Chapter VIII: The Extinct Lakes of the Oasis

Character and Extension of Lacustrine Deposits — Modern Erosion by Wind-borne Sand — The Lakes geologically of Recent Age — Discovery of Pottery and Bones of Domesticated Animals — Area occupied by the Lakes — Maximum Level of their Waters — Lacustrine Strata at Gorn el Gennâh — Relation to the Ancient Monuments — Altitudes of Archæological Sites — Age of the Lakes and their Persistence into Historic Times — Flint Implements — Origin of the Lakes — Their possible connection with the Artesian Waters — Lacustrine Deposits form the Cultivated Lands of Modern Times.

When in 1906 I commenced to make a special study of the geology of the oasis, with a view to elucidating certain questions which had arisen in connection with the water-supply, it came to me as a very great surprise to find indubitable evidence that the greater part of the floor of the depression had at one time or another been the site of an immense lake. No mention had been made by previous observers of the extensive accumulations of lacustrine sediments which cover so large a proportion of the floor, and are found from near Ain el Ghazâl in the north to beyond Beris in the south.

These deposits consist of horizontal finely-bedded alternations of sand and clay, or more frequently of an intimate mixture of the two; local false-bedding is not uncommon, and included fragments of limestone or sandstone are occasionally met with. The beds have a prevailing brown tint, and frequently exhibit well-marked hexagonally disposed shrinkage cracks. Although originally they must have formed an immense compact and continuous sheet, the deposits have since been subjected to considerable denudation, so that at the present day they exist as large isolated patches. Perhaps the most striking of these is that occupying the centre of the depression between Kharga village and Jebel el Ghennîma, covering an area of between 40 and 50 square kilometres, and over the greater part cut by the sand-blast into thousands of isolated hummocks, disposed with their longer axes parallel and in the direction of the prevailing north winds. Individual hummocks have perhaps an average height of 4 or 5 metres, though many exceed this considerably; in length they may measure anything up to 40 or 50 metres. The northern end of a hummock is in nearly all cases the larger, the gradual tapering towards the south being a most distinctive feature. Their present shape and appearance are, of course, entirely due to the eroding and sculpturing action of sand-laden wind.

The finding of these extensive lacustrine deposits naturally opened up a number of questions of the greatest interest and importance. To what level had the waters of the lake attained; what were its limits horizontally; and, most important of all, at what period had it existed, and what were its relations to the ancient monuments of the oasis? With the object of solving these problems it has been my endeavour to collect all possible information concerning the deposits, and, although it is perhaps too soon to draw inferences with any great certainty, I propose to put on record such data as I have been able to obtain, and to indicate the conclusions to which they appear to lead.

It is, of course, obvious that the lake was, geologically speaking, of comparatively recent date; the lacustrine deposits have nothing in common with those of the Cretaceous and Eocene formations which build up the plateau-massif out of which the oasis-depression is hewn. The sands and clays are much softer and less consolidated beds, laid down superficially and unconformably on the uneven surfaces of the older formations; the depression had, in fact, almost attained to its modern dimensions before the beds in question were formed. My first impression was that the lake dated from prehistoric, if not prehuman, times; that it existed before the prevailing desert conditions set in, when the Nile Valley was the site of extensive lakes and the country was partially wooded. The existence of thick deposits of calcareous tufa on the upper portions of the cliffs of the depression suggested a considerable outpouring of water from springs, and it seemed not improbable that the surplus water collected on the floor of the depression below.

For some time I was unable to discover any clue to the age of the deposits, in the shape of organic remains or human relics. Then, in a pit sunk for surface-water in the neighbourhood of Headquarters,[8] some fragments of pottery were found at the base of the deposit. This discovery was followed by the finding of other pieces of pottery firmly embedded in some of the clay hummocks to the south-west of Headquarters. These were without question _in situ_, and proved that the lake was contemporaneous with man.

Further search, in which I was greatly assisted by my friend H. H. Baker, M.B., led to the discovery of human settlements, apparently on the margins of the lake. In these localities large quantities of broken pottery were found associated with the bones of domesticated animals, while the fresh-water gastropod shell, _Melania tuberculata_, was found to be abundant in some of the beds. Careful exploitation of one or two of these sites, about 6 kilometres south-west of Headquarters, enabled us to procure a number of complete earthenware vessels, the chief types of which are shown in one of our illustrations. A comparison of these with pottery of known age must be left until we have referred to the horizontal and vertical extension of the lake.

Judging from the still existing deposits, and utilizing (in the northern part of the oasis) a considerable number of levels, I believe the limits of the lake to have been as shown on the accompanying plan. From its most northerly point in lat. 25° 45′ N., in the neighbourhood of the modern Ain el Ghazâl, its western boundary trended S.S.W., passing about 1½ kilometres west of Meheriq village to the south-west end of Jebel Têr; thence its shore-line lay at the foot of Jebel Tarwan, rounding which it projected somewhat into the Bellaida district between Jebel Têr and Jebel Tarif. South of this it bore slightly west of south, passing about 2½ and 3 kilometres west of Kharga and Gennâh villages respectively. From the latter point the shore-line proceeded almost due south to lat. 25° 2′ N., in the vicinity of Ain Girm Meshîm.

Its eastern limit appears to have been about a kilometre east of Ain el Qasr, and some three times that distance east of Meheriq, so that on this latitude the lake had a breadth of 4½ kilometres. In the neighbourhood of Headquarters its margin lay 2 kilometres to the east, whence it ran almost due south for about 12 kilometres. Here, just to the south-west of Ain Harrân, the lake attained its greatest width—_i.e._, 15 or 16 kilometres, gradually diminishing southwards to Ain Girm Meshîm. The total length of this portion of the lake was just over 80 kilometres.

Both at Ain Girm Meshîm and Ain el Doum the lacustrine deposits are well seen; over the intermediate country they were not observed, and as this area lies at a somewhat high elevation, it seems probable that there was a break in the continuity of the lake of about 15 kilometres in extent.

The southern portion appears to have measured approximately 45 by 15 kilometres, its long axis lying along a line passing through Ain el Doum, Beris, and Maks Qibli. To the south-east the lake had a local extension in the neighbourhood of Dush. It is thus seen to have extended, with a possible break of a few kilometres in the district near Ain Girm Meshîm, almost throughout the length of the depression, over a distance of 136 kilometres, or 85 miles.

It proved a matter of some difficulty to determine the maximum height to which the lake had reached, as in most localities the deposits have suffered considerable denudation, and their margins are usually obscured by blown sand and superficial detrital material. The pottery to the south-west of Headquarters was ascertained to occur at 47 metres above sea-level, the uppermost limit of the deposits in that area being 62 metres. At Headquarters the lake clays occur up to between 65 and 66 metres; at Ain Terfai and Ain Mahmud, north of Meheriq, to 65 and 67 metres respectively. Still farther north I found there were stretches of similar clays at 76 metres near Ain el Qasr, and at 84 to 85 metres in the neighbourhood of Ain el Ghazâl.

On the west side a very well-defined plain formed of the lake beds occurs immediately to the east of the Necropolis, and was found to lie at 70 metres above sea-level. But the most valuable data of all were those obtained at the Gorn el Gennâh. There, as anticipated, the lacustrine deposits, containing _Melania_ and _Limnæa_ in abundance, were found to be well developed and exposed, being piled up on the flanks of the hill on its north, east, and south sides, the exact limits of the beds being best seen on the south-east side near the ruins of a columbarium. Immediately behind the latter the top of the stratified clays was determined as 66½ metres, a figure agreeing remarkably closely with the majority of the maximum heights farther north. Still higher up, however, are other clays, which, while similar in general appearance, are either very irregularly bedded or altogether unstratified. The upper limit of these clays was found to be 82·76, or, say, 83 metres. While these highest beds may possibly have been deposited by the waters of springs issuing from faults and fissures (the line of disturbance mentioned in a previous chapter passes through the Gorn at this point), one must, in the absence of any definite evidence to the contrary, regard them as probably representing the extreme marginal deposits of the lake.

We may, indeed, conclude that, while the lake presumably at one time reached to a maximum level of 85 metres, it stood for a considerable period at about 70 metres above sea-level.

The next step was to ascertain the altitudes of the archæological sites and the relation of the latter to the lacustrine deposits. Up to the present time levels have not been carried through to the south end of the oasis, and aneroid determinations are not of sufficient accuracy for our purpose, so that we must confine our attention in this connection to the northern portion of the depression. The ground-level of the temple of Hibis was found to be at 75 metres and the lowest of the tombs of the Necropolis at 80·6 metres above sea-level. The columbarium already alluded to, on the south-east side of the Gorn el Gennâh, is built on the denuded slopes of the lacustrine series at a level of 57·2 metres, while the base of a number of brick ruins a little to the north was found to be just over 53 metres. Finally, the ground-level at the base of the slope on which stands Qasr Zaiyan was determined as 21 metres, a bench-mark being made on the southern door of the temple at 24·52 metres above sea-level.

There are several points in connection with the disposition of the archæological remains which cannot but strike one. In the first place, they are mostly on or near the extreme margin of the lacustrine deposits; secondly, they are absent altogether from the central portion of the lake site; and thirdly, the older monuments occupy the highest levels, while there are no representatives at all of the earlier Egyptian periods. While the disposition of the monuments may, of course, be entirely fortuitous, we are justified, I think, in assuming that the lake existed well into the historic period, and may have still stood at the 65 or 70 metre level when the temple of Hibis was erected by Darius, about 500 years before the commencement of the Christian era. In the time of the Ptolemies it was certainly considerably lower, while in still later days, when the country became a Roman province, the lake had very much contracted, and probably only existed as a marshy swamp occupying the lower portions of the depression.

What age, it may be asked, is indicated by the pottery, bones, and fresh-water shells which, as already mentioned, have been found in the lacustrine deposits? Unfortunately the shells belong to species which have a wide distribution, and have persisted from early Pleistocene times right up to the present day. They do not, therefore, help us to fix the age of the deposits, except within wide limits. The bones were submitted to Dr. Andrews, F.R.S., of the British Museum Staff, and his conclusion is that they belong to two domesticated animals—one a small, lightly-built ox, the other a small horse, donkey, or zebra; unfortunately they are not sufficiently complete to be determined with certainty. Finally, as to the pottery: the types do not differ in any important respect from those associated with the towns and cemeteries of Græco-Roman age in many parts of the oasis. The barrel-shaped pot has, indeed, persisted to modern times, being, in a slightly different design, the standard water-jar at the present day in the oasis of Dakhla. The pottery, therefore, bears out our conclusions that the lake continued to exist well into the historic period.

Although flint implements have never yet been detected _in situ_ in the lacustrine deposits, I have collected a considerable number from the area originally occupied by the lake. Some of these were found lying on the denuded surfaces of the lake beds, in positions which lead me to suppose that they have weathered out from the deposits. They are decidedly Neolithic in workmanship and character, and were, in my opinion, used by people inhabiting the depression at the time of, and probably prior to, the existence of the lake. A number of examples are shown in one of the accompanying plates; the figure in the bottom left-hand corner is, however, that of a palæolith, a typical example of those which occur on the borders of the plateau, and on or near the eastern escarpments. In the right-hand bottom corner is figured an object of very common occurrence in the oasis—a sandstone hand-grinder, probably of Roman age. With the exception of these two the specimens illustrated are all implements from the site of the lake. Flint implements were evidently in use in Kharga at a very late period, as large numbers of worked flints, mostly in the form of flakes or tools of very poor finish, occur near many of the old sanded-up wells. A considerable amount of detailed work will have to be undertaken before the different flints of the oasis can be satisfactorily arranged in chronological order.

From what period did the lake date, and to what cause did it owe its origin, are questions which cannot, I fear, be definitely answered as yet. While it is quite reasonable to consider, in the absence of any decisive evidence to the contrary, that the lake originated in early prehistoric or Pleistocene times—that it dates possibly from the time of the formation of the tufas of the Nile Valley and oasis-escarpment, when the climate was certainly much moister than at the present day—we must not forget the possibility that it was formed by artificial means during one or other of the Egyptian dynasties between 3000 and 1000 B.C.

There is good reason to believe that the depression was inhabited previous to the formation of the lake, and although we have no information as to when the first deep borings were made, there is some evidence which leads me to suspect that wells existed prior to the time when the surface of the lake began to fall, if not very much earlier. The evidence to which I refer is the discovery of a portion of an earthenware pipe embedded _in situ_ in the lake clays at a height of 42·85 metres, of the type used by the ancients for lining their water-channels. Is it possible, therefore, that, as the result of the industry of the ancient well-borers— following their initial discovery of these deep-seated sources—the long-confined waters welled up with irresistible force, and gradually flooded the country. At the present day, when the pressure throughout the artesian basin must be very much less than formerly, it is not uncommon for new bores to get out of control and flood considerable areas of country. The same thing may possibly have happened on a very much larger scale thousands of years ago, when the first bores were sunk into a previously untapped artesian basin, fully charged with water under great pressure.

There is another explanation which it is advisable to keep in mind, though it has never hitherto, as far as I am aware, been advanced as a possible cause of the formation of lakes. The very existence of artesian water depends on the presence of porous strata overlain by impermeable beds. If one or other of the porous beds, charged with water under pressure, should, through the action of denudation on the overlying beds, become exposed at the surface, the waters would escape through natural springs in very large quantities. This might, indeed, continue for a long period of time, until the bed was nearly depleted and the pressure reduced to nil. There is little doubt that the beds which we have named the ‘Surface-water Sandstone,’ and which are now exposed in places on the floor of the oasis, were originally entirely covered by impervious clays, and contained artesian water under pressure; they may, in fact, have been in the same condition as is the Artesian-water Sandstone at the present day. It is conceivable, therefore, that when those beds became exposed at the surface, owing to the removal of the overlying confining strata, their contained waters escaped in such quantities as to have given rise to a lake of considerable dimensions, if not to one equal in size to that which we have been considering.

No one can be more conscious than myself of how much there is still to be learnt regarding the topographical aspects of the oasis in early historic times, and although the theory that the lake may have been formed by waters which escaped from the underlying water-charged beds (either artificially through bore-holes or naturally as springs) may at first sight appear fanciful, it is one which, in the present imperfect state of our knowledge, is at least worth bearing in mind. A more detailed examination of the lacustrine deposits would probably throw further light on the matter; and a minute analysis of the nature, composition, and arrangement of the individual grains would almost certainly show whether the beds have in the main been formed of sediment carried into the lake by streams from the neighbouring cliffs and plateaux, or whether they represent wind-borne accumulations of sand and clay-dust from the surrounding plains.

At whatever period and in whatever manner the lake may have been formed, it is quite clear that it existed well into historic times, and that on its bed were laid down thick deposits of clay and sand, which at the present day form the bulk of the cultivated lands of the oasis.