337

THE GEOLOGY OF THE MONT CENIS TUNNEL.

BY PROFESSOR D. T. ANSTED, M.A., F.R.S., FOR. SEC. GEOL. Soc. WITH A GEOLOGICAL MAP AND SECTION.

[PLATE LXIV.]

THE approaching completion of the great tunnel through

the Alps, the most interesting and remarkable engineering work-certainly the most daring and speculative workof any that has yet been undertaken for railway purposes, renders an account of its progress and of the condition of the rocks pierced especially interesting. At the present time, more than half the total distance has been completed from the Italian and nearly half on the French, and the rock at both ends being now the same, the structure of the whole mass of mountain penetrated may be estimated with something approaching to certainty. A recent visit by the writer of this article has induced him to believe that an account of the geology of the tunnel rocks would be found to be of general interest. His object will be not so much to describe the numerous and ingenious inventions and mechanical appliances adopted for perforating the rock and ensuring ventilation for the great distance of nearly four miles from each end, as to point out the condition and nature of the rocks pierced in comparison with those seen on the surface. We thus get an insight into the real structure of the Alps at this point, illustrated by a perfect section between seven and eight miles long, obtained by the aid of a tunnel through highly metamorphic rock, the central part being more than 5,000 feet perpendicularly below the actual surface of the ground. Besides the geological structure, several important physical problems will be found to be involved and will receive some explanation.

The part of the Alpine mountain system through which this tunnel is being pierced, is the watershed that separates the waters of the Dora, one of the principal tributaries of the Upper Po, entering the Po at Turin, and those of the Arc, a

tributary of the Isère and ultimately of the Rhone. Crossing the Rhone at Culoz, the railway is carried along the expanded valley forming the lake of Bourget to Chambéry, and thence, in the same direction, without any works of magnitude, but through scenery of very great beauty, till it enters the main valley of the Isère. Proceeding up that valley for a short distance and crossing the river, it turns aside to enter the valley of the Arc, passing Aiguebelle and traversing a narrow strip of granite rock till it reaches St. Jean de Maurienne, a small town in Savoy, celebrated for its numerous and frequent earthquakes, and situated at the extremity of a projection of this granite rock to the east, penetrating the altered jurassic rocks which flank the granite both on the east and west. The rocks here consist of a large series of gypsums and clay, including deposits of iron ore and some veins of galena. A little beyond St. Jean a calcareous rock is crossed, believed to be the representative of the nummulitic series largely developed in southern Europe, and of an age recognised as lowest tertiary. After crossing a long strip of this limestone the jurassic rocks reappear, and beyond St. Michel they are found to be altered and greatly metamorphosed, putting on an aspect so generally characteristic of old rocks, that the geologist not well accustomed to the Alps and to Alpine geology would be easily deceived by them. A very extensive series of these metamorphosed oolites forms almost the whole mass of the two western divisions of the Alpine system, called the Cottian and the Graian Alps, separated by the valley of the Dora Riparia, and except where interfered with by some patches of serpentine and euphotide, they are remarkably connected and uniform throughout. The tract of country separating France from Italy in this part of Europe affords one of the lowest and most accessible of the numerous passes across the crest or culminating axis of the Alps. This occurs near the Mont Cénis, and over the col or pass known by that name is constructed one of the many magnificent roads for which Europe is indebted to the first Napoleon, who, imitating Hannibal in ancient times, had previously brought his army from France to Italy.

The Mont Cénis pass, though long a difficult and troublesome mountain path, seems to have been known and used from time immemorial. It has only within the present century become a great highway connecting north-western Europe with Italy, and is on the whole well adapted for that purpose, its highest point being only 6,890 feet above the sea. It may be crossed, with comparatively few exceptions, during the whole winter, and it is not so subject to destructive avalanches as many of the passes. The general width between the two principal valleys of the Arc and the Dora is here only twelve English miles, but the

length of road over the pass corresponding to this direct distance is nearly twenty-three miles. The valleys on each side are always fit for travelling, except immediately after serious floods, when the torrents coming down from the narrow gorges, chiefly on the French side, sometimes tear up and destroy the road. The Mont Cénis pass is, however, on the whole very approachable on both sides, and offers no particular or exceptional difficulties. The railway now in operation and intended ultimately to pass through the tunnel and connect with the line already open from Susa to Turin, terminates at St. Michel, a small and miserable village, beyond which to Modana, the entrance of the tunnel, is a distance of upwards of ten miles. From St. Michel to Susa the mountain rail known as the "Fell line" completes the communication, and is an admirable substitute for a railway of the ordinary kind; but it must be entirely superseded when the tunnel is once opened.

Geologically, the road beyond St. Michel towards Lanslebourg, the last village before the ascent of the mountain, crosses an important and extensive portion of the metamorphosed jurassic rocks of Switzerland, including (1) a series composed of quartzy conglomerate, quartzite, fine grained gritstone, and some argillaceous and calcareous schistose rocks with anthracite, corresponding as a series, but not in detail, with the middle member of the English oolitic series. These rocks are developed to a great thickness in the Graian Alps, and are identified in places by fossils; but they are very much changed and the stratification is rarely observable. Below this occurs (2) another large series of rocks, among which gypsum is very frequent and remarkable. The gypsum is associated with clays containing here and there organic remains of animals, and the whole appears to represent the lower oolite of England, including the great and inferior oolites and the liassic sands. Underneath this series we have (3) a very remarkable and persistent group of schists, steaschists, talcoseschists, and other varieties of schist, with numerous alternations of quartz and occasional masses of quartzite. In the small fissures and cavities of this rock are crystals of calcite, sometimes very beautiful and perfect. Much of the schist is very hard and compact, of a dark grey or blackish colour streaked by numerous quartzy threads, but the variety of appearance is very great. On the surface many of the schists weather into a paler grey and very fissile rock rapidly decomposing into a powdery mass, resembling at a distance a heap of loose sand. The kind of talus thus formed contrasts singularly by its smoothness with the abrupt and needle-shaped appearance of that part of the rock which has not yet been weathered, and that rises very grandly into mountain peaks.

The rocks thus described may be seen on the valley side

immediately on leaving St. Michel, where the mountain railroad comes into operation, and where the carriages of the French railway are exchanged for the very uncomfortable omnibuses of the American or Fell line. The map and section in the annexed Plate will show the order of the rocks and the way in which they succeed each other. The direction of the valley is at first nearly east, it turns after a certain distance to the south-east, and then towards the north-east, which direction it retains as far as Lanslebourg. The rocks have a general north-westerly dip, averaging about 50° in the tract of country between the Arc and Dora valleys which is the geographical axis of the Alps in this part. The line of railroad rises pretty steadily after leaving the valley of the Rhone at Culoz, at first slightly, afterwards, when the valley of the Isère is reached, more rapidly, the level of the valley bottom near St. Michel being nearly 3,500 feet above the sea.

The project of a railway tunnel to be cut through the crest of the Alps for the purpose of affording permanent and rapid intercourse between France and Italy at all seasons and without change of rail, was due to an Italian engineer, and has been carried on throughout under Italian superintendence. The first suggestion of such a tunnel is said to be due to Signor Medail of Bardonneche, who put it forward in a pamphlet published at Lyons in 1841. This pamphlet coming under the notice of the king (Charles Albert), the Minister of the Interior was ordered to make further investigations. The enquiry was referred to Professor Sismonda and a Belgian engineer (the Cav. Maus), who was at that time in Turin constructing the Turin and Genoa railway. After careful consideration it was agreed that the scheme was practicable, and that the position suggested was the best in that part of the Alps; but it was estimated that the tunnel could not be constructed in less than thirty-five years by the ordinary means then available, even making no allowance for drawbacks and accidents. M. Maus, noticing the large quantity of waterpower on both flanks of the mountain, conceived the possibility of utilising this, and invented machinery adapted for the purpose; but though a model was prepared the machine was not constructed. At a later period the present perforating machine was perfected under the superintendence of the engineers, Messrs. Grattoni, Grandis, and Sommeiller. It is quite distinct from that proposed by M. Maus, and being worked by compressed air (compressed by water-power to a pressure of between six and seven atmospheres), the air is available for ventilation. The exact site selected was so near the ordinary line of road crossing the Mont Cénis, that the tunnel has always been called by that name, but it really cuts through the Alps under the

Mont Fréjus at a distance of nearly sixteen English miles west of the former road. Starting from Modana on the Savoy side, about ten miles higher up the valley than St. Michel (the present terminus of the main line), advantage is taken of the widening of the valley, and the railway is conducted by a zigzag up the side of the mountain into a small ravine to a height of 3,946 feet above the sea, being a rise of 346 feet from St. Michel, showing an average rise of thirty-five feet in a mile, or one in about 150. From this point a direct line through the mountain, drawn at right angles to the axis of the chain at that part, carries us under the Mont Fréjus to an open part of the valley of Bardonneche, an important tributary of the Dora, emerging at a distance of 12,220 mètres (13,365 yards), or about 7 English miles. The direction of the tunnel is about NNW.-SSE. The railroad, after emerging from the tunnel, runs down the Bardonneche valley for another seven miles into the comparatively wide and open valley of the Dora, the ground being afterwards not particularly difficult for a railway, and much less liable than the upper part of the Arc valley to injury from floods and torrents. The road follows the course of the Dora without further divergence to Turin.

It will be evident to the reader, from this account of its position, that the tunnel has been driven through the crest of the Alps, across what would at one time have been regarded as being almost inevitably an axis of elevation, and it would have been presumed that in the course of the work, the primitive granite, considered not long ago to form the nucleus of all great mountain chains, must be crossed under circumstances involving perhaps extreme mechanical difficulties, large and incessant incursions of water, hard, tough, and contorted rocks, and all those indications of convulsion and disturbance inseparable from such conditions.

Fortunately the engineers who first projected this great work had not sufficient faith in the geological theories common then among the multitude, and not unsupported by very high authorities, to be frightened by the prospect of these probabilities. Perhaps they had not made acquaintance with them, and regarded the question as one of ordinary tunnelling through ordinary rock. At any rate they induced the Government of Sardinia (at that time possessed of both sides of the Alps) to listen seriously to their suggestions, and support the work, which was commenced about twelve years ago under auspices favourable in some degree, but with a prospect of a very long, if not endless, work. According to the methods at first adopted, the excavation of the tunnel, although commenced at both ends together, could only proceed very slowly at first, and was likely to be delayed when the great distance from the entry