The important districts of Baku occur on plains over anticlinals of Miocene beds. The petroleum-bearing sands are interstratified with impervious clays, separating the strata into distinct productive horizons.

In Algeria oil occurs in Lower Tertiary beds. The Egyptian petroleum comes from Miocene strata.

Petroleum seems to be unknown in peninsular India; but it occurs in many places along the flanks of the Himalayan range, and also in Lower Burma, generally in Lower Tertiary strata. In Upper Burma and Japan the oil-bearing rocks are probably Newer Tertiary. In all these areas the beds are greatly disturbed, and the same is the case with the great Carpathian field; but it frequently happens that the most productive regions are along anticlinal lines.

In New Zealand oil occurs in Cretaceous and Tertiary strata.

Petroleum and gas almost universally occur associated with brine. This may come wholly or partly from the decomposition of the animal matter which has produced the hydrocarbons, together with the remains of the sea-water originally present in the rocks. But the frequent occurrence of rock-salt in the neighbourhood of petroleum-bearing districts is worthy of note.

Summary.-The main points to be considered in respect to the geological conditions under which petroleum and gas occur in quantity seem to be as follows:— 1. They occur in rocks of all geological ages, from Silurian upwards. The most productive areas are Palæozoic in North America, Miocene in the Caucasus.

2. There is no relation to true volcanic action..

3. The most productive areas for oil in great quantity are where the strata are comparatively undisturbed. Oil, but in less abundance, frequently occurs when the strata are highly disturbed and contorted, but gas is rarely so found.

4. The main requisites for a productive oil- or gas-field are a porous reservoir (sandstone or limestone) and an impervious cover.

5. Both in comparatively undisturbed and in highly disturbed areas, an anticlinal structure often favours the accumulation of oil and gas in the domes of the arches.

6. Brine is an almost universal accompaniment of oil and gas.

3. The Origin of Petroleum. By O. C. D. Ross.

In the course of introductory remarks the author contends that, owing to the mystery surrounding the origin of petroleum, and to the paucity of indications where to seek for it, practical men in this country distrust the permanence of the supply, and hesitate to adopt it for many useful purposes; while the object of this paper is to suggest a way of resolving the mystery which is calculated to dissipate that distrust. The theories suggested by Reichenbach, Berthelot, Mendelejeff, Virlet, Verneuil, Peckham, and others, which are briefly described, make no attempt to account for the remarkable variety in its chemical composition, in its specific gravity, its boiling points, &c., and are all founded on some hypothetical process which differs from any with which we are acquainted; but modern geologists are agreed that (as a rule) the records of the earth's history should be read in accordance with those laws of Nature which continue in force at the present day.. E.g., the decomposition of fish would not now produce paraffin oil; hence we can hardly believe it possible thousands, or millions, of years ago, so long as it can be shown that any of the ordinary processes of Nature is calculated to produce it. The chief characteristics of petroleum strata are enumerated as: I. The

See the Chemical News for October 16, 1891.

existence of adjoining beds of limestone, gypsum, &c.; II. Volcanic action in close proximity; III. The presence of salt water in the wells; IV. The great extent of the production of oil, indicating subterranean receptacles of vast dimensions.

I. The close and invariable proximity of limestone to the wells has been noticed by all writers, but they have been most impressed by its being fossiliferous,' or shell limestone, and have drawn the erroneous inference that the animal matter once contained in those shells originated petroleum, but no fish oil ever contained paraffin. On the other hand, the fossil shells are carbonate of lime, and, as such, capable of producing petroleum under circumstances such as many limestone beds have been subjected to. All limestone rocks were formed under water, and are mainly composed of calcareous shells, corals, encrinites, and foraminifera, the latter similar to the foraminifera of Atlantic ooze' and of English chalk beds. Everywhere, under the microscope, its organic origin is conspicuous. Limestone is the most widely diffused of all rocks and contains 12 per cent. of carbon. Petroleum consists largely of carbon, and there is a far larger accumulation of carbon in the limestone rocks of the United Kingdom than in all the Coal-measures the world contains. A range of limestone rock 100 miles in length by 10 miles in width and 1,000 yards in depth would contain 743,000 million tons of carbon, or sufficient to provide carbon for 875,000 million tons of petroleum. Deposits of bituminous shale have also limestone close at hand; e.g., coral-rag underlies the Kimmeridge clay, which is more or less saturated throughout with petroleum, and it also underlies the famous Black-shale in Kentucky, which is extraordinarily

rich in oil.

II. The evidence of volcanic action in close proximity to petroleum strata is next dealt with, and extracts in proof thereof are given from several writers. In illustration of volcanic action on carbonate of lime, a sulphur mine in Spain, within a short distance of an extinct volcano (with which the author is well acquainted), is mentioned. That petroleum is not far off is indicated by a perpetual gas flame in a neighbouring chapel and other symptoms; and, these circumstances having attracted his attention, he observed that Dr. Christoph Bischof records in his writings that he had produced sulphur in his own laboratory by passing hot volcanic gases through chalk, which fact further led the author to remark that, in addition to sulphur, ethylene, and all its homologues (CnH2n), which are the oils predominating at Baku, would be produced by treating

2, 3, 4, 5......equiv. of limestone (carbonate of lime) with
2, 3, 4, 5......equiv. of sulphurous acid (SO2), and

4 6 8 10......equiv. of sulphuretted hydrogen (H2S);

and that marsh gas and its homologues, which are the oils predominating in Pennsylvania, would be produced by treating

These and all their homologues would be produced in nature by the action of volcanic gases on limestone.

But much the most abundant of the volcanic gases appears (at any rate at the surface) as steam, and petroleum appears to have been more usually produced without sulphurous acid and with part of the sulphuretted hydrogen H'S replaced by H2O (steam), or H2O2 (peroxide of hydrogen), which is the product that results

from the combination of sulphuretted hydrogen and sulphurous acid (H2S + SO2 - H202+2S). Thus

=

Four tables are given at the end of the paper, showing the formulæ for the homologues of ethylene and marsh gas resulting from the increase in regular gradation of the same constituents.

It is explained that these effects must have occurred, not at periods of acute volcanic eruptions, but in conditions which may be and have been observed at the present time wherever there are active solfataras, or fumaroles, at work. Descriptions of the action of solfataras by the late Sir Richard Burton and a British Consul in Iceland are quoted, also a paragraph from Lyell's 'Principles of Geology,' in which he says that the mud-volcanoes at Girgenti, in the Tertiary limestone formation, are known to have been casting out water, mixed with mud and bitumen, with the same activity for the last fifteen centuries.' Probably at all these solfataras, if the gases traverse limestone, fresh deposits of oil-bearing strata are accumulating; and how much may there not have been produced during fifteen centuries!

Gypsum may also be an indication of oil-bearing strata, for the substitution in limestone of sulphuric for carbonic acid can only be accounted for by the action of these sulphurous gases. The abundance of gypsum in the United Kingdom indicates that large volumes of petroleum are probably stored in places where it has never yet been sought for. Gypsum is found extensively in the petroleum districts of the United States, and it underlies the rock-salt beds of Middlesboro' (N.E. Yorkshire), where, on being pierced, it has given passage to oil-gas, which issues abundantly mixed with brine, and under great pressure from a great depth.

III. and IV.-Besides the space occupied by natural gas,' 17,000 million gallons of petroleum have been raised in America since 1860, and that quantity must have occupied 100,000,000 cubic yards; a space equal to a subterranean cavern 100 yards wide by twenty feet high and eighty-two miles in length, and it is suggested that beds of 'porous sandstone' could hardly find room for so much; while vast receptacles may exist, carved by water out of former beds of rock-salt adjoining the limestone.

This would account for the brine; and the increase to the molecular volume of the gases consequent thereon would in part account for the pressure. It is further suggested that when no such open spaces were available, the hydrocarbon vapours were absorbed into and condensed in contiguous clays and shales, and perhaps also in beds of coal, only partially consolidated at the time. There is an extensive bituminous limestone formation in Persia, containing 20 per cent. of bitumen; and the theory elaborated in the paper would account for bitumen and oil having been found in Canada and Tennessee imbedded in limestone, which fact Mr. Peckham (in his article on Petroleum in the Encyclopædia Brit.,' 9th edition) thought was a corroboration of his belief that some petroleums are a 'product of the decomposition of animal remains.'

Above all, this theory accounts for the many varieties in the chemical composition of paraffin oils, in accordance with ordinary operations of Nature during successive geological periods.

4. A Comparison between the Rocks of South Pembrokeshire and those of North Devon. By HENRY HICKS, M.D., F.R.S., Sec. Geol. Soc.

The clear succession from the Silurian rocks to the Carboniferous to be observed in many sections in South Pembrokeshire offers, in the author's opinion, the key to the true interpretation of the succession in the rocks of North Devon, for there

1891.

TT

cannot be a doubt that the post-Carboniferous earth-movements which so powerfully affected and folded the beds in North Devon extended into and produced almost identical results in South Pembrokeshire. In the latter area, however, the succession remains clearer, and can be traced more continuously.

The base of the Silurian (Upper Silurian of Survey) is exposed at many points, and the lower beds, usually conglomerates, repose transgressively on the Ordovician, and even on some pre-Cambrian rocks. Near Johnston and Stoney Slade the conglomerate contains numerous pebbles of the Johnston and Great Hill granite as well as of other igneous masses which were formerly supposed to be intrusive in these beds. From the Silurian conglomerate to the Carboniferous beds there does not appear to be any evidence of a very marked break in the series; moreover, all these beds were folded together and suffered equally by the movements which affected the area. The axes of the folds strike from about W.N.W. to E.S.E. The movements, therefore, at this time were in a nearly opposite direction to those which affected the Ordovician and Cambrian rocks at the close of the Ordovician period. Within the broken anticlinal folds portions of the old land surfaces have been exposed in several places by denudation.

The succession exposed in this area and the effects produced by the earth-movements so nearly resemble those already described by the author as occurring in North Devon, that he is convinced that the beds must have been deposited contemporaneously in one continuous subsiding area, and that the differences recognis able are chiefly in the basal beds, which were deposited on an uneven land surface. He believes that the Morte slates of North Devon are a portion of an old land surface on which the so-called Devonian rocks were deposited, and he also believes that the Devonian rocks are only the representatives in Devonshire of the Lower Carboniferous, Old Red Sandstone (and possibly of some of the Silurian rocks), of Pembrokeshire. A critical examination of the fossil evidence tends strongly to confirm this view.

5. Vulcanicity in Lower Devonian Rocks. The Prawle Problem. By W. A. E. USSHER, F.G.S.

[Communicated by permission of the Director-General of the Geological Survey.]

In the area extending south from the Middle Devonian volcanic series of Ashprington to the Prawle there appears to be no proof of the occurrence of strata older than Lower Devonian. There is no adequate reason for assuming that Lower Devonian rocks as old as the Gedinnian occur on the surface, and there is no certainty that the lowest beds are older than the Lower Coblenzian.

The occurrence of local volcanic action in Lower Devonian time is proved by a series of diabases and tuffs near Dartmouth, in the Kingswear Promontory, near Stoke Fleming, and in the line of country west from Torcross.

In association with the northern chloritic band (running from the mouth of the valley on the north of Hall Sands on the east to Hope on the west) we find volcanic materials identical in character with varieties of volcanic rocks associated with the Devonian slates in the line of country west from Torcross; and here and there in the line of country west from Torcross the volcanic rocks assume a more or less pronounced chloritic aspect. The junction of the slates on the north with the northern chloritic band is a strictly normal one, the chloritic rocks being almost invariably separated from the slates by brown volcanic materials which are everywhere succeeded by the same type of Devonian slate, and in the Southpool Creek and many other sections are found to pass insensibly into the chloritic type. In the Southpool Creek section a hard bluish diabase (? aphanite) occurs in the chloritic band. In the southern chloritic districts of the Prawle the volcanic rocks may still be here and there detected by texture or colour. Volcanic rocks occur in the mica schists of the Start coast, and can be detected even when only a few inches in thickness. At Spirit-of-the-Ocean Cove chloritic rock with much calcspar occurs in association with tuffs and a grey rock with incipient foliation, presenting a slightly gneissoid appearance, and apparently a much sheared diabase. The association of the chloritic rocks with the mica schists is of as intimate a

nature as that of the volcanic materials with the unaltered slates to the north. From these facts it seems evident that the chloritic series is nothing more than a Devonian volcanic group, of which the Torcross, Stoke Fleming, Dartmouth, and Kingswear coast tuffs and diabases were either sporadic offshoots or evidences of more or less contemporaneous local vulcanicity.

The more evident crinkling of the mica schists in contact with the chloritic group seems to be due to their comparative softness and greater fissility during the crumpling and contraction to which both were subjected.

The comparative suddenness of the transition from unaltered to more or less highly altered rocks may be explained by the lessening of strain (in receding from the harder masses of ancient rocks, against which the beds were jammed), being coincident with the thinning out of the volcanic materials northward, and furthermore favoured by the soft character of the grey slates with limonitic interfilmings which everywhere bound the northern chloritic band on the north. It is not the author's present purpose to enter more particularly into the stratigraphy of this interesting region, which is not yet thoroughly worked out. It only remains to acknowledge the prior claim of Mr. Somervail to the suggestion of the identity of the Devonian diabases with the chloritic rocks.1

6. On the Occurrence of Detrital Tourmaline in a Quartz-schist west of Start Point, South Devon. By A. R. HUNT, M.A., F.G.S.

While examining the Devonian cliffs near Street Gate, at the north-east end of Slapton Sands, South Devon, in company with Mr. W. A. E. Ussher, F.G.S., the author selected a hard micaceous sandstone of fine grain, occurring as a band between softer rocks, for comparison with a micaceous quartzite or quartz-schist, previously noticed by Mr. Ussher at a point on the coast south of Start Farm and west of Start Lighthouse. The quartz-schist occurs as an impersistent band among the mica-schists west of Start Point.

Mr. A. Harker, F.G.S., on examining the sandstone, at once pointed out the presence of tourmaline and white mica, of detrital origin; and considered that the rock had the appearance of having been derived from a tourmaline-bearing granite.

On a careful examination of two slides of the quartz-schist, the author detected a single grain of tourmaline. Six additional slides were forthwith prepared, and detrital tourmaline was found in them all. One of these slides contains a pellucid grain of quartz with fluid inclusions and active bubbles; another contains a grain crowded with hair-like inclusions and with one fluid inclusion whose bubble is easily moved by the heat of a wax match. Both these grains could be easily matched in the quartzes of different granites.

The derivation of the quartz-schist from granites of more than one character, but one of which must have been schorlaceous, seems clearly indicated.

The above facts have two distinct bearings, viz., as to the age of the metamorphic schists of South Devon, and as to the derivation of the tourmaline.

The two rocks under consideration, viz., the quartz-schist and the Devonian sandstone, are related to each other in four particulars, insomuch as they contain four constituents common to both, viz., detrital tourmaline, detrital mica, quartz of fine grain, and iron.

It seems difficult to avoid the conclusion that such similar rocks must be of like age and derivation; and that as the sandstone is undoubtedly Devonian, the quartz-schist, one of the metamorphic schists of South Devon, must be of Devonian age also, and not Archæan, as has been supposed by some geologists.

The derivation of the tourmaline is a more difficult question. Whatever may be the age of the mass of the Dartmoor granites, those of a schorlaceous character seem to be post-Carboniferous. Moreover, no tourmaline has been noticed in the

The views above expressed are those to which the author himself has been led, but they have not yet been fully considered and adopted by the Geological Survey. 2 The hand specimen selected for slicing was kindly placed at the author's disposal by Mr. A, Somervail, of Torquay.