On the southern slopes of Al Wein Range, about 16 miles south-south-east of Berbera, and of Ferrio Range about 35 miles south-east of Berbera, the Eocene limestone is overlain by a great thickness[9] of greyish-white massive gypsum beds. The latter lie conformably on the limestone and dip to the south at an angle of about 20°-25°. The Gypsum Series is seamed with numerous ravines, all with very precipitous sides, and, as exposed in these, it consists almost wholly of greyish gypsum indistinctly veined, with patches, streaks and brecciated blocks of anhydrite more or less altered to gypsum. In places, are veins of selenite and satinspar and thin partings of hard shale. In Ferrio Range, at the bottom of the southern slopes, yellow shales make their appearance at the base of the series, and these shales, besides being gypseous, contain in places masses of crystals of celestite. Indeed, celestite crystals can be obtained in every outcrop of the Gypsum series throughout the Guban.

The series also occurs on top of the main scarp south of Hais, the hills between the edge of the escarpment and Erigavo being all capped with greyish gypsum beds, similar to those in Al Wein Range, but by no means so thick. Moreover, the whole of the flat country in the vicinity of Dodab and Las Adey, and round Dabbar Dalol south of El Afweina is floored by gypsum, the surface of the Dabbar Dalol neighborhood consisting of large flat slabs of opaque white alabaster. Gypsum beds of considerable thickness also occur at Jid Ali and further east, and much of the country in the southeast corner of the Protectorate, e.g., round Kirrit, at Ain Abo, etc., is formed of the series.

In the hills north of Erigavo, the beds exposed are somewhat different from those of Al Wein and Ferrio Range. At the base are thin beds of gypsum; these are succeeded in turn by a thin stratum of white clay or marl with fossil bivalves, and a stratum of slaty nummulitic limestone, and on top is a compact somewhat fissile or slaty limestone. In one small hill, on top of the gypsum bed, is nummulitic limestone with oval and discoid flints.


It is evident that at the close of the Eocene period the sea which deposited the limestone retreated, leaving large inland seas over most of the eastern half of the Protectorate, which were liable to inundation and which, on finally drying up left behind great masses of gypsum and anhydrite with associated celestite. Whether, however, the gypsum of Al Wein and Ferrio Range is of the same age exactly as that of Kirrit, Ain Abo and the top of the main scarp at Jid Ali and elsewhere is not known, though some light may be thrown on the question by the determination of the fossil collections made by the writer.


On the southern slopes of Al Wein Range, about 16 miles southeast of Berbers, near the bottom of the outcrop of the gypsum beds, a thick series of sandstones and clays overlies the latter. The series extends for several miles along the slopes of the range, and from it southwards over the flat country to Dagaha Shabell and the east end of Bihendula Range. It lies apparently conformably on the Gypsum Series, the dip of both series in the vicinity of Khal Der ranging from 10°-20° to the south. As the beds stretch for about 10 miles southwards, as there are no signs of any strike faulting and as the dip of the beds gradually diminishes to about 5°, their greatest thickness is probably over 10,000 feet. The series consists of a lower division, exposed on the Al Wein slopes, made up of thick red, brownish-red and yellow sandstones and red, chocolate-colored and in places green clays; and an upper-division characterized by green clays and fewer reddish sandstones. About a mile north of Khal Der Well, and near the contact of the series with the gypsum beds, a seam of lignite is exposed, about one foot in thickness, and another outcrop, probably of the same seam, occurs about a mile further east. The occurrence of this lignite is described later on. The beds exposed in succession from the Gypsum Series southwards are:

  • Yellow sandstones with green and chocolate-colored shaly bands.
  • The lignite seam with associated chocolate-colored shales.
  • Thick brown and yellow sandstones.
  • White gypseous sandstones and clays, and yellow shaly sandstones.
  • Brick-red sandy clays, very gypseous.
  • Thick brick-red argillaceous sandstones.
  • Then follow the members of the upper-division :
  • Green, yellow and brown clays.
  • White clayey sandstones.
  • White sandy slates thinly laminated.
  • Pale yellow gritty sandstone with a seam of black flint one foot thick.
  • Pale purplish sandstone with a thin conglomerate on top.

Throughout the whole series, except in the lowest and in the non-argillaceous sandstones, gypsum is common. A little east of Khal Der, large selenite slabs occur on top of the yellow sandstones. South of the Well, and about 2,000 feet above the base of the series, are yellow and brownish argillaceous sandstones containing numerous fossils — banks of oysters of several species and other mollusca, etc., and at a horizon about one mile west of the Well, large shells of a Nautilus, and silicified wood. These clayey sandstones are overlain again to the south by yellow and white gypseous clays which contain numerous shells, the calcium carbonate of which has been completely replaced by gypsum.

The red, brown and yellow sandstones very closely resemble the Dubar Sandstone, even to the extent of containing similar mammillary forms.

From the nature of the sediments and the character of the fossils — oysters, cephalopods, silicified wood — the series was obviously formed in a large lagoon or inland sea, between which and the main sea, the connection was not entirely broken off, but into which enormous quantities of sediments were discharged. The age of the series will not be definitely known until the fossils collected from it have been determined, but it is probably Pliocene. The upper-division passes, to the south, under beds of pink gritty sandstone and massive conglomerates, but at Dagaha Shabell, green and red sandstones and chocolate-colored gypseous clays again make their appearance with a dip of 10-20° to the south-south-west, with the conglomerates faulted down against them. It is in these sandstones and clays that petroleum occurs, and considerable difference of opinion has arisen between Beeby-Thompson and Ball on the one hand, and Wyllie and Smellie on the other, as to whether they belong to the Dubar Sandstone Series or to the Deban Series. The subject is discussed and further description is given of the Dagaha Shabell outcrop later on, in the section dealing with Petroleum.


In the vicinity of the Daban coffee-shop, the Daban Series passes under a mass of conglomerates and very soft pink and yellow sandstones, which are well exposed in the Daban tug south of the coffee-shop. The conglomerates are composed chiefly of Eocene limestone pebbles, but pebbles of granitic and hornblendic gneiss are also present, especially in the neighborhood of El Jifli well. The pink sandstones forming the banks of the tug form rather a striking feature of the landscape as far south almost as the mouth of the Dagaha Shabell gorge. The thickness of the conglomerates and sandstones amounts to about 500 feet, and in places, the sandstones contain bands of conglomerate and vice versa. Just north of the Dagaha Shabell rock, they have been thrown down by a fault to the level of the Shabell beds. As far as could be seen in the Daban tug, they are conformable on the Daban Series, the dip of both formations where first seen in the tug being about 5° to the south. The conglomerates, however, gradually become horizontal and then, south of Dagaha Shabell, they dip slightly to the north. They are unconformable on the Daban Series of Dagaha Shabell at the east end of the Bihendula Jurassic limestone, and at Dagaha Shabell and elsewhere they appear to contain large solid masses of Eocene limestone. The origin of these limestone masses and their relation to the conglomerates is discussed later in the section on Petroleum.


The beds of this series occur at various places near the coast between Bulhar and Ankor, and there are small masses of limestone between Gedweda and Las Gori, and east of Las Gori as far as Elaiya, which may also belong to the series. The rocks are typically exhibited at Dubar about eight miles south of Berbera, and about 600 feet above sea-level, and at Dubino Hills some 15 miles southwest of Berbera and about 1,000 feet above sea-level. They consist of coarse conglomerates, sands and whitish and yellowish gypseous clays overlain by yellowish coral limestone, in which are numerous fossil corals of different kinds. Some fossil shells occur in the clays under the limestone a little southwest of Dubino Hills. The age of the beds cannot be determined accurately until the corals and shells collected have been examined, but they are certainly post-Eocene and post-Daban, as, at Dubar, they abut on a dissected Eocene-Dubar Sandstone scarp, and their contained fossils—especially the corals—are clearly of later age than the fossils of the Daban Series. The beds have been considerably faulted.


Lavas of several different ages occur in Somaliland. In the extreme west, on the track between Somadu and the Marmar Range, and about 3 1/4 miles southeast of Somadu, two large parallel escarpments striking north-west-south-east and dipping south-west at an angle of 30° were encountered. These escarpments consist of at least two flows of greyish-white spherulitic rhyolite, and as both the ridges are of the same constitution, strike and dip, they probably represent a formation that has been repeated by strike faulting. All the surrounding country consists of yellow sandstone capped by horizontal basalt, and, though nothing certain is known of the age of these rhyolites, they must be considerably older than the basalt. It is noteworthy that both the strike and dip of the series correspond very closely with the strike and dip of the Jurassic limestone scarps of the Gadabursi country.

Again, Mannar Range southeast of Somadu consists in part of extremely sheared greenish slaty rocks which, on microscopic examination, appear to be composed of a porphyritic matrix enclosing fragments of rocks of different compositions, andesite, slate, etc. Moreover in a gorge in the range about halfway along the track through it, is an outcrop of what appears at first sight to be a conglomerate. Closer examination, however, shows that this conglomerate also has an igneous porphyritic matrix, and that like the green slate facies it has been extremely sheared. Though further examination of the range is necessary before a definite conclusion can be come to, it would appear that these greenish slaty rocks are facies of an extremely sheared volcanic agglomerate. If so, the agglomerate is in all probability of great age, certainly older than Jurassic, since Jurassic limestones rest unconformably on parts of the range.

Another rock of distinctly clastic or agglomeratic character occurs in Bihen Gaha Pass at the north end of Jirba Range and close to the outcrop of Jurassic limestone and kerogenous shales west of the outcrop of slates. The microscopic sections of the rock, however, are not yet to hand, and it cannot at present be said whether it is of volcanic origin or not.

At the base of the Jurassic Series in Bihendula Range and resting directly on the gneiss occurs an appreciable thickness of partly decomposed basalt. According to Drs. Wyllie and Smellie, the basalt consists of at least two flows. It is therefore obviously either early Jurassic or pre-Jurassic. At Daba Dulla, however, in the Bur Ad Range, and at Hemal, south-east of Zeyla, the basalt at the base of the Jurassic is not a flow but a dyke that has considerably dislocated and baked the overlying shales. Further, in the extreme west of the Protectorate, west of Abaswein, Hensa and Somadu, all the surface is covered with a thickness of from 20 feet upwards of horizontal black basalt.

West of Berbera, and in the triangle Berbera-Bulhar-Hargeisa, large areas of the country, e.g., the Darnel Plateau, are covered by a series of basalt flows which have a maximum thickness of several hundreds of feet. These lavas lie unconformably, not only on the Eocene limestone, but even on the Coastal limestone. They are everywhere horizontal or nearly so, and the level of their surface is from 1,500-2,000 feet above sea level.

In the extreme east of the Protectorate, just south of Elaiya and about forty miles east of Las Gori, similar basalt lavas occur on the surface (apparently) of Eocene limestone.

The age of the basalts, excluding those near the French border, is at least later than Eocene and even later than that of the Coastal limestone, and it is clearly later than that of any of the main fault systems. The source of the lavas could not be determined, as, in spite of a search as opportunity offered, no necks or plugs were discovered.

Of the basaltic lava near Abaswein, Hensa and Somadu, all that can be said at present is that it is clearly younger than the great sandstone mass in the same district.

In Western Somaliland, especially in the country between Hensa and Somadu, there are numerous volcanic dykes of andesitic or variolitic character, which also intrude the sandstone mass.

There appear, therefore, to be in the Protectorate volcanic rocks of four distinct ages, viz. : (a) the sheared agglomerates of the Mannar Range and possibly the elastics of Jirba Range; ( b ) the spherulitic rhyolites between Somadu and the Marmar Range; (c) the basaltic lava at the base of the Jurassic Series; and (d) the recent basalt flows which form part of the surface of the country west-south-west of Elaiya, near the Italian frontier, and west-south-west of Berbera and south of Bulbar. At the present time, there is no evidence at hand to indicate whether or not the basaltic lava of the extreme west, at Abaswein, Hensa and north-west of Somadu, is of the same age as the lavas of (d).


These comprise the raised coral reefs along the coast; the alluvial gravels and sands of the tugs and those which occur on the plains between the tugs; and blown sand, which occurs in large quantities as yellow ridges north of Biyo Dader Well, and between it and the Dubar Sandstone plateau south of Karam, on the west of Ambal Ridge and on the coast in the vicinity of Karam, Ankor and Hais. Coral reefs and platforms exist east and west of Berbera, behind Ankor, east of Kelma and Mait, and near Elaiya. They are only a few feet above sealevel, but in places extend back into the Coastal Plain for some hundreds of yards. Near Wagderia Well, between Las Gori and Hais, large deposits of shells and corals occur in the banks of the tugs at some distance from the coast. These are now being examined.

It is worthy of mention that in the gravel banks of the Issutugan tug, which runs from near Hargeisa to Bulhar, stone implements have been obtained, which, there is the reason for believing, belong to more than one period. Some, four inches long and about three inches wide and of a round-oval shape, appear to have been used as missiles; others, about half as large and with a more pronounced point, may have been used for cutting or scraping; and others, again, about two inches long at the base and tapering to a fine point, were almost certainly used as spearheads. It is asserted that a collection of these implements was made years ago by Dr. Seton-Kerr, but the present writer has so far been unable to get access to any publication he may have written concerning them.

This site uses Akismet to reduce spam. Learn how your comment data is processed.