and M. Mourlon concludes that the Tangier promontory consists of Eocene beds resting on Cretaceous.

The eastern half of the northern promontory, including Tetuan and Apes’ Hill facing Gibraltar, consists of beds of a different character, for the most part of a hard metamorphic limestone, in which dip and strike are very obscure: these may be a southern extension of the Gibraltar limestone; but I had no opportunity of tracing the connection to Tetuan.

The late James Smith, of Jordan Hill (in ‘Journal of Geological Society,’ vol. ii. p. 41), mentions the occurrence of casts of Terebratula fimbriata and T. concinna, belonging to the Lower Oolite, in the Gibraltar limestone. M. Coquand also assigns to the Jurassic period the beds in the neighbourhood of Tetuan, and divides them into four stages, characterised respectively by marls, dolomites, a calcareous sandstone with the odour of petroleum, and a lithographic limestone containing siliceous concretions. I am of opinion that the Tetuan series, ranging with the Gibraltar limestone, and probably extending far to the south, is separated from the more recent Cretaceous series to the west and north-west by a great north and south fault, which divides nearly equally the Tangier promontory. M. Mourlon, referring to some specimens of shelly limestone in the Brussels Museum, collected near the river Mhellah in the district of Ouled Eissa, between Fez and Tetuan, resembling the Muschelkalk in aspect, and associated with beds resembling those at Tetuan, considers that they may also be of Jurassic age.

The Tetuan limestone has given rise to enormous beds of brecciated tufa, on terraces of which the city is built. The flow seems to have taken place from the hills to the north-west of the city, and has produced beds of a collective thickness of 60 or 70 feet. This is evidently true tufa, due to aqueous deposition, and is of a different character from the great calcareous sheet, to which I shall have occasion further to refer, which shrouds over the entire plain of Marocco.

Respecting the Mediterranean coast-line of Barbary, I will not add much to a paper read before the British Association at Liverpool, in which I remarked on the singular absence of coast-cliffs of any height. The undulating contour of the land-surface extends down to the water’s edge, a continuation of the form of the bottom of the straits without the intervention of cliff-escarpments, from which I surmised that the present sea-level and coast-line of the straits had not been of long duration.

Of frequent changes of level on the Barbary coast there is abundant evidence. The more recent seem to be, first, an elevation of from 60 to 70 feet along the entire coast, implied by the existence of concrete sand-cliffs with recent shells exactly similar to the raised beaches of Devon and Cornwall. These occur in Tangier Bay to a height of 40 feet, resting on the upturned edges of nearly vertical mesozoic beds; to the south of Cape Spartel, as a long cliff nearly 50 feet high; as low shoals near Casa Blanca; as a compact cliff about 50 feet high at Saffi, and as a coast-cliff and islands at Mogador, where the concrete sand-beds attain a height of 60 or 70 feet above the sea-level. It seems probable that this elevation of coast-line was coincident with a similar rise, implied by the existence of concrete sand-cliffs, all along the Spanish and Portuguese coasts, viz. on the eastern face of Gibraltar, where stratified raised beaches are seen cropping up at a considerable height from under the great mass of drift-sand in Catalan Bay; at Cadiz, as low cliffs 40 to 50 feet high, forming a hard coarse freestone of which the city is built; and also at the Rock of Lisbon, where, at a height of from 150 to 180 feet, isolated fragments of stratified concrete sandstone are seen clinging to the sea-escarpment of the older rocks.

The great range of latitude included in this simultaneous coast-rise suggests the probability that the elevation of similar coast-beds in Devon and Cornwall may pertain to the same movement.

Judging from the evidence afforded by the coast near Mogador, a subsequent submergence appears to be taking place. The island is probably diminishing in bulk; and, from observations made by M. Beaumier, the French Consul, it appears to have been reduced about one-fourth in area in twenty years; but whether from denudation or subsidence is not clear. The sea is, however, sensibly encroaching, as an old Portuguese fort and some Moorish buildings are now environed with sand and salt-marsh close to the sea, in a position where they would not have been built. This submergence of the coast at Mogador may perhaps be contemporaneous with the subsidence at Benghazi, Barbary, described by Mr. G. B. Stacey in the twenty-third volume of the ‘Quarterly Journal of the Geological Society.’ The general absence of cliffs characterises nearly the whole of the Barbary coast. A few low cliffs occur at scattered intervals west of Tangier; but from Cape Spartel to Cape Cantin a low monotonous coast shelves under the waters of the Atlantic, and not a cliff is to be seen, save an occasional raised beach. After rounding Cape Cantin the coast trends nearly north and south; and here the first good coast-section presents itself as a vertical cliff nearly 200 feet high (fig. 1), consisting of nearly level stratified alternations of grey and reddish marl, and fine-grained sandstone with beds of argillaceous carbonate of iron resembling the cement-stone of the Kimmeridge clay.

At a distance the cliff has a massive rocky aspect due to the vertical infiltration of tufaceous seams, which support the softer beds and stand out in prominent masses. The cliffs continue southwards to Saffi, where I obtained a small series of fossils from the section represented in fig. 1, amongst which Mr. Etheridge has determined Exogyra conica, Ostrea Leymerii, and 0. Boussingaulti. He considers the beds to be of Neocomian age. The hard band c is almost entirely made up of Exogyra conica.

I am indebted to the late Mr. Carstensen, H.B.M. Vice-consul at Mogador, for a specimen of Ostrea Leymerii, brought

to him by a Moor from Agadir, and obtained, at a height of 1,500 feet, on the flanks of the maritime termination of the Great Atlas range, 160 miles south of the Saffi section.

Two or three miles south of Saffi another section occurs, known as the ‘Jew’s Cliff;’ and from this Dr. Hooker, who landed on his homeward voyage, obtained a few fossils, viz. several undeterminable species of Pecten; an Ostrea allied to 0. Virleti, and a scutelliform Echinus of an unknown type, which Mr. Etheridge proposes to place under a new genus, and names Rotuloidea fimbriata. All these Mr. Etheridge supposes to be of Miocene age; and the ‘Jew’s Cliff’ section may probably give the key to the age of the beds of the Marocco plain in which we found no fossils. In connection with the occurrence of these Tertiary beds at Saffi, I must refer to MM. Desquin and Mourlon’s observations in the neighbourhood of Mazagan to the north-west, near which, at a place called Sidi Moussa, calcareous tufas associated with flints occur, containing

Solen, Venus, Modiola, Cardium, &c.; the deposit in its main characters resembling the description given by M. Coquand of the fluvio-marine travertines of the north of Marocco, and also the Sahara beds described by M. Ville; with the difference that the Sahara deposits are characterised by the presence of little Paludinas, whilst those of Sidi Moussa are full of vermiculiform perforations. The depressions are occupied by a very porous conglomerate, passing into a calcareous sandstone used for building. This conglomerate contains an abundance of Helix vermiculata, a species living in the country, and also found in the calcareous sands which are supposed to be of post-Pliocene age. The plain of Doukala (Ducaila of Washington), at a level of about 140 feet above the sea, is covered with these sands. At Sidi Ammer an escarpment was observed, the base of which consisted of clay and red ferruginous marls, containing a stratum formed for the most part of oysters, in which also Teredina personata occurred, supposed by M. Nyst to belong to the Eocene formation; succeeded by another fossiliferous bed containing

supposed by M. Nyst to be Miocene, the upper part of the escarpment resembling the beds of later age before described.

An examination of the higher points of the western coast near Saffi, and at Azfi in the province of Abda, near Mazagan, tended to establish the fact of the occurrence of Pliocene beds in the district.

At Cape Saffi, 180 metres in altitude, a reddish calcareous sand was met with abounding in Cyclostoma, Cylindrellas, and a species of Helix differing from that at Mazagan; and at other points, including the hill of Aher and at Sidi Bousid, white marls and sands associated with calcareous sandstones were met with analogous to the supposed quaternary beds in the neighbourhood of Mazagan.

The only other point in the geology of the coast-line I have to refer to is the great mass of blown sand surrounding Mogador, presenting a weird expanse of sea-like waves of sand, on a scale vastly greater than anything of the kind on our own coast, mimetic of mountain-chains and bold escarpments in miniature, differing only from true hill-and-valley structure in the absence of continuous valley-lines, the hollows being completely surrounded by higher ground. Many of the ranges of sand are from 80 to 100 feet in height, and their perfectly straight scarped faces are produced by the violent westerly gales blowing the sand up the angle of repose, and accumulating it in fountain-like showers over the rounded backs of the sand-hill ranges.

It is worthy of note that the sub-aërial ripple-markings superimposed on the greater undulations, occupy a reversed position with reference to the prevalent winds, their long side facing the wind, with the more vertical straight scarps on the lee side. The moving sand in this case is drifted up the long side, and falls over the scarp at the angle of repose.

The Plain of Marocco.—We now turn inland; and before referring to the details of the structure of the Great Atlas range, it will save repetition if I briefly describe the general contour of the district under consideration. Leaving the sand-hills, which die out inland, and travelling westward, we gradually ascend over an undulating country, in aspect somewhat like the Weald of Sussex, covered for 30 miles with Argan Forest, till we reach, at 60 miles inland, the average level of the plain, about 1,700 feet above the sea.

The fundamental rock is here rarely to be seen; for the entire face of the country is shrouded over by a sheet-like covering of tufaceous crust (fig. 3), rising over hill and valley, and following all the undulations of the ground. Only in river-beds and here and there by the side of a hill were the fundamental beds visible, and seen to consist of alternations of hard and soft cream-coloured calcareous strata, dipping and undulating in various directions at low angles, and so closely resembling the surface crust that it was difficult to distinguish the one from the other, unless the surface crust happened to lap unconformably over the scarped exposures of the stratified beds. This singular deposit varies in thickness from a few inches to two or three feet, and is taken advantage of by the Moors for the excavation of cellars in the soft ground, over which the crust forms a strong roof. These are termed matamoras, and are used for the storage of grain, and as receptacles for burying the refuse from the villages. The calcareous crust in the neighbourhood of Marocco is extensively burned for lime. In section it presents a banded agatescent structure, often much brecciated. It is impossible it can have been deposited by any waterflow, as completely isolated hills are shrouded over by it as thickly as the valley bottoms; and the only satisfactory explanation of its origin I can suggest is, that it results from the intense heat of the sun rapidly drawing up water charged with soluble carbonate of lime from the calcareous strata, and drying it layer by layer on the surface, till an accumulation several feet thick has been produced. The rapid alternations of heavy rains and scorching heat which take place in the Marocco plain are conditions favourable to this phenomenon, which is unknown in northern temperate climates.

A familiar illustration of the same kind of action is seen in what brickmakers term ‘limewash.’ A brick formed of marl containing soluble carbonate of lime, if rapidly dried or placed in the clamp in a wet state, will have on its upper surface, after burning, an unsightly white scum or crust, by the accretion of soluble matter driven upwards and outwards by the quick evaporation. Before we left Mogador on our journey inland, we were told of great beds of shingle covering the plain, and fully anticipated some interesting drift phenomena; but these shingle-beds were found to be nothing more than the broken débris of the surface tufa, covering the plain for hundreds of square miles with stony fragments. Of marine drift there is not a vestige, the few isolated patches of waterworn stones and alluvial shingle being always connected with river valleys, excepting only the huge boulder deposits of the Atlas hereafter to be referred to.

About midway between Mogador and the city of Marocco, the monotony of the plain is broken by a curious group of flat-topped hills, which rise two or three hundred feet above its

general surface. They present straight scarped sides, on which are exposed cream-coloured calcareous strata capped with a flat tabular layer of chalcedony, which seems, in arresting denudation, to have determined their peculiar and symmetrical form. In these we found no fossils; and I am doubtful whether they are an inland extension of the Miocene beds observed by Dr. Hooker at the ‘Jew’s Cliff,’ near Saffi, or are some members of the Cretaceous series, of which there are sections on the coast north of Saffi and on the flanks of the Atlas.

At this point the main boundaries of the plain come into full view,—on the north a rugged range of mountains trending east and west, which we estimated at from 2,000 to 3,000 feet in height; and on our right the great chain of the Atlas, rising 11,000 feet above us and between 12,000 and 13,000 feet above the sea, bounds the view to the south, framing-in the great plain, here some 50 miles broad, which is lost as a level horizon in the eastern distance.

The Atlas Range.—Commencing at Cape Guer, on the Atlantic sea-board, the range, which at a little distance has the aspect of a single ridge, averages at its western extremity from 4,000 to 5,000 feet in height, from which it slightly falls off in height for a few miles, and then gradually increases in height as it recedes from the coast. In the eastern part of the province of Haha the summits probably attain to a height of about 10,000 feet. At a point about 60 miles from the sea there is a comparatively deep breach in the range, through which runs the main road to Tarudant. Eastward of that pass the projecting summits appear to lie between 11,000 and 11,500 feet above the sea to a distance from the coast of about 100 miles, and about SW. of the city of Marocco, where a second depression occurs, affording a pass to the south, at an altitude of about 7,000 feet. Immediately east of this, and due south of the city of Marocco, the range for 30 miles in length presents a long unbroken ridge, 12,000 feet in height, on which are deposited a few isolated crags and peaks rising from 500 to 800 feet above the general level; and it is doubtful whether this part of the chain attains an extreme height of 13,000 feet. Still farther east the ridge-like character is lost, the range becoming broken up into a series of less continuous peaks (including Miltsin, estimated by Lieut. Washington to be 11,400 feet in altitude, and supposed by him to be the highest point in the chain) of diminished height: beyond this, eastward, little or nothing is known either of the altitude or character of the range, excepting that it trends NE. by E. towards the southern borders of Algeria on the Sahara.

Rohlfs, in his journal of his overland journey from Marocco to Tripoli, speaks of mountains to the east of Marocco being covered with perpetual snow; but this is a character which has been erroneously attributed to the Maroccan section of the Atlas range. When we arrived at Marocco in the first week of May, the snow was limited to steep gullies and drifts—all the exposed parts, including the very summit, being entirely bare. There were, however, frequent storms, which intermittently covered the range down to 7,000 or 8,000 feet; but it is certain that these occasional falls would be rapidly cleared off by the summer heat; and we came to the conclusion that there was nothing like perpetual snow on any portion of the chain we visited, included in the section (apparently the highest part) lying due south of the city of Marocco.

As seen from the city, the great ridge appears to rise abruptly from the plain some 25 miles off; and so deceptive is the distance, that it looks as though it were a direct ascent from the plain to the snow-capped summit, even too steep to scale; but in reality this wall-like ridge represents a horizontal distance of 15 miles or more from the foot to the summit. As we approached it, an irregular plateau four or five miles wide was seen to form a sort of foreground to the great mass of the chain, from 2,000 to 3,000 feet above the plain, and 4,000 to 5,000 feet above the sea-level. This is intersected by occasional narrow ravines, which wind up to the crest of the ridge; and its face, fronting the plain, is for the most part exposed as an escarpment of red sandstone and limestone beds dipping away from the plain, and again rising from a synclinal against the crystalline porphyrites of the centre of the ridge, and unconformably overlying nearly vertical grey shaly beds with a strike ranging with the general trend of the Atlas range. Against the plateau escarpment rest enormous mounds of boulders spreading down to the level plain.

These, then, are the general features of the chain of the Atlas and plain of Marocco, the further details of which it will be convenient to consider under the following heads:—

(a) Surface Deposits and Boulder-beds.—Next to the Tufa crust already described, which extends over almost the entire plain of Marocco, perhaps the most remarkable feature in the physical geology of the country is the enormous deposit of boulders that occurs in the lateral valleys, and flanks the great chain on its confines with the plain. Of marine drift there is not a trace; and alluvial drift and valley gravels are very limited in their distribution, being confined to the borders of a few insignificant rivers that intersect the plain and the localities of occasional waterflows; but as soon as the flanks of the Atlas are reached, new and distinct drift phenomena present themselves. It was on our second day’s journey from Marocco to the Atlas that the great boulder-beds came under our notice, first in a valley leading up from Mesfioua to Tasseremout, as scattered blocks of red sandstone, remarkable for their large average size, many of them of from ten to twenty cubic yards; but here the method of their disposition scarcely enabled us to decide that they were other than stream-borne masses from the higher ground. From Tasseremout we turned west, and at the

mouth of a second valley, two miles from the village, suddenly came upon a huge development of these Red Sandstone boulder-beds as great ridge-like and very symmetrical masses with terminal faces three or four hundred feet high, and, like the more scattered blocks NW. of Tasseremout, intermixed with but a very small proportion of fine matter. From this valley we turned out northwards, skirting the escarpment facing the plain; and for more than ten miles no lateral valley breaks into the cliff-like face; but below it the great boulder-beds (figs. 5, 6) still occur in huge masses not resting directly against the escarpment, but as isolated mounds two or three hundred feet in advance, sloping down towards the escarpment in one direction, and in the other rolling away in great wave-like ridges and undulating sheets, which terminate at a well-marked line of demarcation, just where the level portion of the plain commences. I measured by aneroid the height of these mounds; and at one point their summit was 3,950 feet above the sea-level, from which they spread down uninterruptedly to the edge of the plain nearly 2,000 feet below. They bear a striking resemblance to the glacial ridges or escars between Edinburgh and Perth; their mound-like structure is distinctly visible from the city of Marocco, twenty-five miles off, appearing like a row of pyramidal tali resting against the face of the escarpment as though they had been cast down from its edge on to the plain. The internal structure of the mounds also suggests such a deviation from the

disposition of the boulders in layers sloping away from the escarpment towards the plain; and on a nearer approach it is seen that the individual mounds are not connected with channels or valleys breaking through the escarpment.

The depression between the escarpment and the drift-mounds is a remarkable feature, and suggests an entire change of conditions since the boulder-beds were deposited. If they are a mere sub-aërial talus, they should rest directly against the cliff face, and the depression separating them must have been formed after the accumulation had ceased; and yet no satisfactory reason can be assigned for such cessation, if rain and river action were the only operating causes. The form of the mounds in the valley west of Tasseremout at once conveyed to me the impression that they were of glacial origin; and the discovery of undoubted moraines in the higher valleys strengthened my conviction that the boulder-mounds and ridges flanking the Atlas plateau can only be satisfactorily explained as the result of glaciers covering the escarpment, leaving on their recession the intermediate depression.

(b) Moraines of the Higher Atlas.—Kindred phenomena occur higher up in the Atlas valleys, most notable in the case of unquestionable moraines, commencing at the village of Adjersiman, in the province of Reraya, at an altitude of 6,000 feet. Here we met with a gigantic ridge of porphyry blocks, having a terminal angle of repose of between 800 and 900 feet in vertical height, and grouped with several other mounds and ridges of similar scale, all composed of great masses of rock with little or no admixture of small fragments, and completely damming up the steep ravine and retaining behind it a small alluvial plain 6,700 feet above the sea-level.

We failed to detect any scratched blocks or striæ; but that these ridges are true glacial moraines no one who has seen them and compared them with other glacial phenomena would for a moment doubt; and their interrupted occurrence at various heights is strictly in accordance with the distribution of moraines in many of the Swiss and Scotch valleys.

Lieut. Washington, in referring to the pointed mountainous hills NW. of the city of Marocco, crossed on his homeward journey, describes one of them as being ‘covered with masses of gneiss and coarse-grained granite (? diorite), many of the blocks being several tons in weight,’ and asks, ‘how got they there?’ ‘If granite, the nearest granite mountains are at a distance of twenty-five to thirty miles: can they be boulders?’ As far as my own observations go, there was no rock in situ in the part of this range I visited near Marocco resembling granite or diorite; and in connection with the boulder-mounds of the Atlas, the occurrence of foreign blocks north of the plain of Marocco so far from the parent source, is a circumstance of great interest.

(c) Stratified Red Sandstone and Limestone Series.—A long line of comparatively low and flattish hills, forming a plateau, with an average height of about 4,500 feet above the sea, and 2,800 feet above the plain of Marocco, intervenes between it and the main ridge of the Atlas. The edge of this plateau facing the plain is for some distance an escarpment, exposing stratified beds of limestone containing bands of chalcedonic concretions, underlain by grey and puce-coloured marls. As this plateau is crossed from north to south towards the Atlas ridge, its central line would represent a synclinal, from which the beds rise northwards towards the plain and southwards towards the Atlas; but it is locally broken and contorted, and near Tasseremout the limestone beds stand up nearly on end. South of the synclinal, i.e. between the centre of the irregular plateau and the Atlas, great deposits of red sandstone and dark-red conglomerate, interstratified with cream-coloured shelly limestone, occur, which appear to be inferior members of the series of limestones and marls exposed in the escarpment facing the plain. Lieut. Washington, who ascended Miltsin to a height of 6,400 feet, describes hard red sandstone with an east and west strike dipping 10° south, as occurring at this elevation, which is nearly 2,000 feet higher than we observed the Red Sandstone series in the province of Reraya farther west, and also both in his approach and descent from Miltsin of ranges of limestone running NE. and SW. dipping 70° SE. with abrupt sterile sandstone mountains rising above them. From the few obscure fossils, including an Ostrea, I was able to collect from the limestone bands, Mr. Etheridge considers that they are of Cretaceous age. They are, like the beds of the plain, remarkable for containing great deposits of chalcedonic concretions; but the latter may possibly be of more recent age. They rest unconformably on the upturned edges of grey shaly beds, and extend also over the porphyries that form the great mass of the Atlas chain. They appear to have been deposited subsequently to the porphyry ridge assuming its present hill-and-valley contour, as little isolated fragments are seen clinging to the sides of a narrow ravine leading out of the valley we ascended through the province of Reraya to the Atlas. Their relation to the few exposures of stratified beds in the plain is somewhat uncertain, as no fossils were obtained in the latter, and there are no direct connecting links; but, judging from petrological similarity, and from the fact that Neocomian fossils occur in exposed beds on the coast cliffs, and Cretaceous fossils in the beds forming the crest of the plateau, it seems possible that an unbroken series occurs from the cliff north of Saffi to the plateau skirting the Atlas, representing the whole of the Cretaceous epoch; but it is also open to question whether the level beds of the plain may not be an inland extension of the strata of Miocene age from which Dr. Hooker obtained fossils at the Jew’s Cliff south of Saffi.

(d) Grey Shales.—At several points on entering the lateral valleys of the Atlas, almost vertical shaly beds are crossed, having a strike nearly east and west, corresponding with the trend of the chain. They clearly underlie, and are unconformable to, the Red Sandstone and Limestone series; and their almost vertical position appears connected with one of the several upheavals that have affected the chain. Of their geological age there is no evidence, except that they are pre-Cretaceous. In places, as at Assghin, they abound in nodules of carbonate of iron. Pale shales, containing quartz veins, crop up near the village of Frouga, in the plain south-west of Marocco, which may possibly belong to this series; and if the porphyries forming the mass of the Atlas are contemporaneous, they are probably interbedded with these grey shaly beds. Lieut. Washington speaks of the occurrence of clay-slate dipping 45° east between El Mansoria and Fidallah, and again of a hilly country of clay-slate near the plain of Smira, and at Peira, farther south; but it is impossible to say whether these beds are related to the grey shales of the Atlas.

(e) Metamorphic Rocks.—The most important development of metamorphic rocks in the neighbourhood of Marocco is on the north side of the city. In its immediate neighbourhood, three miles to the north-west, a low rugged hill occurs, composed of a very hard and compact dark-grey rock, containing knotted white concretions elongated in the line of stratification, which dips from 50° to 80° south-west, the strike being north-west and south-east. The whole of the north side of the plain is bounded by ranges of rugged hills of similar form, and apparently rising from 2,000 to 3,000 feet above the plain. We had not an opportunity of visiting them; but, judging from their outline, they are identical in formation with the hill close to Marocco. We observed nothing in the Atlas resembling it. Lieut. Washington, who crossed these hills on his journey to Marocco at about the point I visited, and again forty miles to the east, near the source of the river Tensift, on his homeward journey, speaks of them as from 500 to 1,200 feet in height, consisting of micaceous schist and a schistose rock with veins of quartz dipping 75°, with a strike north by east and south by west. The strike may vary a little at different points, and, taking Lieut. Washington’s and my own observations together, would average about north and south; and it is worthy of note that these apparently ancient rocks are nearly at right angles to the strike of the rocks of the Atlas chain a few miles to the south.

The only other metamorphic rocks that came under our notice were:—first, white marble or metamorphic limestone, intercalated with the porphyrites at the summit of the ridge of the Atlas south of Arround; secondly, mica-schists, pierced by red porphyry dykes, forming the mass of Djebel Tezah, a peak 11,000 feet in height, and fifteen miles farther west, ascended by Dr. Hooker and Mr. Ball after my return. It is possible that the mica-schists may be a portion of the grey-shale series, metamorphosed by the intrusion of the porphyry dykes. Lieut. Washington, on his first day’s journey south of Tangier, refers to the occurrence of rounded schistose hills about 300 feet high, strike north-west and south-east, dip 75° south-west, containing mica-slate with veins of foliated quartz; but I have no recollection of observing any such metamorphic rocks between Tangier and Tetuan.

(f) Porphyrites.—Of the eruptive rocks of the Atlas, porphyrites and porphyritic tuffs occupy by far the most prominent position, forming the great mass of its ridge.

On entering the lateral valleys, after crossing the vertical shaly beds, great masses of red porphyrites and tuffs are met with, associated with specular iron and occasional green porphyries. The harder portions of the latter are seen as Verde antique pebbles in the river-beds; but we failed to detect this in situ. From the large proportion of tuffs that occurs the porphyrites appear to be interbedded, and are possibly contemporaneous with the vertical grey shales to which they are adjacent. They are overlapped unconformably by the Red Sandstone and Limestone series of Cretaceous age. The late Mr. D. Forbes informed me that they bear a strong likeness to the porphyrites of the Andes, of Oolitic age; but beyond the fact that they were in existence and had undergone denudation into hill-and-valley contour before the Cretaceous beds were deposited over them, there is no certain evidence as to their age.

There may have been at least one or two subsequent intrusions of red porphyrites, viz. of the dykes of Djebel Tezah, metamorphosing grey shales into mica-schists, and of the dykes that break up through the stratified beds of the plain east of Sheshaoua—which may probably be more recent than the porphyrites of the Atlas, as they appear to penetrate strata which extend over the denuded surface of the Atlas mass; but I cannot speak with certainty as to the relative age of the stratified beds and the porphyritic bosses which rise up out of the plain.

(g) Eruptive Basalts.—Of these we met with three distinct species:—

(1) Black vesicular basalt (porous and compact pyroxenic lava with olivine) on the coast near Mogador, and imbedded in the base of the post-Tertiary concrete sandstone cliffs: but it was nowhere seen in situ; and I think it possible that the fragments may have been derived from the Canary Islands, which are only 70 or 80 miles distant, or possibly from some point of eruption nearer the land.

(2) Amygdaloid green Basalt, which rises up in dykes, in many places penetrating the Red Sandstone and Limestone series on the flanks of the Atlas, and also piercing the diorite of the Arround valley. We observed numerous dykes at Tasseremout, Tassgirt, and Asni, south-east and south of Marocco city. Beyond the fact that they are probably post-Cretaceous, there is no evidence as to their age. From what we could see of their distribution, the whole range of the Atlas seems abundantly intersected by these dykes.

(3) Diorite rises up in considerable masses among the porphyrites in the valley of the Arround, due south of Marocco, but forms no great proportion of the bulk of the ridge. Its intrusion may have been contemporaneous with the dislocation and upturning of the Red Sandstone and Limestone series overlying the porphyrites.

General Summary.—It now only remains briefly to recapitulate the order of sequence of the geological phenomena observed in the plain of Marocco and the Atlas.

The oldest rocks that have been noticed are:—

(1) The ranges of rugged metamorphic rocks north of the city of Marocco, and forming the northern boundary of the plain, respecting the age of which, and the period of their upheaval and metamorphism, there is no evidence.

(2) The interbedded porphyrites and porphyritic tuffs of the Atlas, forming the backbone of the ridge, the age of which, and of the grey shales with which they seem to be interbedded, is also uncertain.

(3) Mica-schists of Djebel Tezah, in the Atlas, south-west of Marocco, pierced with eruptive porphyritic dykes, which may be an altered condition of the vertical grey shales adjacent to the interbedded porphyrites.

These rocks are our starting point, respecting which there is no evidence of their age, or even relative age.

(4) We now come to a long period of denudation of the Atlas ridge, and its sculpturing into hill-and-valley contour, before the deposition of the Red Sandstone and Limestone series.

(5) The deposition over what is now the Marocco plain, of the Cretaceous Red Sandstone and Limestone series (and beds possibly of Miocene age), which also occupies pre-existing valleys in the older porphyrites of the Atlas.

(6) The intrusion of diorite into the porphyrites and porphyritic tuffs, probably accompanied by a further elevation of the Atlas range, disturbing the stratified Red Sandstone and Limestone series, throwing them into a synclinal trough, from which the beds rise northwards towards the plain, and southwards towards the Atlas.

(7) A further long period of denudation of the Red Sandstone and Limestone series, rescooping out the lateral valleys of the Atlas, in continuation of the valleys that existed in the porphyrite ridge prior to their deposition, and also denuding the beds in the Marocco plain to the extent of at least 300 feet, leaving isolated remnants as flat tabular hills rising above the present general level of the plain.

(8) A further possible emission of red porphyrites through the stratified beds of the plain, which may have been contemporaneous with the eruption of the red porphyry dykes of Djebel Tezah, in the High Atlas; but I could not clearly ascertain whether these bosses really pierced the stratified beds, or were existing before their deposition.