I.—The Lake of Van
It is about six times as large as the lake of Geneva, having an area of some 1300 square miles. Its western shore is erroneously laid down in existing maps; and this necessitated a particular survey of that region during my second journey, the result of which has been to invest the lake with a shape of greater symmetry—a central body with two arms, one on the north-east, the other on the south-west. The remainder of the outline I have borrowed from the best available sources, adapting them to the position of Van, of which the latitude and longitude are approximately known, and correcting them as well as possible by sketches, and readings to the principal points from the summit of Sipan. If my reader will turn to the map which accompanies this work he will, I think, be able to transfer, with the aid of a few illustrations, the features which are there conventionally delineated into a picture visible by the mind’s eye.
How strange it seems that at the end of the nineteenth century one should be engaged in exploring and mapping this fine country, one of the fairest and most favoured of the Old World! How should we be able to explain, still less to justify, the circumstance to some visitor from another planet? It lies about in the centre of the land area of our hemisphere; the climate is bracing, water is abundant, the sun is warm. Yet it is so little known to the more civilised peoples that their travellers journey thither with the aid of a compass through districts which are now deserts, but which are well capable of supporting the races that are highest in the human scale. The case would appear to have been much the same during the period of the expansion of Greek culture and of the later and beneficent sway of Rome. The knowledge displayed of these regions by representative writers like Strabo, Pliny and Ptolemy is, to say the best of it, vague and fabulous. Yet Strabo, the contemporary of Augustus, was a native of Asia Minor; the countrymen of Pliny had carried the Roman eagles to the Araxes; and Ptolemy wrote during the reign of the Emperor Hadrian whose statue, commemorating his journey through the interior, looked out upon the waves above Trebizond. The first of these authorities plainly confuses the position of Lake Urmi with that of Lake Van; but he is well acquainted with the essential characteristics of both sheets of water, different and strongly marked as these are. The former is described as largest in area, and second in size to the sea of Azof; its name is interpreted to signify the deep blue (κυανῆ ἑρμηνευθεῖσα). The water is salt; and there are salt works in the neighbourhood.[1] The peculiar properties which actually distinguish the latter exactly tally with the language of Strabo, who, speaking next of the lake Arsene or Thopitis, says that it is charged with nitre, which word would seem with him to signify carbonate of soda,[2] and that it washes clothes as though they had been scoured. He adds that the water is undrinkable and supports only one kind of fish. And he proceeds to relate a circumstance which is repeated and embroidered by Pliny, and which is so curious that I cannot refrain from extracting the whole passage from the work of the last-named writer.[3]
“Meet also and convenient it is to say somewhat of the river Tigris. It begins in the land of Armenia the greater, issuing out of a great source; and evident to be seen in the very plaine (fonte conspicuo in planitie). The place beareth the name of Elongosine (or Elegosine, Elosine, Elegos). The river it selfe so long as it runs slow and softly is named Diglito; but when it begins once to carry a more forcible streame it is called Tigris, for the swiftnesse thereof; which in the Median language betokens a shaft (sagitta). It runs into the lake Arethusa; which beareth up aflote all that is cast into it, suffering nothing to sinke; and the vapors that arise out of it carry the sent of nitre. In this lake there is but one kind of fish, and that entreth not into the chanell of Tigris as it passeth through, nor more than any fishes swim out of Tigris into the water of the lake. In his course and colour both he is unlike, and as he goes may be discerned from the other: and being once past the lake, and incountreth the great mountain Taurus, he loseth himself in a certain cave or hole in the ground, and so runs under the hill, untill on the other side thereof he breaketh forth again, and appeares in his likenesse, in a place called Zoroanda. That it is the same river it is evident by this, that he carrieth through with him, and showeth in Zoroanda, whatsoever was cast into him before he hid himselfe in the cave aforesaid. After this second spring and rising of his he enters into another lake, and runneth through it likewise, named Thospites; and once again takes his way under the earth through certain blind gutters, and 25 miles beyond he putteth forth his head about Nymphæum. Claudius Cæsar reporteth, that in the country Arrhene, the river Tigris runs so neere the river Arsania, that when they both swell, and their waters are out, they joyne both their streams together, yet so, as the water is not mingled: for Arsanias being the lighter of the twain, swimmeth and floteth over the other for the space wel-neere of 4 miles: but soon after they part asunder, and Arsania turneth his course toward the river Euphrates, into which he entreth.”
We need not discuss in this place the phenomenon last mentioned, except to remark that the story may well have been suggested by the propinquity of the sources of the Diarbekr branch of the Tigris to the stream of the Murad, the ancient Arsanias. The country of Arrhene is probably the same as that better known as Arzanene, which is comprised within the present vilayet of Diarbekr. Our present interest in the passage lies in the statements relative to the Tigris, that it flows through two lakes called Arethusa and Thospites. Strabo, in speaking of the same phenomenon, attributes it to one lake only, namely that of Arsene or Thopitis. The river, according to him, rises in the Niphates mountains, by which name he seems to be referring to the Nepat of Armenian writers, the modern Ala Dagh. After flowing through Lake Thopitis it disappears in a chasm at the corner of the lake. It comes to light again in the province of Chalonitis; and, although later on he attributes that province to the Zagros, I cannot help thinking that the sense which his informants wished to convey was that it came to light in the mountains of the peripheral region. The mention of two lakes by Pliny need not perplex us over-much; for his Arethusa no doubt denotes the Arjish arm of Lake Van, and his Thospites the principal body of water with the city of Van, the Dhuspas of the cuneiform inscriptions, upon its eastern shore. Ptolemy, on the other hand, entangles the subject still further by separating the lakes of Areesa—no doubt the Arethusa of Pliny—and Thospitis by four degrees of longitude. This geographer does not give us any indications as to the properties of the lake waters; but he tells us that the Tigris is partly a river of Armenia and that its sources constitute Lake Thospitis. The position which he assigns to the town of Artemita—which is probably the modern Artemid—is further evidence that in speaking of Lake Areesa or Arsissa he was in fact referring to Lake Van. One cannot help concluding that his Thospitis with the town of Thospia was actually the self-same sheet of water. The discrepancy in longitude finds a parallel in the degrees assigned by this writer to Lakes Sevan and Urmi. They are really upon the same degree. Yet Ptolemy, under the names of Lychnitis and Martianes, assigns to them the difference of over four degrees.
I think it is plain that the names Thopitis, Thospites, Arsene, Arethusa, and Areesa or Arsissa, are all applied to the great basin with the two immemorial cities, Dhuspas—the modern Van—and Arjish. Moreover, I should be surprised to learn that any lake exhibiting the same properties had been discovered in the belt of mountains south of Lake Van in which the present sources of the Tigris are found. Put together, the scraps of information retailed by the classical geographers go to show that in their days there existed a widely spread belief that the Tigris drew its waters from the tableland of Armenia, flowed through a lake strongly impregnated with soda, and disappeared in a chasm at its further and narrow extremity (μυχός) to come to light again on the further side of the barrier of Taurus or, in other words, of the parapet of mountains which are aligned upon the south coast of Lake Van. The mention of two lakes by Pliny and Ptolemy may point to a former isolation of the Arjish arm. I have taken the trouble to set forth these accounts—though not with all the care that I should desire—because they have an important bearing upon the subject to which I now proceed—a brief notice of some of the peculiarities which distinguish Lake Van.
It may not be out of place to cast one’s look a little further so as to include the other great lakes. That of Urmi in the Persian frontier province of Azerbaijan has an area of 1823 square miles. Its extreme length from north to south is about 80 miles, and its breadth from east to west 24 miles. It resembles its neighbour on the west in constituting an isolated basin, many rivers flowing in but none out. On the other hand its insignificant depth invests it with the character of a lagoon; the average being probably not more than 20 feet and the maximum some 45 or 50 feet. Evaporation must be very rapid over such a sheet of water; and it is at once situated further south than the lake of Van and at a level which is lower by 1500 feet (Lake Urmi, 4100 feet; Lake Van, 5637 feet). Abnormal salinity is the special feature about the waters of Lake Urmi; and extensive beds of rock salt are found in their vicinity. It has been estimated that they are six times as salt as the ocean, though only three-fifths as heavily charged with saline matter as the waters of the Dead Sea. Viewed from a height they are coloured a deep azure, a characteristic usual with salt lakes. If they are allowed to dry upon the body of the bather it is as though he had been covered with flour, and neither fish nor molluscs can live within them. The shores of the lake, which are in general low, are impregnated with salt; and the margin, upon which are found fragments of fossil coral and shell, shines like a white ribbon by the side of the blue. Three boats of not more than 20 tons burden compose the entire fleet of this inland sea.[4]
Very different is the description which may be given of Lake Gökcheh (the blue) or Sevan—the Lychnitis of Ptolemy, the lake of Gegham or of Geghark in Armenian literature. It is situated at a level of 6340 feet, and is therefore the most elevated, if also the smallest, of the three great sheets of water upon the surface of the tableland. It lies at a distance of about 130 miles north of the northern shore of Lake Urmi, and close to the barrier of the mountains of the northern peripheral region. Its waters are sweet and support delicious salmon trout; they are said to attain a depth of 360 feet, or, according to another observer, of 425 feet.[5] Gökcheh is in fact essentially an Alpine lake, lying restfully in the lap of a circle of mountains of which those on the southern shore are of eruptive volcanic origin. It has an outlet on the west to the river Zanga, and a portion of its waters find their way through this channel to the Araxes. The balance of opinion inclines to the view that this connection is of artificial origin; and when the lake is low, especially in autumn, the stream will be almost dry.[6]
But both Urmi and Gökcheh sink into obscurity when compared to the lake of Van. Almost as large as the one and perhaps deeper than the other, it at once combines some of the characteristics of either basin and adds others essentially its own. Like Urmi its waters are heavily charged, though with soda rather than with salt. Its great elevation and its juxtaposition to the mountains of the peripheral region recall corresponding features in Gökcheh. But like a book which may borrow much from the work of other writers, and yet produce an effect on the reader which is wholly new, so one opens the landscape of Lake Van with that particular emotion which only very beautiful and original objects can produce. With the wondrous pieces of natural architecture about the margins of this inland sea my reader will become perfectly familiar as this work proceeds. My present object is to fly very low to the ground, and to notice such facts as appeal to the mind rather than to the eye. The extreme length of the lake would seem to measure 78 miles, and the breadth from north to south of the principal body about 32 miles. To all appearance it is very deep except at the north-east and south-west extremities; but no systematic soundings have been taken to my knowledge, though it would be extremely interesting to know whether indications can be traced of the Arjish arm having once composed a separate unit. The principal streams enter the easterly portion of the basin; they are the Erishat or Irshat near Akantz, the Bendimahi Chai, the Marmed and the Khoshab. Several little rivers are collected in the delta below the old Akhlat, and quite a nice stream cascades into the lake at the neighbouring village of Karmuch, which probably collects a portion of the drainage of the plain between Nimrud and Lake Nazik. No issue of the sea has yet been discovered. None of the copious springs which feed the Tigris on the southern side of the parapet of mountain, quite close to the flood washing its northern slopes, has yet been shown to possess any of the strongly marked qualities characteristic of the waters of Lake Van. One of the most remarkable of these springs is situated near the south-west corner of the lake, at Sach in the Güzel Dere or beauteous valley—a valley with a specially appropriate name.[7] It has been examined by Major Maunsell, who describes it as issuing from the base of a cliff and immediately constituting a stream 50 yards wide and 18 inches deep. It is quite possible that this source of the Tigris may have given colour to the belief of the ancients that the river flowed through the lake and found an exit at its further end by an underground channel. Another scarcely less interesting fountain in the neighbourhood is that of Norshen at the head of the plain of Mush. It rises in a circular pool with a diameter of 105 feet, from which it wells over into a stream which runs to the Euphrates. The natives hold that it is in connection with the lake in the crater of Nimrud, and relate how a shepherd, whose staff, weighted with a small parcel of coin, had sunk below the surface of that deep mere, had one day been astonished to see the lost object eddying in the current of the pool of Norshen. Careful scrutiny of the spring during my second journey established the conviction that it affords no outlet to Lake Van. Moreover, its position and the delicious flavour of its water point to its being derived from the limestones of the range on the south of the plain.
Analysis of the waters of Lake Van has furnished results which are described as remarkable by the eminent chemist to whom I submitted the sample which I brought home with me, and which I obtained by swimming out from the rocky shore at Erkizan, some distance east of the abandoned Ottoman fortress of Akhlat. The amount of suspended matter has been found to be very trifling; while the proportion of solids in solution, principally carbonates of potassium and sodium, chlorides and sulphates, is very large indeed. It is estimated that the alkalinity is equal to rather more than 3¼ ounces of ordinary soda crystal dissolved in a gallon of water. The presence of a little silica accompanies the alkali. The account given by Strabo of the cleansing properties of the lake is thus confirmed in a striking manner. Indeed, the bather issues from his swim as though his limbs had been rubbed with soap—but with a soap of extremely agreeable quality, leaving a velvety feeling upon the skin. The great buoyancy of the waves enhances the pleasure of such exercise, and they are at once pellucid and sparkling under the ruffle of the breeze. On the other hand they are most unpleasant to the taste. The colour of the sheet of water cannot be given in a single word; and indeed it varies with extraordinary range of scale. A cobalt of great brilliancy is perhaps the most normal hue; but a certain milky paleness is seldom quite absent, becoming invested at morning and evening with an infinite number of delicate tints.[8]
Only one kind of fish is found in Lake Van, resembling a large bleak. But, often as I have bathed, I have never seen one gliding through the water, or surprised a shoal while following the shore. It is possible that they adhere to the estuaries of the rivers, up which they make their way in large numbers to spawn during the season of spring freshets. It is then that they are caught in great quantities by means of barriers placed at the mouth of the streams with baskets resting against one side. The fish leap the barrier and fall into the baskets, after which they are dried and salted. Seagulls and cormorants haunt the lake, but are not very numerous; nor have I observed a pelican, although these birds are conspicuous on the adjacent lake of Nazik together with many varieties of smaller waterfowl. The main body of the sea never freezes over in winter, rigorous as that season is at this high altitude.
A feature which has occupied considerable attention, especially among German writers, is the fluctuation in level of these Armenian lakes. There can be no doubt that they are all three subject to more or less pronounced periodical changes; and various reasons have been assigned. Do these fluctuations arise from the opening or closing of subterraneous issues or from movements of the earth’s crust? Or may they be accounted for by ordinary climatic conditions, such as the fall of snow and rain and the consequent variation in the volume of the rivers and in the activity of springs? The economic state of the country and the extent of irrigated land within the watershed has been recognised as a factor, but a factor of insufficient importance to produce the recorded results during the period reviewed. In the case of Lake Van we are precluded from attributing these fluctuations to the agency of subterraneous issues. Not a single one of such has yet been discovered. Nor am I aware that any such outlets to Gökcheh or Urmi have been noted by any traveller. The evidence which may be collected in the case of all goes to show that the islands are as much affected as the adjacent shores. It may therefore seem unlikely that the changes arise from movements at the bottom of the lake; for these would lift or depress the islands to some extent.[9] If I venture to join in the discussion I would submit the suggestion that we should for convenience group the phenomena under two heads. Temporary variations should be distinguished from any differences of a more permanent nature the existence of which it may be possible to prove.[10]
It cannot be expected that we should be able to collect evidence of a satisfactory nature in respect of the changes which would fall within the first category. We have to rely upon the statements and even upon the inferences which may be derived from the writings of travellers. Even if we could rest contented with the accuracy and sufficiency of such testimony in the case of lakes which are so much affected by the melting of the winter snows, it would not establish, except in a very approximate manner, the beginnings and ends of the successive phases. Still, the subject is so interesting that it is worth while to collate the observations of which record may be found. In the subjoined table I have endeavoured to perform this task; and it has already been undertaken with great diligence by Dr. Sieger. It will be seen that a certain correspondence may occasionally be traced in the periodical fluctuations which have affected the three sheets of water.[11] Perhaps the most remarkable evidence in this sense is that which is furnished by the almost simultaneous observations for 1898. Messrs. Belck and Lehmann for Lake Gökcheh, Mr. Günther for Lake Urmi, and my companion, Mr. F. Oswald, and myself for Lake Van, all bear witness to a rise in quite recent years. Our own investigations were made during the month of July of that year, and were confined to the westerly inlets of the lake. A prominent feature about these inlets was the tendency of the streams to form shallow lagoons behind a narrow barrier of alluvial sand. On the margin or even in the bed of such lagoons one might often see a group of willows. Some had been immersed a foot or two by the rise in the waters; and, while their neighbours on dry land were green and thriving, these were quite dead. The most notable example was observed by Oswald within the little broken-down crater on the southern shore opposite Akhlat. It receives the lake within its enfolding arms. We have called it Sheikh Ora after a little village of that name which was discovered in its south-east corner. Oswald sailed across to examine this interesting spot while I was busily engaged at Akhlat. Between the village and the water he came across a small grove of willows upon which the lake had gained. Those above the water line were evidently flourishing; but those which stood in the lake had been killed and their bark withered, so that many of the stems were quite gaunt and bare. The average diameter of the trunks of the dead and the living was not appreciably different. It was therefore not a question of an advance of the lake dating back very many years. On the other hand there had been time for the chemical properties of the water to exercise their destructive effect.
TABULAR STATEMENT OF THE EVIDENCE OF TRAVELLERS IN RESPECT OF THE FLUCTUATIONS IN LEVEL OF THE THREE GREAT LAKES.
| Year. | Lake Van. | Year. | Lake Urmi. | Year. | LakeGökcheh. |
| 1806 | Jaubert attests a gradual rise in the waters,threatening Arjish and the suburbs of Van (Voyage enArménie, etc., p. 139). | 1811 | Morier attests a relapse. The former island of Shahi hasbecome joined to the mainland by a swampy isthmus during the last twoor three years (Second Journey, p. 287, seq.). | ||
| 1812 to 1829 | Progressive relapse of about 10 feet during this period attested by Monteith(J.R.G.S. 1833, vol. iii. p. 56). | ||||
| 1838 | Brant attests a relapse which, according to thenatives, has effected a gain of one mile in ten years to the plain onwhich Arjish stands (Journal R.G.S. 1840, x. p. 403). | 1834 | Relapse attested by Fraser since his last visitin 1822 (Travels in Kurdistan, pp. 47 seq., andNarrative of Khorassan, p. 321). | ||
| 1830 | A low level, perhaps aminimum, is attested by Monteith. The canal to the Zanga is aninsignificant runnel, supplying the river with the smallest portion ofits waters (J.R.G.S. 1833, vol. iii. p. 43). | ||||
| 1838 | Loftus records a rise on native authority,commencing during the winter. In twelve months, viz., by the winter of1839, the lake is said to have risen nearly 6 feet. In the next two years, viz., by 1841,it is said to have risen altogether 10 to 12feet, necessitating the evacuation of Arjish by the inhabitants,the place becoming an island (Quarterly Journal Geol. Soc. 1855,p. 318). | 1838 | Autumn. Rise attested in general terms byRawlinson (J.R.G.S. 1840, vol. x. p. 8) and more precisely in1839, by Perkins on native testimony (Residence in Persia,Andover, 1843, p. 394). Rise has been gradual. | ||
| 1856 | Lieut. Owerin of the topographicalstaff of the Caucasus, estimates that nearly ⅛th of the waters ofthe lake find an egress through the canal to the Zanga (Petermann’s Mitt. 1858, p. 471). Other evidence goes toshow that in the forties and fifties the lake was certainly higher thanin Monteith’s time. | ||||
| 1847 | Hommaire de Hell attests a relapse (Voyage enTurquie, etc., quoted by Sieger, Schwankungen, p.6). | ||||
| 1850 | Layard attests a rise “during the last fewyears.” Many villages on the margin are partly submerged. Iskele,the port of Van, is still in- habited; but the greater part of thevillage is under water (Nineveh and Babylon, p. 408). [Layardwas perhaps only witnessing the effects of the rise which commenced1838.] | ||||
| 1859 to 1879 | Relapse during this period isassigned to the lake by Brandt (Zoologischer Anzeiger,ii. 523 seq.), from whose observations we may infer aminimum about 1879. Islands had formed; these again had become apeninsula. The canal to the Zanga seems to have been scarcely operativeat all. | ||||
| 1852 | Loftus attests a considerable relapse duringrecent years, said by the natives to have commenced in 1850. Arjish isconnected by a passable isthmus to the mainland for eight months in theyear (op. cit. p. 318). | 1852 | A relapse is attested by Perkins to Loftus (QuarterlyJournal Geol. Soc. 1855, p. 307). | ||
| 1856 | Rise may be deduced from N. von Seidlitz whoseems from a distance to have seen Shahi, an island in October(Petermann’s Mitt. 1858, pp. 228, 230). | ||||
| 1863 | Strecker records a continuous rise during the yearspreceding his writing, as evidenced by Turkish officials of hisacquaintance (Petermann’s Mitt. 1863, pp. 259seq.) | 1891 | Relapse has continued. Treesplanted thirty years ago on the margin of the water at the island ofSevan are now standing some 50feet away, and some 7to 10 feet above thelake level. (Belck in Globus, vol. lxv. p. 302). | ||
| 1875 | A maximum at about this period may be inferredfrom the accounts given by Bishop Poghos of Lim to Dr. Belck(Globus, vol. lxiv. p. 157), and by the Rev. Mr. Cole of Bitlisto Dr. Butyka (Globus vol. lxv. p. 73). From this period thereappears to have been a gradual relapse until 1892, and possiblylater. | ||||
| 1898 | Rise dating backseveral years is attested by Belck and Lehmann. The trees alluded toabove are now standing in the water (Zeitschrift fürEthnologie, 1898 p. 414). | ||||
| 1898 | Günther chronicles arise during the last two years on native evidence(J.R.G.S. November 1899, p. 510). | ||||
| 1898 | Evidence of Oswald and myself infers arise during the last few years. |
The same phenomenon of a rise in level was apparent on the margin of the large lake in the crater on Nimrud. There the brushwood, representing the growth of many years, was submerged; and much had already perished from want of sustenance. All the evidence points to the fact that such changes are of a temporary nature, and that a period of increase is followed by one of decline. The most probable explanation is that they are due to climatic conditions, which, it is well known, are variously operative over cycles of years. In the absence of any observatory in these countries this question is largely a matter of surmise or, at best, of inference. The existence of such periodical fluctuations may be regarded as having been established; it remains to consider the changes of a more permanent order.
We must not forget that at a period relatively recent in geological time this lake of Van was but a part of an extensive inland sea, which appears gradually to have become divided up into a series of basins. There can be little doubt that down to quite a late geological epoch no such barrier had been constituted between this basin and that of the plain of Mush, which immediately adjoins it upon the west. The waters have left their mark upon the rocky boundaries of that plain; and to their action I do not think we should err in attributing the peculiar appearance of the basal slopes of the Kerkür Dagh, where they face the great depression of Mush. To the same period perhaps belong several terraces which may be traced upon the bush-grown face of the southern coast of Lake Van between Garzik and the Güzel Dere. The highest of these is perhaps the most conspicuous, and may be situated at an elevation of a hundred feet or more above the present level. Just as the waters of the plain of Mush were drained away through a narrow opening in the mountains which hem it in upon the west, so it is quite likely that a similar vent was offered by the gorge which cuts through the parapet of Taurus in the direction of Bitlis, and at the present day affords an easy passage to the caravans from the plains of Armenia into the defiles of Kurdistan. Loftus chronicles a tradition that the waters of Lake Van cover a plain that was once studded with villages and gardens. The streams of Arjish and the Bendimahi Chai—and presumably the Khoshab—are said to have met and formed one large river about midway between Arjish and Bitlis. His informants were under the belief that it had issued from the plain through a hole in the earth; and that when this passage had been closed up by a sudden convulsion the present lake formed.[12] This story is at least not lacking in verisimilitude, so far as the existence of a former river is concerned. This river would have probably flowed to the Tigris, of which it would have been the principal branch. The cause of its being dammed up was perhaps the outpouring of lavas from Nimrud, which have formed the plateau between Tadvan and the head of the plain of Mush—a plateau which rises to a height of 680 feet above the lake, and, extending across from Nimrud to the face of Taurus in the south, chokes the entrance to the Bitlis gorge. It is this barrier which actually maintains the lake of Van. No eruptions on this scale are recorded during the historical period; and, of course, it is not impossible that they were originally submarine.
These phenomena, which are partly attested by the ancient lake terraces and in part suggested by the general structure of the country, belong to an epoch which, if quite modern from the standpoint of the geologist, probably lies beyond the range of the archæologist as well as of the historian. Much the same conditions as at the present day appear to have prevailed during the historical period—a vast sheet of water, deep and translucent, dammed up by the volcanic barrier at its westerly extremity. I think there can be no doubt that the permanent tendency of this sheet of water has been to rise in level. Moreover, all the evidence is to the effect that this tendency has been operative in the case of the other two seas. Dr. Belck has recorded that in the year 1890 during the month of July he came across a little lake at the eastern end of Lake Gökcheh, separated from it by a tongue of land scarcely more than 55 yards broad, and connected with it by a stream descending from the mountains and piercing through the isthmus. On the margin of this shallow lagoon, near the outflow of the stream, he discovered an ancient Armenian graveyard of which the stones were under water. When he returned in August of the following year they were only just dry. His visit coincided with the latest stage of a period of decline; and it seems certain that since the time when the cemetery was constituted the norm about which the fluctuations oscillate had risen in a marked degree. The same traveller draws our attention to the interesting circumstance that the three last lines of the cuneiform inscription of Rusas the First (c. 730–714 B.C.), cut in the face of the rock overlooking that same northern lake, have been almost completely destroyed by the erosion of the waters, although placed just above their level in 1891. It seems incredible that the Vannic king should have engraved his memorial in a situation where it would be exposed to the periodical floods.[13] As regards Lake Urmi I need only recall the important discovery of Mr. Günther in 1898. In the islands of that sea he found many species of living animals which could not have crossed the stretch of salt water, amounting to a distance of some 10 miles, that at present separates their homes from the shore. In his opinion the zoology affords conclusive testimony of these islands having been joined to the mainland at no very distant date. Upon one of them he found the skeleton of a wild sheep.[14] The evidence which may be collected upon the shores of Lake Van all points in the same direction of a progressive upward tendency.
Strecker has thrown out the suggestion that this process may be accountable for the junction of the Arjish arm to the main body; and that we may therefore attach some credence to the statements of Pliny that in his time there were two lakes.[15] However this may be, we are not dependent upon such hypotheses, or upon the stories current of submerged causeways or bridges. The three old fortresses of Akhlat, Adeljivas and Arjish all bear testimony to a considerable rise in the level of the lake since the days when they were built. The walls of the first two on the side of the water have either fallen in or are being slowly undermined. Arjish has been permanently abandoned by its inhabitants. Immemorial villages, like that of Kizvag between Akhlat and Tadvan, are being menaced by the latest periodical increase, which seems to have commenced about 1895. Nature herself speaks eloquently in the same sense. An ancient walnut-tree which stands on the rocky bank of the lake in the gardens of Erkizan, a quarter of Akhlat, had already been deprived of a great portion of its foothold when we encamped beneath its boughs in 1898. In the Sheikh Ora crater a giant mulberry, which may have been some 500 years old, was standing with half its roots in the water and was already doomed. The most obvious explanation of this gradual rise in the norm of the lake level is furnished by a cause, which must be constantly operative, namely the increase of sediment deposited upon the bottom. But whether this factor by itself be sufficient to have produced such important changes is a question upon which I am not qualified to pronounce an opinion.[16]