The beginning, as given by Temple and Guérin, is DIVO OPTIMO . . . . . SEVERO. Bruce’s rendering, besides having been made at a much earlier date than the others, is also more probably correct,

Near this arch is another, but possessing no particular interest. The arch itself is entire, but no portion of the entablature remains in place. Its length, according to Guérin, is 10·90 mètres, breadth 2·75 mètres, opening of arch 5·45 mètres, and height about 10 mètres.[195] We were too busy with other important matters to have time to verify these measurements.

A third arch existed at the other side of the town facing the W.S.W. This is now almost entirely destroyed, but the dimensions are recorded by Sir Grenville Temple[196]—

From this there may still be observed a paved street running through the town.

The theatre has been a very remarkable building, constructed of immense and beautifully cut stones. Its form is indicated by the arches round the circumference which still exist; two parallel galleries surrounded the orchestra, and a considerable concentric portion of the proscenium and postscenium is still entire. We picked up a small fragment of sculpture amongst the ruins, representing a serpent coiled up in an attitude of repose, rudely but boldly executed.

Next in importance to the larger triumphal arch, is the Corinthian temple shown in both of Bruce’s drawings of the former building. He has left an excellent sketch of this, and two plates giving finished Indian-ink drawings of the decorative bands.

On the sketch itself he has added—somewhat unjustly, as it seems to me—a pencil remark, ‘Bad taste: it will do for a distance to arch; ornaments bad.’

The monument was more complete in his time than it is now; his drawing exhibits two entire sides of the cella, as far as the top of the pilasters, but no fragment either of the entablature or of the portico. At present the rear wall of the cella remains very much as it then was, but there is very little indeed existing of the other sides. On each side were four pilasters, the angular ones joining on each side; they were crowned with Corinthian capitals, very richly detailed, and having equally rich mouldings at their base; but the peculiar features of the building, which exist in no other one, as far as I am aware, in Africa, are the bands of sculpture occurring between the pilasters at about two-thirds of their height above the plinth. Bruce gives beautifully executed and highly finished drawings of the six which existed. They are all similar in composition, but differing in detail; they are bordered by a very elegant moulding, and contain the usual emblems of the sacerdotal office, such as the bucrane, or victim’s skull, from which garlands depend, supported in the centre by candelabra or vases; there are also introduced the knife, poleaxe, flagon for libations, the lituus, or augur’s wand, the flamen’s cap of office, the aspergillium for sprinkling lustral water, and several other emblems. Similar decoration is very frequent in the friezes of temples, but such sculptured bands on the walls are by no means common; the nearest approach to this feature, of which I am aware, is on the outside wall of the Pantheon at Rome, on each side of the principal entrance.

The other remains of Assuras are of less importance; they consist of several tombs and cisterns, private edifices and defensive works. Two bridges crossed the river, of which the upper was built of cut stone, that lower down the stream being of rubble masonry and brick, probably of a much later date.

The situation of this ancient city has been admirably chosen; it is built on a peninsula of land, surrounded on every side but the south by two water-courses, with deep and precipitous banks, which not only constitute a strong natural defence, but supply an abundance of fresh water.

In front of it stretches the plain of Es-Sers—no doubt, as Bruce remarks, a corruption of the ancient name Assuras. This basin, enclosed by hills on every side, contains about 50 square miles of rich, highly cultivated and irrigable land; but the plateau on which the city stood was cut off from the plain by the river, and was itself, or rather is now, perfectly barren.

The Khalifa of the district had sent his brother, Omar bin Amer el-Ayari, from a great distance to receive and entertain us. Without such hospitality it is extremely difficult for a traveller to encamp amongst ancient ruins in this country; they are usually shunned by the people, and no supplies are procurable within a radius of several miles. We also received a visit from Si Mohammed Esh-Shabi, Kaid of a fraction of the Drid tribe, who was encamped not far off; he was a really grand specimen of the Arab patriarch. He took quite an intelligent interest in the object of my journey, and was delighted with the extracts I translated to him from Bruce’s Diary. There is a considerable divergence between Bruce’s route and mine at this point; he came from Kef by Zouarin. I extract all that he says regarding his journey:—

A portion of this frieze is in the Kinnaird collection; the following is the inscription, as recorded by Bruce:—

Several other inscriptions were copied by Bruce. Some have since been published by Guérin and others. The following appears to have escaped the observation of subsequent travellers:—

Our friend Si Mohammed esh-Shabi confirmed in many curious respects the narrative of Bruce; he was particularly struck with the mention of the fact that this district was occupied by the Oulad Ali. His own tribe, that of the Drid, is one of makhsin, or hereditary soldiery. They wander all over the country with their flocks and camels, and as a rule possess no fixed residence and own no land in fee simple; his section of it however, that of Esh-Sabiah, purchased the territory which they now inhabit in the plain of Es-Sers, from the Oulad Ali, not very many years after Bruce’s visit, and the latter tribe has totally disappeared, at least from this part of the country. Then the ‘wood of firs,’ through which Bruce passed, was in the mountain of Bou Seliah; he was well aware that it had been covered at one time with Aleppo pines; but he assured me that not a single tree now remained. The Oulad Aoun still remain in their own frontiers, and as formerly, are exempt from taxation, like the Sidi Bou Ghanim, the Khomair, and several others, who are not required to pay taxes, simply because there is every reason to believe, that they would not consent to do so, and Government is not strong enough to enforce obedience.

CHAPTER XXVI.

After quitting the plateau of Zanfour we paid a visit to the tents of our friend Si Mohammed esh-Shabi. The good Sheikh was quite distressed that we would not spend a day at his encampment, or even alight to partake of his hospitality; but time was precious, so we contented ourselves with quaffing a lordly bowl of milk to his health, and he filled the haversacks of our escort with dates, bread and fresh cheese, the only luxuries of a portable nature that he possessed. He accompanied us several miles on our way, and we parted with, I hope, mutual feelings of goodwill and regret that our visit had been so short. His encampment did not materially differ from all other Arab villages, if so they may be called. It consisted of a circle of black tents, in the centre of which were the cattle, horses and mules of the family. The intervals between the tents were filled up with prickly brushwood; but the great protection of an Arab camp, which makes approach to it extremely dangerous to a stranger, is the cloud of yelping dogs which rush out from every direction, and can hardly be pacified by the utmost efforts of the owner. The tents are made of strips, sewn together, of black and brown woollen material like sacking, very open in texture, but perfectly impervious to the rain. The Arab knows nothing of the exigencies of fashion; he dresses as his forefathers have done for countless generations, and his tents are exactly the same as when they were described by Sallust as ‘oblonga, incurvis lateribus tectâ, quasi navium carinæ sunt.’[202]

We traversed the plain of Es-Sers from north-west to south-east, over soft springy meadow land, or amongst fields of corn, and crossing a low range of hills which bounds it, entered a valley scored in every direction with deep ravines, only just practicable for laden mules, and so descended into the plain of El-Gharfa. This is drained by a considerable river, the Oued Tessäa, which lower down becomes the Oued Khalad. It contains a small quantity of cultivated land, but much excellent pasturage. A considerable part of it is overgrown with a thick scrub of lentisk, once a favourite resort of brigands, but since the accession to power of the present Wuzir, General Kheir-ed-din, the roads in Tunis have become nearly as secure as those in Algeria.

The heat all day was insupportable; a strong sirocco, I will not say blew, but existed, for the worst siroccos are characterised by an utter absence of air in motion; the atmosphere is deprived of every particle of moisture, the sky is leaden, the mucous membranes of the body get parched, dry and incapable of relief by perspiration; everything one touches is hot and brittle, and the leaves of a book curl up and expand like those of a fan. It is no use to rest under the shade of a tree; the heat is not only in the direct rays of the sun, but everywhere; escape is hopeless and life is a burden.

About four miles before reaching Teboursouk we passed Ain Edjah, where is an interesting ruin, sometimes called Bordj Ibrahim, after a late Kaid of the Drid tribe, who built a stone house here, and surrounded it with a beautiful orchard of fruit trees.

The ruin was evidently a Roman fort, restored by the Byzantines, and converted into a fonduk, or wayside inn, by the Arabs. In some few places the original Roman foundations are still visible. Of the Byzantine restoration a great part of the wall of circumvallation remains; it was of a rectangular form, with square towers at the angles. In one of these the vaults which covered in the rooms on the ground floor and first storey still exist, as well as the stairs which conducted to the upper terrace. From the foundation of the eastern tower issues a beautiful spring, which waters the orchard below. This is the source of Edjah, which gives its name to the district. There is no doubt that this name is simply a corruption of the ancient one, Municipium Agbiensium, or Agbia, mentioned in the Tables of Peutinger. A short distance beyond, on the road to Teboursouk, we observed two milliary columns lying half buried in the earth, with their bases still in place close beside them. One was too deeply covered to permit us to see the inscription, and we had no instruments with which to dig it up. The other was only partly legible, the upper part of the inscription being much defaced by the action of weather. It has been recorded, both by Pelissier and Guerin.[203]

After a very fatiguing journey we reached Teboursouk, and were at once conducted to the house of the Khalifa. Our host was extremely courteous, but at first cool and reserved. He gave up one of the rooms in the ground-floor of his own house for our accommodation, and appropriated an empty building close by for the use of our attendants. He sent us our meals from his own kitchen, where certainly the art of cookery is thoroughly understood, and very soon, through the instrumentality of his son, a dear little fellow about four years old, to whom we presented a few trifles, and whose curiosity was insatiable, the good gentleman thawed considerably and our intercourse became more friendly and unrestrained.

The modern town is no exception to the general law, which seems to have doomed all Mohammedan cities to decay. Its situation is naturally most beautiful, being built on the slope of a hill which commands a valley of singular fertility, covered with groves of olives, and orchards of fruit trees; but the houses are half-ruinous, and the streets in a filthy and neglected condition.

Teboursouk, the ancient Thibursicum Bure, was so-called to distinguish the city of the pro-consular province from another of the same name in Numidia, Thibursicum Numidarum, the modern Khamisa.

Several bishops of this place are recorded, and one of them is mentioned by St. Augustine in his book Contra Cresconium.[204]

It can hardly be said to be now surrounded by a wall, though in some places the wall built on Roman or Byzantine foundations still exists. The most interesting part is towards the north, where one of the ancient city gates still remains entire, though built up and buried to a great part of its height in débris.

This is curious, as it differs from all the triumphal arches I have yet seen in North Africa. The piers on each side consist of two fluted Corinthian pilasters, supporting a complete entablature, from which the central arch rises. As nothing except the arch itself now remains, it is impossible to say how this was crowned. Amongst the stones used in the re-construction of this part of the walls are many fragments of sculpture and inscriptions. The latter have frequently been published. One of them contains the ancient name of the city, THIB. BVRE, another a record of the reconstruction of its walls by Thomas, prefect of Africa during the reign of Justinus II. and the Empress Sophia, A.D. 565 to 578.[205] This Thomas is honourably mentioned by Carippus Africanus in the first book of his poem, De laudibus Justini minoris:—

Three others have been recorded by Bruce, but the precise locality where they were found is not given.

The principal water-supply of Teboursouk is derived from a very fine spring, which issues from a small chamber and flows into a large square reservoir, both of Roman work. The overflow runs through a subterranean passage, and waters the gardens to the north of the town. On the lintel of the door of the chamber from which the water flows may be traced a few letters, which Guérin gives as VG. ARA. They are hardly legible now, but the entire inscription is recorded by Peyssonnel:—

On the summit of the hill above the town is a koubba, from which a very fine view is obtained.

From Teboursouk we made an excursion to Dougga, دُقّة, the ancient Thugga, Cives Thuggenses, or, as it is given in one inscription in the wall of a house near the temple, Respublica Coloniae Liciniae Septimiae Aureliae Alexandrinae Thuggensium.

This city must have been of very considerable consequence, to judge by the extent and magnificence of its remains, which cover an area of about three square miles; and when its temples and palaces were standing, and clear of the cactus and brambles that now invade their remains, it must have been a most striking object in the landscape. The city was built high up on the hill, which bounds on the west the extensive valley watered by the Oued Khallad, a tributary of the Medjerda. A wretched modern hamlet is built amongst its ruins, and the traveller has to wade through the accumulated filth of years to visit the various objects of interest which the village contains. This, however, occupies but a small extent of the ancient city; the remainder is overgrown with cactus and briars, or laid out in delightful groves of olive-trees.

The ancient name probably had the same signification as the modern word Dougga still bears in the Berber language, green grass; and, indeed, it would be difficult to find a more charming position, or one which, from its abundant water-supply, was more likely to be always verdant, than the hillside on which the city was built.

The most beautiful of all the ruins here—and, I am tempted to add, the most exquisite gem of art I have seen in North Africa—is the temple built from the private resources of two brothers, and dedicated, as will be seen, to Jupiter and Minerva. Bruce calls this

And again he says—

Bruce’s illustrations of this temple consist of nine sheets:—

1. A perspective view in pencil of the front of the portico, showing the door of the cella and the apse beyond.

2. Details and measurements in pencil of capital and base of column, and of entablature.

3. Rough pencil sketch of plan, with details in ink of various parts.

4. Perspective view in Indian ink of front; podium in pencil only. The triumphal arch of Septimius Severus is seen to the left.

5. Perspective view in Indian ink of the side of portico, showing the construction of the cella, or the building which has replaced it (Pl. XXII.)

6. Highly finished drawing in Indian ink, taken from an opposite point of view to the last, with landscape and figures of the Balugani type (Pl. XXIII.)

7. Finished Indian-ink plan and elevation, with dimensions.

8. Finished Indian-ink drawing of a capital, details of abacus and base.

9. Finished drawing in Indian ink of soffit of cornice, six of the fifteen flowers for centre of abacus, details of head of doorway of cella, and the same given in petto.

10. Finished drawing in Indian ink of details of pediment and podium of temple.

This temple seems to have suffered very little if any injury during the last century; but it is in a very incomplete condition, and the portico, the noble entrance-door of the cella, and a small portion of the wall, are probably all that remain of the original structure. I am inclined to attribute to a later period, probably that of the Byzantines, the masonry now existing on the site of the cella. It is in a style very commonly met with in Africa.

A course of cut stone was laid horizontally, a long cut stone was erected at intervals of four or five feet on this, and the interstices filled in with rubble masonry, exactly like the long and short bond found in Roman and early Saxon work in Britain. Some excellent specimens of this still exist at Teboursouk, twenty or thirty feet in height. This explains the occurrence of so many upright stones in all ruined cities throughout Africa; the masonry of inferior quality has fallen away, bringing down the whole of the superstructure, and leaving only the uprights resting on the foundations. As this part of the building has a semi-circular apse at its extremity, it is probable that it was intended for a Byzantine basilica.

To return, however, to the temple. It is a tetrastyle, with a noble portico, of the Corinthian order. The columns are fluted, and with one exception, monoliths. The dimensions of the buildings, as given in Bruce’s plan and elevation, are as follows:—

In saying that the ornamentation of this temple was the richest he ever saw, Bruce was no doubt right, especially if he alluded to the monuments still retaining a certain purity of style, which he had met with in Africa; but in Italy the temples of the latter half of the second century were generally most highly decorated. In the case of this temple the frieze has an inscription, but otherwise it is without ornaments of any kind. The architrave is divided, as usual, into three bands, but the mouldings are simple, without oves, pearls, or other ornaments. On the other hand, the cornice is highly decorated, and the pediment is enriched with a grand piece of sculpture.

Bruce, evidently misled by the occurrence of the letters NERVAE in the inscription on the frieze, imagined that these alluded to the Emperor Nerva, and inferred that the temple had been built by Hadrian, and that the sculpture, which represented a male figure on the back of an eagle, was

Sir Grenville Temple’s supposition, that it was intended to represent the rape of Ganymede by the eagle of Jupiter, is much more likely to be correct.[207]

On the frieze of the order is an inscription, now almost effaced, but which has been recorded by Bruce more fully than by subsequent travellers. It is as follows:—

The door of the cella is nearly all that remains of that part of the temple. It consists of three huge stones, a lintel and two door-posts, the former of which projects a considerable distance beyond the latter. These are enriched with a moulding on the exterior edge of the stones, which, instead of mounting in a straight line from the ground to the top of the lintel, as would probably have been the case in an earlier period of Roman art, follows at right angles the course of the projecting portions of the lintel. A similar style is often met with in Etruscan architecture, but in such cases the line of moulding under the projections hardly ever formed horizontal straight lines and right angles, unless painted or cut in the inside of tombs.

On this lintel is a second inscription:—

From these inscriptions it is evident that the temple was built by two brothers, L. M. Simplex and L. M. Simplex Regillianus, at their own expense, sua pecunia fecerunt, in honour of Jupiter and Minerva, during the reign of Marcus Aurelius and his colleague in the Empire, L. Aurelius Verus, between the years A.D. 161 and 169.

No strong marks of decadence strike the eye at first sight, but the antæ at the end of the cella walls are wanting, and perhaps other signs of decadence would have been apparent in the cella itself, had it been preserved.

Bruce states that the material of which the temple is built is white marble. If this is not actually the case, it is a very compact and crystalline limestone, full of fossil shells, and susceptible of receiving a high polish; when new it must have been even more effective than the finest description of marble. I am inclined to believe that it is none other than the Lumacchella antica, one of the lost Numidian marbles, of which only two or three specimens are known to exist, and which at the present moment is worth something like its weight in gold. A bold and conspicuous hill was pointed out to us as the spot whence the stone of the temple was obtained, but it was too remote to be included in our visit. I leave the solution of this question to some future traveller.[209]

Altogether this grand monumental fragment is a most interesting historical specimen of the workmanship and architectural genius displayed by the Romans in their African possessions.

The inhabitants of Dougga were not ungrateful to these good citizens for their unusual liberality; they appear to have voted a statue to one of them at least, the pedestal of which was discovered by M. Guérin.[210]

The precincts of this temple are in a state of great filth and neglect. It is surrounded on every side by rude Arab huts, and is used as a yard for sheep and oxen, the handsome flight of steps leading to it are partly buried in a thick layer of manure, and we found it impossible from any point to obtain a view showing the whole columns from base to capital. Nevertheless my companion made some excellent photographs of it, which testify in a remarkable manner to the conscientious accuracy of Bruce’s drawings, when not marred by the unfortunate accessories of Balugani.

One day at Dougga is quite insufficient. I would recommend anyone having abundant time at his disposal to encamp for a week amongst its delightful shady groves.

Unrivalled as this temple is as a work of art there is another monument of even greater interest, the celebrated mausoleum from which the Dougga bi-lingual stone was obtained.

Bruce has left a pencil sketch of this in outline merely (Plate XXIV.); but it is exceptionally interesting, as the monument itself was almost entirely destroyed by Sir Thomas Reade, the British Consul-General at Tunis, in extracting the stone on which the inscription was carved. We met people on the spot who were present at this sacrilege, and who described to us the manner in which stone after stone of this, beyond all question the most interesting, because the only pre-Roman, monument in the Regency of Tunis, was hurled by levers and crowbars into the valley below.

Such a proceeding would have been indefensible had the object been to enrich some great public museum, but destruction so wanton to secure an object of interest for a private collection cannot be too strongly reprobated. Half the sum expended in destroying the mausoleum devoted to making a plaster cast of the inscription would have sufficed for every purpose, and even if some future traveller had carried away the precious relic, at least the guilt would not have been chargeable to the British nation. The deed being done, it is fortunate that the two slabs containing the inscription were purchased by the British Museum at the sale of Sir Thomas Reade’s collection in 1852.

The monument was square in plan, two storeys in height, forming what might almost be regarded as two distinct tombs superposed one on the other, the whole surmounted by a graduated pyramidal roof of receding steps. It is supported on five steps, averaging 16·2 inches in height, with treads 10 inches wide. The lower storey 21 or 22 feet square, and 10 ft. 8½ in. high, with semi-plain and slightly projecting pilasters or antæ at the angles, surmounted by Ionic capitals, one volute being on each face, from which spring two lotus-like flowers, one from above and the other from below. A plain fascia, surmounted by a few bold simple mouldings, forms a cornice to the lower order, 1 ft. 9 in. high, or one-fifth the height of the pilaster; it runs in an unbroken line along the four faces of the tomb. A false window appears on three of the sides.

The arrangement of the courses of stone is peculiar. A narrow square plinth or base, rather less than eleven inches high, from which the pilasters appear to spring, is carried all round. Above this is a course four times as high, then a narrow band half as high again as the base, followed by another high course slightly lower than that beneath it. Above is a fifth course, half the height of that immediately beneath it, which contains the capitals of the pilasters, and formerly bore the inscription. The upper storey, of which very little now remains, resembled the lower one in its general divisions and style of construction. It was supported by three steps, 16½ inches high, with treads 10 inches wide, and appears to have been a sort of tetrastyle of the Ionic order, probably with pilasters at the angles, certainly with two intermediate, attached, fluted columns, 15 inches in diameter on each front, separated by an interval of 3 ft. 5 in. On the north and east faces, between these columns, were small doors, closed by portcullises of stone, giving access to the interior. They had dressings all round, and the architrave above them was very high. The construction of this storey is similar to that of the lower one, consisting of four unequal courses, dividing the height of the columns into six parts. The entablature in Bruce’s drawing (Pl. XXIV.) equals one third of the height of the columns. It is divided into architrave and cornice, the former consisting of a plain fascia, with narrow but bold mouldings. The latter has a distinctly Egyptian character, with a high and boldly projecting cavetto, surmounted by a fillet.

In a more recent view of this tomb, taken by Mr. F. Catherwood in 1832 with the aid of the camera lucida, for which I am indebted to Professor Donaldson, there appears an intermediate frieze between these two features, which would give the very unusual height of half the column to the entablature.[211]

Above this rose the pyramid, with a large upright block of stone at each angle as high as three or four of the steps, crowning classically the tomb underneath. Probably this never rose to an apex, but was truncated to receive a group of statuary, like the mausoleum of Halicarnassus. Sir Grenville Temple[212] mentions having dug up at the base, a portion of a draped figure, and at no great distance a small mutilated equestrian statue. He also says that on the base of the monument was a coarsely executed alto-relievo representing a quadriga with a warrior and the charioteer.

The monument contained two tiers of sepulchral chambers, one in each storey, divided into cells by vertical walls.

This mausoleum and the Medrassen in Algeria are the only monuments in North Africa of a pre-Roman origin, and the only examples remaining of the style employed by the earlier aboriginal races. The height of the courses, and the capitals of the antæ in the lower storey, would indicate a Greek origin, as does also the upper storey, which recalls the Tomb of Theron at Agrigentum, but the large cavetto of the cornice, and the lotus flowers with which the volutes of the capitals are ornamented, are identical with the Egyptian type. The whole is of a purer style than Roman tombs generally, and excels in this respect every other similar edifice in North Africa.

The inscription has frequently been published. The first copy of which we have any record was one made by a Frenchman of the name of Thomas d’Arcos in 1631, which he delivered to the learned scholar Isaac Peirese; afterwards it was entirely forgotten until Camillo Borgia copied it again in the year 1815. This copy became known to the world through Münter, Humbert and Hamaker. Two other copies were made by Sir Grenville Temple[213] and Honegger, and published by Gesenius.[214] The last facsimile was taken by order of the Duc Albert de Luynes, from an impression of the stone then in the British Museum, and published in M. Guérin’s work.[215]

The mausoleum appears to have been erected in honour of a Numidian, and not of a Carthaginian, which is supposed to be the reason why the Lybian version of the inscription is more carefully executed than the Punic one, and the place of honour, the right side, assigned to it. The Punic text appears to be the translation of the Lybian one.[216]

The following translation of it is given by Gesenius:—[217]