Charles, on vertical move- ments of the earth's crust, 114. Bacculites associated with Tertiary plants in Rocky Mountains, 102. Baily, W. Hellier, opposes "JEozoon" xiii. Bala limestone, 84. Ballinahinch (Connemara) crystal- line limestone, 90. Baltimorite, 5, 27. Baretti, M., on the metamorphics of the Central Alps, 31. Barker, Arthur E., xxxi. See Max Schultze. Barna Oran (Connemara), 90. Bastite, 5, 6. Beaches, ancient sea-. See Terraces. Beaumont, Elie de, on metamorphics of France and Switzerland, 31. Beauty Hill (Scotland), lime and serpentine associated, 55. Behring Sea, its volcanoes in the line of the earth's great volcanic girdle, 117. Belemnites in metamorphics of the Mont-Cenis district, 32. Ben Nevis, terraces of, 113. Bigsby, Dr. John J., on Canadian crystalline limestones, 73. Biharite, 5. Biotite, 7. Bischof, Gustav, vii, 24, 25, 26, 31, 32, 36, 37, 40, 44, 47, 54, 55, 62, 64-66, 70, 91. Blum, vii, xiv, 24, 37, 45. Bonn Lough (Connemara), crystal- line limestone of, 90. Boltonite, 61, 63. Bonnard, M., on metamorphics of France, 31. Bonney, Rev. Prof., on serpentine, peridote, &c., xliv, 11, 37, 40. , accepts " Eozoon? xliv. Bowenite, 5. Brackish lakes, 103, 104. Brazil, sea-board of, 111. Breithaupt, M., on serpentine and peridote, 37, 62, 64. Bristol dolomitic conglomerate, 93, 121. British Association Reports, x, xiii, xxxii, lii, 6, 75. Isles, absence of deep-water or third stage of Triassic system in, 102. Broegger, M., on hornblende changed into calcite, 51. Brongniart, Alexandre, u amphibo- lique " and " pyroxenique " rocks, 47. Bronzite, 61. Brucite, 7. Bryozoans, Permian, 103. Buch, Von, 91. Bufaure (Tyrol), dolerite of, 37, 46. Bunsen, his names " hydatothermic " and t{ pyrocaustic," 35. Bunter sandstone, 102. Burbank, L. S., at first accepts and next opposes " JEozoon," xxiv. , eozoonal limestones of C helms- ford not true stratified deposits, xxix, xxxviii. Burgess (Canada), et JEozoon" from, xv, Ivii, 41. Burmah, mountain -ridges of, 112. Burren of Clare, terraces of, 113. Byers's Quarry (Durham), crystal- lized dolomite of, 94. C. Calcaire saccharoide. See Limestones. Calcareous drift of Galway, 93, 120. dykes, or calcitic vein-like masses, 51-53, 56, 62, 64, 78, 79, 81. fossils, rarity of, in Cambrian rocks, 85. matter held in chemical and mechanical suspension in water of Lough Corrib, 120. INDEX. 129 Calcareous rocks. See Limestones. Calci-feldspathic gneiss, 13. Calciferous rocks, 84, 88. Calci-granitoid gneiss, xlv. Calci-hornblendic gneiss, xiii, xlvi, 14, 72, 73. Calci-micaschists, 2. Calciphyres of Alex. Brongniart,47. Calcite. See Calcareous dykes and Hemithrenes. interlaminated with siliceous minerals, xiii, xlv, xlvi, 13, 14, 72, 73. pseudomorphic after serpentine and other minerals, xvi, Ivii, 21, 49-53, 57-59, 67, 81, 83. — , polarized, 17. Calcitization of granite &c. near Galway, 27, 41. Calumet (Canada), " Eozoon" from the, ix. Camarophoria Kingii, 103. — '• — multiplicata , 103. Cambrian crustaceans, 85. limestones, their rarity, 87. organisms, 85. — rocks, 83-89. Canadian Archseans. Se.e Archaean rocks. Geological Survey, 29, 73, 76. group (? Upper Cambrian) of rocks, 85. < Canadian Naturalist/ 29, 48. Canadian ophites, 45, 57, &c. "Canal system." See " Eozoon Canadense." Cannaver Isle (Lough Corrib), ser- pentine after tremolite, 39, 40, 125. Cape Colony, terraces of, 114. Cape St. Roque, 111. Cape-Verd Isles, peridote of, 61. Caradoc limestones, 84. Carbonate of magnesia in Permian dolomite, 93. Carbonic acid or carbacid solutions, their action upon silacid rocks and mineral silicates, 25, 26, 53, 79, 80. Carboniferous metamorphics, 32, 90, 94. system and its formations, 101, 102, 104, 108. Carmoney Hill (Antrim), doleritic dyke of, 71. Carpenter, Dr. W. B., his attack on the authors, xiii. , on " Eozoon Canadense" x, xi, xix, xxi, xxii, &c. Carpenter, Dr. W. B., versus H. J. Carter, xxx, xxxi. versus T. Mellard Reade, xxvii. versus Otto Hahn, xiii. ~ versus Mobius. See Corrigenda. , remarks on Max Schultze, xxxii. , on new Lurentian fossil in graphic granite, xl. , " Final Note on Eozoon Cana- dense" xxxiii. Carrarite (Carrara marble), 2, 32, 82, 90. Carter, H. J., opposes " Eozoon," xxx, xxxi, xxxvii. Caspian Sea, 92, 99. Cassiterite, pseudomorph after feld- spar, 23. Central Italy, metamorphosed rocks of, 32. Cerolite, 5. Ceylon, " Eozoon " in, xxix, 23, 57, 87. Chazy limestone, 85. Chelmsford (Massachusetts) crystal- line limestones, "Eozoon " &c. in, xxiv, xxviii, xxix, xxxviii. ' Chemical and Geological Essays ' (Dr. Hunt), 29, 32, 33, 35, 44. 1 Chemical and Physical Geology ' (Bischof ), 24, 25, 26, 32, 48. Chemical changes in minerals. See Pseudomorphism . - deposits, 29, 75, 120. Chippal (Vosges), hemithrene of. 51, 53. Chlorargillytes, 2, 43. Chlorite (see Corrigenda), 29, 50, 65. pseudomorphic after augite,ll. schist, 2, 29. Chondrodite, 7, 61, 72, 79. Chonicrite an alumino -calcareous serpentine, 5. Chrysolite, 61. Chrysotile an allomorph of serpen- tine, 8, 19. , changes of, into " proper wall " of "Eozoon Canadense,'' xiv, xxxiii, xxxviii, xlv, li, 8, 9, 14-19, 21,26, 125. from Reichenstein (Silesia), 15, 16, 123. — , Dr. Dawson on, xxxv, xxxvi, xxxvii, li. , M. Delesse on, 15. " Circle of fire," 117. Clare (Barren), terraces of, 113. M2 130 INDEX. Clay-slate, 2, 32. Cleavage, mineral, xv, xvi, 12, 57, 67-70, 108. , slaty, 107, 109,. 110, 111. Cleggan dyke (Connemara), 48, 49; 50,51,53,78,81, 124. Climatal amenities, 116. — changes,great recurrent,115,116. Clyde district, peridotic trap rocks of the, 70. Coal a methylosed product, 43. Coast-lines. See Continents. Cocchi, M., on fossils of the crystal- line limestone of Carrara, 32. Coccolitic minerals and marbles, 7, 48, 55. See Crystalloids. Cocos Isles in relation to the earth's volcanic girdle, 117. Coalenterates of the Cambrian system, 85. Colafirth chrysotile, 8. Colloidal serpentine, 6. Configurations, mineral. See " Eo- zoon" " canal system." Coniston limestones, 84. Connemara, " Eozoon " in ophites of, xi, xxi, xxv, 87. ophites. See Ophites. Continents, their coast-lines and tri- angular form, 110, 111. Conzocoli (Tyrol), junction of diorite and dolomitic limestone, 41. Coquand, Henri, fossils in crystalline limestone of Carrara, 32. Coralloids of magnesian limestone of Sunderland, xviii, xix, Ivi, 93, 94, 105, 106, 118, 119. Cordier, M., 13. Cork County, iointed limestone of, 108. Cornwall, serpentinyte of . See Lizard. Corrosion of minerals. See Crystal- loids and "Eozoon" "canal system." Corundophyllite, 5. Cotta, Bernhard von, on pseudomor- phism, 37. — — , on crystalline limestone, 89. Coulonge river (Ottawa), Archaean conglomerates of, 77, 87. Credner, II., advocates " Eozoon" xxii. Cretaceous system, pelagic formations of, 102. Croix-aux-Mines (Vosges), metamor- phics of, 51. Crouza stone, 21. Crustaceans, Cambrian, 85. Crystalline limestones. See Lime- stones and Hemithrenes. Crystalloids, xv-xvii, xx, xxvi, xxviii,' Iv, Ivii, 21, 22. 41, 50, 55-59, 79, 83, 84, 89, 124. Cumberland, Permians of, 104. Cunninghaine on serpentine, 37. Cyclical (regional) vertical move- ments of the earth's crust, 100- 102, 104, 113-115. D. Dakyns, J. E,., on terraces on the Dovrefjeld, 113. Dalmein (Scotland), calcareous mar- ble of, 89. Damon, Robert, specimens of (( Eo- zoon " received from, xv, xxxvii, 18. Damour, M. A., on " tremolite " &c., 40, 61. Damourite an alkaliferous mineral ,31. Dana, Prof. James, on the West- ch ester crystalline limestones, 33. 42,59. , on the earth's continents, 110, 115. , on loganite, xv. D'Argenville, priority of his name peridote, 61. Darwin, Charles, on terraces of Pata- gonia, 113. Daubre"e, Prof. A., on rock-iointing, 107, 108. , on action of alkaline waters, 35. < Dawn of Life ' (Dr. J. W. Dawson), xxxviii. Dawson, Dr. J. W., on "Eozoon Ca- nadense" x, xi, xxi, xxiv, xxvii, xxviii, xxxiii, &c., 13, 18. , on "imitative forms of Eo- zoon" xlv. , on pseudomorphism, xxxv, xxxvi, xlix, 1, li, 19. , on chrysotile and serpentine, xxxv, xxxvi, xliii, li. versus Mobius, xlvii. versus Otto Hahn, xliii. versus T. Mellard Reade, xxvii. versus Carter, xxxvii, xxxviii. versus Roemer, liii. Decalcification, its effect on a dolo- mitic limestone, 94, 95. De la Beche on ancient crystalline rocks, 29. Delabole slates, their cleavage corre- INDEX, 131 spends with depositional partings 107. Delesse, M. Achille, on pseudomor- phism, 23. , on metamorphism, 28. , on change in chrysotile 15. , on slate changed into schistose serpentine, 44. , on hemithrenes of the Vosges. 52,53. Demagnesiation of Durham Permian dolomite, 94. Dendritic shapes assumed by minerals. See Flocculite, Crystalloids, &c. Derryclare Lough(Connemara), crys- talloids in limestone of, 40. Deweylite, 5, 6. Diaclasite, 7, 61. Diallage rock, 37, 38, 54. Diamond, 7. Dimetian rocks, impure limestone bands in, 83. Diopside, 35. Diorites and their methylosis, 2, 26, 27, 37, 38, 41, 48, 54, 64. Dolerites and their methylosis, 2, 37, 38, 48, 49, 51, 64, 71. Doleritic dyke at Cleggan. See Cleg- gan dyke. Dolomites and dolomitization, 2, 3, 32, 41, 43, 84, 91-95, 102-106, 118, 119, 121. Dolomitic conglomerate near Bristol, 93,121. Donegal, crystalline limestones of, 14, 90. Dovrefjeld, terraces of the, 113. Duncan, Dr. P. Martin, adopts t( Eozoonf xxii. Dunglow. See Donegal. Dunyte a peridotic rock, 2, 62. Durham coal-measures, their place in the Carboniferous system, 101. magnesian limestone, 91-95, 102-106, 118, 119. and marl slate, their place in the Permian system, 92, 102, 121. Durnes (Sutherlandshire) limestone, 33, 84, 89. Dwarfed organisms of the Permian period, 103. Dyke, doleritic. See Cleggan dyke. Dyke-like masses of crystalline lime- stone, xxviii, xxix, 51, 52, 53, 78-82. Dyke-like masses of granite. See Granitic veins. E. Earth-crust movements. See Cyclical and Stratal. Echinoderms, Cambrian, 85. ' Edinburgh New Philosophical Jour- nal,'25. Elevations and subsidences. See Cy- clical and Stratal. Elfdalen (Delecarlia), peridote of, 12, 62, 63. Emmons on " primary limestones " (hemithrenes), 53, 88. on crystalloids, 56. Enderby's Land in relation to the earth's volcanic girdle, 117. Engadine. See Kalkgebirg. Enstatite, 7, 11, 37, 61. Eocene deposits of Central Italy con- verted into ophites, 97. "Eozoic "rocks, 29,32. 11 Eozoon" preserved in loganite, x, xv, xxiv, Ivii, 41. "Eozoon Canadense," v, vi, vii, and Introduction generally, 87. , " acervuline " and " lami- nated" varieties, 10, 11, 13, 14. , " canal system," xviii, Iv, 10,19,20,22,57,84. , "chamber-casts," 56, 64,79. , " intermediate skeleton," 10,13,21,22. , "properwall,"9, 17,18,70. , summary of evidences and arguments against, liv, Iv, Ivi. Eozoonal features in chrysotile rock of Reichenstein (Silesia), 15, 16. in hemithrene of Akei',23, 87. Amity (New Jersey), xxiv, 57. Ceylon, xxix, 57. Chelmsford (Massa- chusetts), xxiv, &c. Connemara, xi, &c. Isle of Skye, xxviii &c. Mont St. Philippe (Vosges), xlvii, 22. — Switzerland, 31. present in intrusive ser- pentine rock of the Lizard, xliii. Epidote,26, 29, 51,65. Equatorial. See Jointing. Eribol, Loch, See Durnes limestone. 132 INDEX. Essex Co. (New York) limestone. See Westchester Co. Estuarine deposits, their position in a rock-system, 100-102, 114. Etheridge, Robt., on organic origin of graphic granite, xi. , accepts " Eozoon" See Corri- genda. Eulysyte (a peridotic rock) of Tuna- berg (Sweden), 62. Euphotide, 2, 38, 42, 54, 97. Europe under Pliocene climatal con- ditions, 115. Expailly (Auvergne), peridote of, 63. F. Fahlun (Sweden), primitive schists of, 32. Fallou, M., on serpentine, 37. Fassaite, its relation to augite and serpentine, 37. Faults in relation to systems of rock- jointing &c., 109-111. Favre, Alphonse, his discovery of " Eozoon" in the crystallines of the Jungfrau, 31. Fayalite a peridotic mineral, 62. Feldspars, 24, 26, 29, 31, 39, 49, 50, 52, 60, 62, 72, 75. interlamellated with calcite, 13. with quartz. See Granite, graphic. striping or striation of, xxxiii, xxxix, xl, xli, xlii, 12, 126. Feldsyte of Galway serpentinized, 41. Fenestella retifarmis, a deep-water organism of the Permian period, 103. Ferruginous serpentines, 5. Figures de corrosion. See "JSozoon" "canal system." Findelen glacier moraine, peridote and calcitic veins in blocks of, 62. Flocculite (flocculent serpentine), its figures de corrosion form the "'canal system " of "JSozoon," xviii, 6, 10, 19, 20, 23, 50, 59, 84. Fogo (Cape-Verd Isles), peridote of, 61. Formations, Cambrian, deficient in limestones, 83-88. , Permian, 99, 102. , r-, as divisions in a rock- system, 100, 101, 102. Forsterite a peridotic mineral, 61. Fossils of the dolomites of Tyrol, 32. of the (?) Permian and (?) Car- boniferous rocks of N. India, N. America, and Austrian Alps, 104. of the Permian and Carboni- ferous rocks, 103. Franz- Joseph Land, its climate under altered physical conditions, 116. Froisset, M., on metamorphic rocks of Europe, 31. Fuegia, its relation to the earth's volcanic girdle, 117. Fundamental rocks, origin of, 29. Funkite, 7. G. Gabbro of Portsoy, 38. Gages, Alphonse, on serpentine, 4, 6. Galway Co., rock-jointing of, 108. , glacial drift around, 93, 120. , igneous rocks near, serpenti- nized and calcitized, 27, 41. , crystallized limestones of West (Connemara), 90. , Queen's College, specimens de- posited in Geological Museum of, 18, 27, 94. Gangri range (Thibet), terraces of the, 114. Garnet, pseudomorphosed, 24. Gastaldi, Prof. Bartolomeo, on meta- morphics of Central Italy, 31, 32. Geikie, Prof. Archibald, on deposits dredged by the < Challenger,' 75. u Geikie, Prof.," stated to be in favour of the organic origin of the Harris graphic granite, xL Gemaingoutte (Vosges), hemithrene of, 51. Genoa. See Spezzia. Geography, physical, in relation to geological phenomena, 99-117. < Geological Magazine,' 23, 28, 57. Geological Survey of Canada, 29, 73, 76. time, 100, 117. phenomena in relation to sys- tems of rock-jointing, 107-117. Glassillaun. See Cleggan dyke. Glauconite, 6. Glenelg (Scotland), calcareous marble of, 89. Glen Tilt (Scotland), 89. Glinkite a peridotic mineral, 61. Gneisses, 2, 31, 51, 52, 60, 72, 73, 81. Goniatitidae of Northern India and the Austrian Alps, 104. INDEX. 133 Gothard, St., tremolite of, 40, 125. Grandjean, M., on dolomitization, 94. Granite, 34, 60, 62, 65. , graphic, the structure of that of Harris presumed by Eozoonists to be of organic origin, xxxix, xl, xli, xlii, 12, 13, 126. near Galway, serpentinized, 27, Granitic veins, 78, 81, 82. Graphite, 7, 52, 72. Griesbach, Rudolph, on terraces of Natal, 113. Grindelwald, gneisses of, 32. Grinnell Land, its flora and climate during the Miocene or Pliocene period, 116. Groppite a serpentine mineral, 5. G umbel, Dr. C. W., on crystalloids in hemithrenes, xx, 56. — , on u Eozoon" xx, xxii. Gypsum, its occurrence in the pri- mitive schists of Sweden, 32. H. Hahn, Otto, opposes " Eozoon? xxxviii, xliv. , < Die Urzelle,' xlvi. , on peridote, 63. Haidinger, M., on dolomite, 91. Hall, Prof. James, on serpentine lime- stones of Can ad a and Northern New York, xlii, xliii. -, on the age of the crystalline limestones of Westchester and Es- sex Cos., 33, 88. Halleflinta in hemithrene of the Vosges, 52. Hardman, E. T., on hullite, 71. , on dolomitization, 91, 94. Harkness, Prof. E., on serpentine limestones of Connemara, xx, 42. , opposes ll Eozoon f xiii, xx. , on iointing in Cork limestones, 108. Harlech grits, their equivalents in N. America, 84. Harris (Hebrides), marbles of, 58, 59. — . See Granite, graphic. Hastings (Canada) limestone zone, its thickness, 74. Haughton, Rev. Dr., on peridote in Mourne granite, 62. (in conjunction with Prof. E. Hull), report on Vesuvian lavas, 69. Hebrides. See Harris. Heddle, Dr. M. Forster, 13, 16, 22, 37, 38, 39, 48, 54, 55, 58, 59, 89. Hemithrenes, 2, 40, 47-59, 61, 63, 72-91, 96. Herculaneum, serpentine among the ruins of, 11. Hiasen(Norway), hornblende changed into calcite,51. Hicks, Dr. Henry, on impure lime- stone bands &c. of Porthlisky, 83. Himalayas, their relation to systems of rock-jointing, 111, 112. Hochstetter, Dr. F. von, advocates "Eozoon," xx. Hoffmann, M., on metamorphics of Carrara, 32. Hoffmann, Robert, advocates "Eo- zoon" xxii. Hornblende, 26, 41, 49-52, 65, 57, 60, 62, 66-72, 89. Hornblendic gneiss or schist of Cleg- gan, 49, 124. of Madawaska, Canada, 72. Horner, Leonard, on "granular lime- stone," 48. Hudlestone, W., on crystals of " altered enstatite " in serpentine, Cornwall, 11. Hull, Prof. E., on origin of Lauren- tian limestones, xxiii, xxv. , on crystalline limestones of Mayo and West Galway, 90. , on peridote, 69, 70, 71. Hullite of Carmoney Hill (Antrim), 71. Humbleton Hill fossiliferous lime- stone, its place in the Permian system, 102. Humite, its relation to peridote, 61. Hunt, Dr. T. Sterry, on origin of Archaean rocks, see Metamorphism, Pseudomorphism, vii, 23. , on crystalloids, 56, 79. , his " novel doctrine " in oppo- sition to metamorphism, vii, xiv, 1, li, 28-33, 36, 64, 74. , on dolomitization, 91. , on " Eozoon Canadense" ix, xi, xv-xviii, xxiv, xliv, Iviii, 10. , on origin of serpentine &c., vii, 29, 36. , on origin of limestone, xxv, xxvi, 75, 76, 78. , on granitic and calcareous " vein-rocks," 56, 64, 78-81. 134 INDEX. Huronian rocks, 74, 88. Hyalite in Roman masonry at Plom- bieres, 35. Hyalosiderite, its relation to peridote, 61. Hydration of minerals &c. See Sol- vents. Hydatothermic, Bunsen's name for hydrothermal, 35. Hydrocalcite, 7. Hydromagnesian silicates (serpen- tinous minerals, &c.), 4-7. Hydromagnesite, 7. Hydro-phlogopiteschists, 2. Hydrothermal action on rocks and minerals. See Solvents. Hydrous aluminate of magnesia (volknerite), 7. magnesia (brucite), 7. silo-magnesian marl. See Sepi- olyte. Hypersthene, 7, 62. Hypersthenyte of Elfdalen, 12, 62, 123. I. Idocrase, 7, 14, 51. Imitative shapes in minerals. See " Eozoon^ " canal system." in dolomite. See Co- ralloids. Imprunetta, ophite of, 46. India, Northern, fossils of, 104. Infra-Liassic fossils in metamorphics of Mont-Cenis district, 32. Intrusive or igneous serpentine rocks, xiv, xliv, 46. Ireland, Geological Survey of, 39, 48, 90. , west coast of, affected by Gulf- stream, 116. Irish Academy, Proceedings of Royal, 17, 20, 29, 32. dolomites, 94. Isle of Skye, Eozonal marble of, xvi, xxxviii, 11, 32, 41, 57, 87, 89, 96. Italy, Central, ophites of, 97. , Northern, vegetation of, 116. J. Jannettaz, M. Edouard, referred to, 13. Jardin des Plantes, specimens in Geological Museum of, 13. Jersey, diorite of, changed into cal- cite, 26, 60. , New. See Amity. Jervis, Chevalier, on serpentine rocks of Italy, 46. Johnston, Prof. James F. W., on mag- nesian limestones of Durham, 91. Jointing in rocks, 53, 80, 90, 94, 106- 112, 118, 119. Jones, Prof. T. Rupert (" T. R. J."), on " Eozoon" xi, xiv, xxii, xlii, xlvi. Judd, Prof. J. W., volcanic rocks of Scotland, 97. Jungfrau, occurrence of " Eozoon" near the, 31. Jurassic ophite of Skye, occurrence of " Eozoon " therein. See Isle of Skye. metamorphics, 31, 32, 87. Jurolla, vapours still disengaged from lava of, 65. K. Kalkgebirg (Todte Alps, Engadine), lamellar ophi-calcite of, 11. Kammererite a serpentinous mineral, 5. Kaolin, Prof. Heddle on, 39. Kara Sea, its genial temperature, 116. Karsten's 'Archiv fur Mineralogie,' 32. Karstenite pseudomorphosed into selenite, 23. Keilhau, M., on hemithrene of Scan- dinavia, 53. Kerguelen's Land, its relation to the earth's volcanic girdle, 117. Keuper of Cheshire, its rock-salt, 92. Kilkenny coal-beds, their place in the Carboniferous system, 101. Killas, serpentinized, 2, 43. Kinahan, G. H., on serpentine of Cannaver Isle, 39, 46. , opposes "Eozoon" xxvii. King, Dr. Clarence, on Wahsatch mountain-mass, 111. King, Prof. William. See Rowney. , on metasomatosis, 28. , on a silo-carbacid rock in Ceylon &c., xxix. , reply to Dr. W. B. Carpenter, xiii. , on Permian limestones of Dur- ham, 94, 118, 119. INDEX. 135 King, Prof. William, Xera and Tha- lassa in the Permian period, 99-1 21. Knocktopher sandstone, Carboni- ferous system, 101. Knop, M., on metasomatosis, 28. Kobell, M., on chrysotile. 8. Kuntze, O., on " Eozoon, xlvi. Kyschtimsk, peridolyte of, 62. L. Labradorite. See Feldspars. Lacustrine deposits, their place in a rock-system, 100, 101, 104. Ladak, elevated lakes of, 114. Lancashire "Bunter," its relation to the Permian system, 102. Land and sea features in relation to different stages of vertical move- ments of the earth's crust, 100. Larval evolution, 85. Lasaulx, Prof, von, on metasomato- sis, 28. Laurentian rocks. See Archaean rocks. Lava, 65, 66, 69. Laveline (Vosges), hemithrene of, 51. Leibnitz, on metamorphic rocks, 28. Leipervillite a peridotic mineral, 61. Leonhard, Von, on origin of hemi- threne masses, 53. Letterfrac (Connemara), crystalline limestone of, 90. Leuchtenbergite, 5. Leucite, its origin, 66. Levis dolomites and limestones of Canada, .48. Lewis (Hebrides), hemithrene of, 58. Lherzolite, 2, 37, 62. Liassic metamorphics, 32, 96. Liguria, ophites of, 97. Limbilite a peridotic mineral, 61. Limehillock (Scotland), serpentine &c. of, 55. Limestones, Carboniferous, of Ireland dolomitized, 94, 101. , Archaean, origin of. See Hunt, Hull, Ramsay, and Chap. xii. , " Eozonal." See Ophites. , magnesian. See Dolomites. , mineralized, 89, 90. , methylosed. See Methylosis. of mechanical origin, 120, 121. of chemical origin (their organic origin ' * a fallacy ") . See H unt, also 120, 121. Limestones rare in Cambrian system, 83-87. Lisoughter (Connemara) crystalline limestone, 42. Lizard (Cornwall), serpentine rocks of, xliii, 11, 21, 37. Llandeilo flags, impure limestones of, 84. Lochaber terraces, 113. Loch Eribol, subcrystalline lime- stones, 33. Logan, Sir W. E., on Laurentian rocks of Canada, ix, 45, 73, 74. , on "Eozoon," viii, ix, x, xiv, xxi. Loganite, a pseudomorph after horn- blende, xv, 6. . See " Eozoon " in loganite. ( London and Edin. Phil. Magazine/ xliii, 2, 11, 37, 43. Longinynd rocks non-calcareous. 75,83., Lough Corrib. See Cannaver. water contains lime in mechanical and chemical suspen- sion, 120, 121. Lower Cambrian limestones, 83-87. Silurian limestones, 33, 84, 88, 89. metamorphics, xlii, xliii, 33, 88. Lyell, Sir Charles, on metamor- phism, vi. , accepts " Eozoon" xxix. M. Macalister, Dr. A., accepts u Eozoon,'' xxix, xxx. MacCulloch on coccolite marble of Tiree, 48, 55, 56, 59. M'MuUen, J., early discoverer of " Eozoon," ix. Madawaska (Canada), crumpled layers of limestones at, 72, 73. Magnesian limestones. See Dolo- mites. of Durham, their con- figurations. See Coralloids. — - — silicate minerals, their origin, 24, 25, 26, 29. Magnesiated rocks, Prof. Heddle on, 39 j J. Arthur Phillips on, 43. Magneso-argillyte. See Sepiolyte. Malacolite, xv, xvi, 7, 21, 22, 23, 50, 57, 58, 60, 61, 83. 136 INDEX. Malacolite replaced by calcite, xvi, Ivii, 21, 50-53, 67, 58, 83. Malacolophytes, xvi, 2. Malbay nags, their place in the Car- boniferous system, 101. Manchester marls, their place in the Permian system, 92, 102. Marbles, calcareo - crystalline. See Hemithrenes. Marl-slate of Durham, its place in the Permian system, 92. Marmolite, 5, 6. Marsden limestone of the Permian system, 107. Maryland (U.S.), terraces of, 113. Massachusetts, hemithrene of, 63. Mather on origin of hemithrene masses, 53. Max Schultze on " JEozoon" xxix- xxxii. Meridional. See Jointing. Metam orphic rocks, methylosed. See Methylosis. , mineralized. See Meta- morphism. — of Canada, 72-83. Metamorphism, v-vii, 1, lii, 7, 28-59, 72-83, 91-95, 105, 106, 118, 119, 121. Metamorphosed " igneous " or irrup- .tive rocks, 39. Metasomatosis, 28. Metaxite, its structures and relation to serpentine, xviii, 6; 10. Meteoric peridote, 63. Methylosis, vi, xxix, xliii, 28-83, 91- 95, 105, 106, 118, 119, 121. Mettenbach (Switzerland), "Eozoon" of, 31, 32. Mettenberg (Switzerland), Ammo- nites in gneiss of, 32. Miask, peridote in metamoi-phics north of, 62. Mica-schists, Dr. Sterry Hunt on their origin, 29. Midderidge limestone, its place in the Permian system, 102. Miemite, the crystalline form of the rock dolomite, xv, 29. Milan Cathedral built of ophi-cal- cite, 3. Millstone-grit of the Carboniferous system, 101. Minerals that occur calcitized, 26, 51. Mineral carbonates, their part in hemithrenes, 29. cleavage. See Cleavage. " Mineral Resources of Italy " (Che- valier Jervis), 46. Mineralized limestones, 89, 90. metamorphics. See Metamor- phism. ' Mineralogical Magazine,' 8, 16, 48. 'Mineralogy of Scotland,' Heddle's, 38. Miocene climate and vegetation of the Arctic regions, 115, 116. Mobius, Prof. Karl, accepts "Eozoon" xliv. , rejects "JEozoon" xlv. , in reply to Dr. Dawson, xlvi. Modum (Norway), hemithrene of, 40, 63. Mollusks, Cambrian, little lime in their shells, 85. 1 Monograph of Permian Fossils of England,' 92, 99. Monradite a serpentinous mineral, 5. Mont-Cenis district, metamorphics of, 31, 32. Mont St. Philippe (Vosges), hemi- threne &c. of, xvi, xlvii, 22, 51, 52. 53, 60, 78, 87, 124. Monticellite a peridotic mineral, 61. Monzoni (Tyrol), serpentine pseudo- morphic after augite at, 36. Mourne Mountains, peridote in gra- nite of, 62. Movements, vertical, of the earth's crust, 100-102, 104, 113-115. Miiller, H., on the origin of serpen- tine, 37. , Prof., of Basle, referred to, 11. Murray on the 'Challenger' dredg- ings, 75. Muscovite, its relation to phlogopite, 60. N. Nantwich (Cheshire) rock-salt, its origin, 92. Natal (Africa), terraces of, 113. Natrolite, its occurrence in granite, 34. ' Nature,' xxvi, xxvii, xxxiii, xl, xlv, liii, 30, 75. Naumann, C. F., on pseudomorphism, vii, xi, xiv. , on serpentine, 37. , on rounded crystalloids of augite &c., 56. Neolite, 5, 29, 30. Neptunists, Sterry Hunt on, 30, Nevis, Ben, terraces of, 113. INDEX. 137 New Brunswick, metamorphics of, 88. New Jersey, "Eozoon " &c. in hemi- threne of. See Amity. New- York State, crystalline lime- stones of, 56, 59, 88. Nicholson, Prof. H. Alleyne, on or- ganic remains in Harris graphic granite, xxxix, xli. Nickeliferous serpentine (Nouemite), 4,5. North Pole, present vegetation 16° or 17° below, 116. , its proximity to the earth's volcanic girdle, 117. Northern Italy, niethylosed rocks of, 96. , relation of its present flora to that of Grinnell Land in the Miocene period, 116. Norway, raised shell-beaches of, 113. Nouemite, a nickeliferous serpentine, 4,5. 0. Odern (Vosges), homogeneous slate serpentized at, 44. Oligoclase. See Feldspars. Olivine. See Peridote. Olivinoid a peridotic mineral, 61. Olivinyte a peridotic rock, 71. Ontario, Eastern, " fundamental gneisso-syenite " of, 73, 77. Ophi-calcites, Ophi-dolomites, Ophi- euphotides, Ophi-magnesites. See next reference. Ophites (see Serpentine rocks) defined and classified, 1, 2. , their minerals, 1, 2, &c. , their structural characters, Introduction generally, 8-22, 56, 64,70,79,122-126. of Connemara, xi, xxi, 46. O'Reilly, Prof. J. P., on sepiolyte of Vallecas, Madrid, 29. t Organisms, Cambrian lime-elabora- ting, 85. Ossipyte a peridotic rock, 2, 71. Ottawa, Laurentian conglomerate of, 77, 87. Oughterard (Connemara), crystalline limestone near, 42. P. Pacific Ocean, deposits in abysses of, during the various geological periods, 115. Palaeozoic metamorphics, 31, 32, 33. 88, 89. rocks, fossils of earliest, 85. Pamir (High Asia), in relation to equatorial jointing, 112. Paradoxides, rarity of lime in the sheU of this trilobite, 85. Pargas (Lapland), heniithrene of, 40, 56. Paris basin, sepiolyte of, 44. Parker, Mr., accepts " Eozoon" xxii. Patagonia, terraces of, 113. Patagonian Andes, their relation to the earth's volcanic girdle, 117. Pebidian series, 83. Pelagic deposits, their place in a geological rock-system, 101, 102, 115. Penninite a serpentinous mineral, 5. Pennsylvania, terraces of, 113. Peridolytes (peridotic rocks), 2, 40, 62, 71. Peridote (olivine), 7, 17, 24, 25, 37, 40, 50, 61-71, 75, 123, 124. under the polariscope, 17, 50. 63, 64, 70, 71. Permian rocks, 91-95, 101-106, 118, fossils, 99, 103, 104. physical geography, 99-106. Perry, John B., opposes "Eozoon" xxviii, xxix. Philippe, Mont St. See Mont St. Philippe. Phillips, J. Arthur, on serpentinized killas, 2, 43. Phillips, Prof. John, on "Eozoon," xxvii. , on rock-jointing, 107. Phlogopite, 7?>24, 29. Phosphate of lime in Cambrian fossils, 85. Physical geography of the Permian period, 99-106. , its phenomena in relation to rock-jointing, 107-117. — geology, 107-117. Pic dyEridlitz (Pyrenees), calci-feld- spar rock of, 13. Picrosmine a serpentinous mineral. 5,6. Picryte a peridotiferous rock, 2, fc>2. Pisani on leipervillite, 61. Plagioclase, its presence in the * Challenger ' soundings, 75. 138 INDEX. Plagioclase, striping or striae of. See Feldspars. Plateaux, continental, in relation to the intervening oceanic depres- sions, 110, 112. Pliny, his name chrysolite, 61. Plombieres, action of alkaline water on Roman masonry at, 35. Plutonists, Sterry Hunt on, 30. Podermo (Tuscany), argillaceous schist changed into serpentine, 44,97. Polarization of serpentine, 50, 63. of peridote. See Peridote. Polinally (Scotland), " lime and ser- pentine beds " of, 55. Pompeii, its ruins contain fragments of ophi-calcite, 11. Pontefract sandstones of the Permian system, 102. Porphyritic serpentinyte of the Li- zard. See Lizard. rocks near Galway, serpenti- nized, 27, 41. Porthlisky ( Pembrokeshire) , ' l impure limestone beds " of, 83, 84. ( ), malacolite .replaced by calcite, xvi, 83. Portsoy, serpentine and other rocks of, 30, 54, 55. Post-Archaean ophites and henii- threnes, 87, 96. Potsdam rocks, rarity of non-crystal- line limestones in, 84, 85, 88. Prato (Central Italy), ophites of, 46. Pre-Cambrian metamorphics of East- ern North America (Hunt), 33. Predazzite (hydrous dolomite) of Conzocoli (Tyrol), 7, 41. Presidential Address of Leonard Homer, 48. of H. C. Sorby, 109. of Dr. A. Macalister, xxix, xxx. Pre-Silurian metamorphics according to Sterry Hunt, 33. "Primary limestone." See Hemi- threnes. " Primitive water " in granite and metamorphics, 34, 35. Prochlorite, 5. Productus horridiis, P. giganteus, P. ponderosus, their habitat, 103. " Protogsea" of Leibnitz, 28. Protogines, 2. Protozoans of the Cambrians, 85. Pseudo-diallage, 5. Pseudomorphic serpentine minerals, Pseudomorphism, vi, xiv, xxxv,xxxvi, xlix, 1, li, 11, 19, 23, 24, 57. Pseudophite, 5. Pusyrewski on "Eozoon" xx. Pyrallolite a serpentinous mineral, 5,6. Pyrocaustic rocks according to Bun- sen, 35. Pyrosclerite a serpentinous mineral, 5, 27, 52. in " Eozoon" xlvii, 124. Pyroxene, white. See Malacolite. " Pyroxenique '* rocks according to Alex. Brongniart, 47. Pyroxenites, their association with hemithrenes, 78, 81, 82. Q. Quarry-water (" eau de carriere), 35. 1 Quarterly Journal of the Geological Society,' 22, 28, 29, 35, 37, 42. Quartz, pseudomorphic and original, Quartzites &c. of Loch Eribol, 33. Quebec fault, its extent and direc- tion, 111. "Quebec rocks" in Canada and Northern New York, 48, 88. Queen's College, Galway, specimens in Geol. Museum of, 18, 27, 94. R. Ragged Chute (Madawaska, Canada), corrugated layers of gneiss and limestone, 72. Ramsay, Dr. A., on rock metamor- phism, v. on origin of Archaean lime- stones, xxh. on Permian formations, 91. 99, 101, 103-106, 119. adopts " Eozoon" xxii. Reade, T. Mellard, opposes "Eozoon" xxvi, xxvii. Regional cyclical vertical movements of the earth's crust. 100-102, 106, 113-115. metamorphism. See Metamor- phism. Reichenstein (Silesia), chrysotile of, 8, 15, 16, 123. Regard, Abbe, on the < Challenger ' soundings, 75. INDEX. 139 Kensselaerite a serpentine mineral, 5,6. forms an ophitic rock, 2. Retepora cellulosa simulated in Cor- nish serpentinyte, 11. Retinalite, 5. Rhodochrome, 5. Richardson, T., on dolomite, 91. Ripidolite, 5, 6. Roches homogenes et heterogenes of Alex. Brongniart, 47. Rock-jointing. See Jointing. Rogers, Professors, on decomposition of serpentine, 4, 25. Roman baths at Plombieres, minerals generated in brickwork of, 35. Rose, Gustaf, origin of serpentine, vii,37, 41. Rossie (St. of New York), crystalline limestone of, 56. Rottenstone, a product of rock-alter- ation, 43. Rowney, Dr. T. H. (in conjunction with Prof. W. King), xii, xiii, xxii, xxiii, xxviii, xxxiii, xxxix, xliii, xlvi. ( ), on regional metamor- phism, 34, 35. ( ), on dolomitization, 92, 93, 94. Rozel on hemithrene of the Vosges, 53. Rutley, Frank, on the serpentine of Cannaver Isle, 40. S. Sahlite, 7, 55, 57. St.-Bees' red sandstone, its place in the Permian system, 101. Sainte Marie-aux-Mines. See Mont St. Philippe. St. Paul's Islands, their relation to the earth's volcanic girdle, 117. Sandberger, F., on origin of serpen- tine rocks &c., 70. Sandford, W. A., on « Eozoon " in Connemara marble, xl. Sanidine, its occurrence in deep-sea deposits, 76. Saponite, xliii, 5, 11. Sapphire, its occurrence in hemi- threnes &c., 7. Scheerer on pseudomorphism, vii, 24. on aqueous origin of granite, 34. on hemithrenes, 48, 53. Schimper, Prof. W. P.. opposes "Zozoon," xlix. Schizodus-li-mestones, their position in the Permian system, 102. Scouler, Dr. R., on dolomite, 91. Sedgwick, Rev. Prof. Adam, on ser- pentine rock of the Lizard, 87. on coralloids of the Durham magnesian limestone, 93, 106. Sedimentary ophites, 45, 46. Selwyn, Alfred R. C., on Levis metamorphics, 48, 88. , on Huronian rocks of Canada, 74. — — , on thickness of the" Archaean rocks, 73. Sepiolyte (sepiolite, " argile mag- nesienne), 2, 5, 29, 44. Sequoias, fossil, of Grinnell Land and Spitzbergen, 116. Serapis, columns of verd antique in temple of, 11. Serpentine minerals, their pseudo- morphic origin &c.. xiv, 1, 4-7, 24, 50, 63, 64, 71. rocks (serpentinytes), their methylotic origin. See Methy- losis. generally. See Ophites. , kinds originally igneous, intrusive or eruptive, xiv, xliv, , kinds originally sedimen- tary, 45, 46. Serpentinized porphyry, granite, sy- enite, and other rocks near Gal- way, 27, 41. rocks. See Methylosis. Serpentinytes, xliii, 2, 11, 21, 37. Seyoertite, 5. Shallow-water deposits, 100, 101. Sharpe, Daniel, on slaty cleavage, 109. — , on terraces of the Alps, 113. Shetland, serpentines &c. of, 38, 48. , South, its relation to the earth's volcanic girdle, 117. Siberia, meteoric falls of, 63. , Boreal, under genial climatal conditions, 116. Sideromelane, its occurrence in deep- sea deposits, 75. Silacid or ordinary metamorphics, 2, 75, 76. and silo-carbacid ophites. See Serpentine rocks. 140 INDEX. Silicate of magnesia, anhydrous, 24, 25. , hydrous. See Serpentine minerals. Silurian period, calcareous deposits of, 86. Siphonotreta, spines of, containing glauconite, 6. Six, A., on " Eozoon" xlvi. Skye, Isle of. See Isle of Skye. Slaty cleavage, 107, 109. Sleat (in Skye), Post-Cambrian metamorphics of, 33. Smaragditje, 7. Smyth, Warington W., on "Eozoon" xx. Snarum (Norway), peridote of, 63, 65. Sollas, Prof. W. J., on Bristol dolo- mitic conglomerate, 121. Solvents or solutions, their dissol- ving or wasting action on crystals &c., 25, 26, 34, 35, 51, 53, 65, 79, 80. Solway red sandstones, their position in the Permian system, 102. Sorby, H. Clifton, on dolomitization of Permian magnesian limestone, 91. , on liquid cavities in metamor- phics, 34. , on slaty cleavage, 109. South Pole, its proximity to the earth's volcanic girdle, 117. Shetland in relation to the earth's volcanic girdle, 117. Spadaite a serpentine, 5, 44. ' Spanish Geology, Notes on,' 29. Spezzia (Central Italy), Alberese limestone near, ophitized by dykes of euphotide, 42. Sphene in hemithrene, 7, 55. Spinel in ophites &c., and its inti- mate association with eozooual structures, xxiv, xxv, 7. Spitzbergen under genial climatal conditions, 116. Staurolite, pseudomorphic after an- dalusite, 23. Steatite, 5, 55. Stevenson, Prof. J., on Ammonites &c. of the Rocky Mountains, 102. , on*" horizontal benches" of the Alleghanies, 113. Straits of Belle Isle, calcareous Aca- dians of, 84. Stratal disturbances of the earth's crust, 109-112. Strathdon (Scotland), marbles of, 89. Striping of feldspars. See Feldspars. Studer, Prof. Bernard, on metamor- phisrn &c., 31, 32. ' Study of Rocks,' Rutley's, 40. Stylolitic structure in rocks, 12. Sullivan, Dr. (Pres. Queen's Coll., Cork), on origin of sepiolyte of Vallecas, 29. Sulphuric acid, the part it plays in methylosis, 93. Summer-day of sunlight in Arctic regions, 116. Sunderlaud (Durham) coralloidal limestones, xviii, xix, 93, 102, 118, 119. Sutherlandshire, Lower Silurian me- tamorpbics of, 33. Swamp-cypress. See Taxodium. Swinaness (Unst), altered rocks of, 38. Symes, R. Glascott, on crystalline limestones of Mayo, 90. Synersk (Ural), peridote of, 62. Systemal periods, 100. T. Tabergite a serpentine, 5. Taimyr peninsula in relation to High Asia, 112. , its climate and vegetation, 116. Talc a hydrous mineral occurring in granite, 5, 34. Talc-gneisses or protogines, 2. Talc-schists, 2, 62. Tarbert (Harris) graphic granite, its presumed organic structure. See Granite, graphic. Taunus slate alkaliferous, 31. Taxodium (swamp-cypress) accli- mated to Arctic conditions, 116. Terraces of mountains, their relation to changes of ocean-level, 113, 114. " Thalassa and Xera in the Permian period," 99-106. Thermophyllite a serpentine, 5. Thibet, terraces of, 114. INDEX. 141 Thomson, James, his belief in organic structures in Tarbert graphic gra- nite, xxxix. Thuringerwald, alabaster of the, 43. Tiree, marble of, 48, 55-59. Titaniferous peridote of Zermatt, 61. Torrin (Isle of Skye), eozoonal ophite of, 41. Totaigite a calcareous serpentine, 5. Tourmaline, 7. Towanrieff hill (Scotland), serpen- tine of, 38. Trans. Cambridge Phil. Society, 37. Trans. Royal Soc. of Edinburgh, 38. Tremadoc rocks, scarcity of limestone in, 83. Tremolite of St. Gothard, 39, 125. Trenton metamorphics. See West- Chester Co. Triassic metamorphics, 31, 32. deposits formed in inland seas, 92, 93. Trinity Land, its relation to the earth's volcanic girdle, 117. Tschermak, G., on serpentine, 40. Tudor " Eozoon" xxiii. Tunaberg (Sweden), metamorphics of, 62. Tuscany, recent formation of serpen- tine in, 97. Tweedian beds, their position in the Carboniferous system, 101. Twining, Frederick, of Cleggan Tower, 48, 50. Tyrol, dolerite of Bufaure in the, 36, 37, 43, 91. , pseudomorphs in serpentine after augite at Monzoni in the, 36. U. Uddevalla (Sweden), peridolyte of, 62. Unkel, peridote of, 68. Unst (Shetland), serpentine rocks of, 38. Upper Cambrian metamorphics of Eastern North America, 33. Ural talcose rock, 62. Urkalk, 47, &c. See Corrigenda. V. Yaalite a serpentine, 5. Vallecas. See Sepiolyte. Vegetation of Arctic regions, 116, 117. Vein-masses of granite and limestone. See Dyke-like masses and Hunt on ditto. Venerite a serpentine, 5. Vennor, Henry G., on thickness of Archaean rocks, 73, 74. , his discovery of "Eozoon" on the Gatineau, xlvi. Vermont, North-western, calciferous formations of, 84. Vertical movements of the earth's crust. See Cyclical ditto. Vesuvius, peridots in lava of, 65, 69, 70, 71. Vigan (Dept. du Gard), calci-feld- spathic gneiss of, 13. Vilanova y Peira, Juan, opposes "Eozoon" xxxv. Villa Rota, its slate changed into schistose serpentine, 44. Villarsite a serpentine, 5. Virginia (U. S), terraces of, 113. Voigtland, diorite of, changed into serpentine, 37. Volcanic girdle of the earth, 117. Volcanoes of the western borders of the two Americas, 117. Volknerite, 7. Volterra, dykes of ophite at, 46. Von Buch, his theory of dolomiti- zation, 91. Von Leonhard on origin of hemithrene masses, 53. Von Morlet on dolomite, 91. Von (Eynhausen on calcite dyke near Auerbach, 51. Vorhauserite a serpentine, 5. Vosges. See Mont St. Philippe. W. Wahsatch, mountain-mass of, 111. Westchester Co. (New York) meta- morphics, their age and origin, 33, 59, 88, 89. Westoe ^Durham) Sigillaria-saiid- stone, its place in the Permian system, 102. Wick, M., on the age of the crys- tallines of the Central Alps, 81. Wilson, Dr. James, an early discoverer of " Eozoon" ix. Winter-night of darkness in the Arctic 142 INDEX. regions in relation to vegetable growth, 116. Wisembach (Vosges), rocks of, 51. Wollastonite occurs in altered rocks. 7, 61. X. Xera (and Thalassa) in the Permian period, 99-106. Xerothermal rocks, 35, 71. Y. Young, John (Hunterian Museum of Glasgow), referred to, 70. Z. Zermatt, peridote of, 61, 62. Zircon occurs in methylosed rocks, 7. Zirkel, Prof. F., on rounded crystal- loids in hemithrene, xxvi. , on "JEozoon," xxix. Zittel opposes "Eozoon" xlix. Printed by TAYLOR and FBANCIS, Red Lion Court, Fleet Street. RETURN EARTH SCIENCES LIBRARY 642-2997 LOAN PERIOD 1 1 MONTH 2 3 4 5 6 ALL BOOKS MAY BE RECALLED AFTER 7 DAYS Books needed for class reserve are subject to immediate recal DUE AS STAMPED BELOW FORM NO. DD8 UNIVERSITY OF CALIFORNIA, BERKELEY BERKELEY, CA 94720