GEOGRAPHICAL

Previous

My purpose in the present chapter is to collect the threads of that part of the narrative which was occupied with the natural features, and to endeavour to weave them together into a composite but single fabric, capable of being appreciated as a whole. In the pursuit of this object I shall postulate familiarity on the part of my reader with the contents of the companion chapter dealing with the same subject which belongs to my first volume; and it is not without misgiving that I compare the scantiness of my present material with the multitude of facts with which the researches of Hermann Abich have enriched our knowledge of the Russian provinces. I am dependent almost entirely upon the gleanings of my own journeys and of those accomplished by my friends within quite recent years; and it has been impossible to commence the writing of this chapter before the completion of the map embodying these results. What it may, perhaps, be hoped without excessive presumption is that the framework, at least, of our subject, the geography of South-Western or Turkish Armenia, can now be established with some degree of certainty; and that succeeding travellers may be enabled to recognise at a glance the more imperfect parts instead of losing themselves in the almost unknown or falsely known.1

No better standpoint could be selected from which to commence a survey of the geography than the spine of that range whence we descended into Turkish territory during our journey southwards from Kagyzman (Vol. I. Ch. XX. p. 409, and Ch. XXI. p. 436). It carries the present frontier between the Russian and Turkish Empires, and in fact divides the area of Armenia into two parts. In a political sense it forms a boundary of considerable significance, shutting off Russia from the waters which issue in the Persian Gulf. More than once have her victorious armies flooded across this barrier, and not less often have they been compelled by the provisions of the ensuing peace to withdraw to its further side. The length of the range, its ruggedness and the relative height of the passes, compared with the plains on either flank, are features which must have operated throughout history to invest it with an importance unrivalled by the other systems which furrow the surface of the Armenian tableland. From the Kuseh Dagh (11,262 feet) in the west to Little Ararat (12,840 feet) in the east is a distance of nearly 100 miles; and throughout that space the chain is made up of such lofty peaks as the Ashakh Dagh (10,723 feet), Perli Dagh (10,647 feet), Sulakha Dagh (9644 feet) and Khama Dagh (11,018 feet). The passes reach from 7000 to 8500 feet; while the level of the plain of the Araxes does not exceed 3000 feet, nor that of the plain of Alashkert 5500 feet. In appearance the barrier as a whole resembles the mountains of the peripheral regions; there are the same deep valleys, jagged outline, precipitous slopes. It seems some daring invasion of those mountains into the plateau country; and the semblance is accentuated by the beds of marl along its northerly base into which the long transverse parapets plunge (Vol. I. Fig. 106, p. 419). Highly crystalline rocks, such as diabase, and even syenite, of which the spine of the more westerly portion is probably composed, have played the principal part in its configuration, where recent eruptive action has not built up a sequence of volcanic fabrics, such as Kuseh Dagh, Perli Dagh, the peaks about Lake BalÜk, the Great and the Little Ararat.

This range, to which collectively we may apply the name of Aghri Dagh or Ararat system, constitutes the principal intermediate line of elevation between the northern and the southern zones of peripheral mountains. It has been subjected to intense folding pressure, and during the process of bending over from an east-north-easterly to a south-easterly direction a partial fracture of the arc it describes has taken place. From the western shore of Lake BalÜk, an upland sheet of water lying at a level of 7389 feet, we are, perhaps, justified in tracing the extension of one branch of the system along the water-parting between the Murad and the Araxes south-east to the Tendurek Dagh, and through that volcano into the line of hills which divides the basin of Lake Van from the streams which find their way into the Araxes. Thence the elevation may be followed into the southern peripheral region, forming, as it were, a splinter from the chain of Zagros which has struggled upwards through the plateau country to its very heart. The prevalence of crystalline rocks, which have been classed by Loftus as granite, has been attested along the inner edge of Zagros all the way from near Khorremabad in Persia past Hamadan to the sources of the Great Zab; and they extend from the western borders of Lake Urmi at least as far as the district of Bayazid.2 It seems probable that they are in connection with the granite rocks of the Aghri Dagh, where they are found to the west of the Perli Dagh along the axis of this northern intermediate system.3

The more northerly and principal branch in an orographical sense would appear to consist almost exclusively of recent volcanic mountains, stretching from Perli Dagh in an east-south-easterly direction to the Pambukh Dagh, west of Great Ararat. In this neighbourhood the line is taken up by the fabric of Ararat, raising the barrier by slow stages to nearly 17,000 feet, and having an axis from north-west to south-east.4 The sequence comes to an end in the Little Ararat, whose slopes descend on three sides to fairly level plains. An interesting feature about the range in its more westerly portion are the outbreaks of andesitic lava along its base upon the north. These eruptions appear to have culminated in the peak of Takjaltu (8409 feet) near Kulpi, which forms a landmark to the districts on that side. Thence the fissure which gave issue to the andesite may be traced westwards, keeping parallel to the chain. The eruptions have disturbed the sedimentary rocks, and their incidence can be certainly attributed to the Miocene period.5 Further east the upwellings of lava along the slopes of the mountains have all the appearance of having been discharged into a sheet of water spread over the surface of the Ararat region.6

West of the Kuseh Dagh, the bell-shaped mountain, this intermediate line of elevation may be plainly followed upon the map along the southern confines of the plain of Pasin through the limestones which the Araxes threads in a landscape of savage grandeur before its entry upon the level expanse. From the left bank of the river the heights are continued for many a mile, until they are distinguished by the PalandÖken-Eyerli Dagh volcanic system (10,694 feet) just south of Erzerum. A slight inclination southwards through the Karakaya Dagh into the volcanic Keupek Dagh, and further south into the Khach Dagh, the southern boundary of the province of Terjan, takes the line with clear definition through the Girdim Dagh and the Baghir Dagh into the lofty and extensive barrier of the Merjan-Muzur Dagh (about 12,000 feet), facing the plains about Erzinjan. The progress of the elevation across the Euphrates through Asia Minor to the Mediterranean appears to be indicated on the map of Kiepert by the Sarichichek Dagh, west of Egin, whence it is probably protracted between the Taurus and the Anti-Taurus chains. The Anti-Taurus would appear to be represented in Armenia by the system which enters the country in the Chardaklu Dagh (long. 39, lat. 39.55), and extends in the form of an elevated block of tableland through the Sipikor Dagh, Dadian Dagh (11,000 feet), Kop Dagh into the DÜmlÜ Dagh, north of Erzerum, and the Chorokh region.

The importance of the orographical system which we have now traced from Ararat to Muzur Dagh, and from Lake BalÜk to the Zagros range, may be appreciated in a geographical sense by one or two reflections. In the first place it provides the natural frontier between the country about Lake Van and the Persian province of Azerbaijan. This frontier may probably be regarded as the natural eastern boundary of Armenia during its course from behind Bayazid to the Avrin Dagh, overlooking the valley of the river of Kotur. At the present day it forms the Turko-Persian border; while the more northerly branch, which effects a junction in the neighbourhood of Lake BalÜk, divides the Russian and Turkish Empires. As the most pronounced constituent of the Asiatic structural design within the limits of the tableland, the system carries over the Tauric lines of elevation into those which have determined the configuration of the Iranian highlands. It encompasses this result in a most impressive manner, standing up from the plateau region with precipitous slopes on either side and suggesting to the mind the conception of a backbone to the country as a whole. It is at this point that in the Shatin or Aghri Dagh it effects the bend over into Persia, but not without partial fracture and consequent dislocation. At the same time we should be mistaken in attributing to the system functions analogous to those of the mountains of the peripheral regions. Even the Aghri Dagh is deprived of many of the qualities essential to a barrier by its narrowness and by the extension of the open plains on either flank. The border between the Lake Van basin and Azerbaijan consists of a line of hills rather than of mountains in the proper sense. The extension of the elevation along the southern confines of the plains of Pasin and of Erzerum takes the form of the lofty rim of the central region of the tableland, and not of a mountain range. That term might, perhaps, be applied to the cretaceous heights of the Merjan-Muzur Dagh; but these again are probably due to the resistance of the Dersim block, the plateau-like country which they limit upon the north.

I have already traced the course of the mountains of the northern peripheral region, the effective barrier between Armenia and the coast of the Black Sea, throughout their prolongation upon the confines of the tableland, and have drawn the natural frontier inwards in the neighbourhood of Ispir across the valley of the Chorokh to the northern border heights of the plain of Erzerum (Vol. I. Ch. XXI. p. 431). The analogous zone upon the south is composed by the main chain of Taurus, separating the highlands from the low-lying plains of Mesopotamia and buttressing them up on that side. This chain appears to have succeeded in accomplishing the curve into the Iranian direction without undergoing fracture to any material extent. The symmetry of the arc described as seen from the plains about Diarbekr has already enlisted our admiration (ibid. p. 424). The spine of the range may be followed along the southern shore of Lake GÖljik to the Palu Dagh, east of the town of Palu. Thence it is taken along the plain of Chabakchur and the left bank of the Murad to the confines of the plain of Mush. Conspicuous with sharp peaks which are seldom free from snow, it stretches past the depression of Mush into the landscape of Lake Van, where it recalls the sombreness of the Norwegian coast. Through the Karkar Dagh (long. 42.47), and, further east, through the Bashit Dagh, west of Bashkala, it makes steps southwards to the threshold of the basin of the Great Zab; and the elevation may be traced on the further side of the river in the peaks of the Jelu Dagh, said to attain a height of between 13,000 and 14,000 feet.7

An impressive feature of this Taurus range, and one which ought not to escape the attention whether of geographers or of political students, is the manner in which it appears to have sunk down along its southern edge between the 39th and 42nd degrees of longitude. In places the girdle of mountains becomes so narrow that its effectiveness as a barrier is much impaired. From the town of Arghana, which must lie almost at the southern foot of the chain, it is a direct distance of not more than 28 miles to the confines of the plains about Kharput. These may be attained from Diarbekr on the lowlands without encountering a greater altitude than less than 5000 feet. The position of the town of Haini (2800 feet) appears to correspond to that of Arghana; and thence the Murad may be reached in 22 miles direct by a pass of only 4200 feet. In such a climate heights like these are quite insignificant, and they would not offer at any season an obstacle of much importance to an army operating from the lowlands in the direction of the Armenian plains. This sinking-down of Taurus has been accompanied, as indeed one might expect, by volcanic action on a considerable scale. The Karaja Dagh, which lies to the south-west of Diarbekr, is not a mountain of much relative height. You may ride at a trot across its long-drawn undulations, admiring the sea-like expanse of the plains around. Yet it represents an extensive outpouring of lavas in recent geological times. It would appear to be in connection with some of the greatest of Armenian volcanoes, and with a string of depressions extending across the plateau. The line may be easily recognised through Nimrud and Sipan to Tendurek and Ararat.

With the exception of the Dersim block, lying to the south of the Merjan-Muzur Dagh, which has not yet been satisfactorily explored, the remaining lines of elevation within the limits of the tableland are probably for the most part derived from the Taurus system. In this connection it is most interesting to take due note of the phenomenon that, side by side with the results of the later earth movements which have most largely determined the existing configuration of the land, an older movement may be discerned with a wide extension in Turkish Armenia, rearing mountains along a south-west—north-east line. We ourselves remarked this phenomenon on an impressive scale in the Akh Dagh, an elevation of highly marmorised limestone, which may well be older even than the Cretaceous period. It rises up on the north of the plain of Khinis (Ch. VIII. p. 186, Fig. 159), which it confines in an east-south-easterly direction. Though we were unable to test the strike of the stratification, the appearance of the ridges of which it is composed almost demanded the conclusion that they were originally members of a series of heights with a north-easterly course. Even as far east as the region to the south-west of Lake Van, where the Taurus is pursuing a general trend towards east-south-east, the strike of the older rocks was ascertained to be north-east. A glance at the map will show that the heights which confine the course of the Gunek Su pursue a north-easterly direction. Those on the right bank, extending to the basin of the Kighi or Peri Su, may be clearly traced into the Taurus on the west of Palu, to be represented further south by the Chembek Dagh and Mastikan Dagh, constituents of Taurus to the south-west of Kharput. In the opposite direction the line may not unreasonably be regarded as extending beneath the volcanic accumulations of the BingÖl Dagh through the Akh Dagh into the hills confining the plain of Alashkert upon the south, known as the Mergemir or Khalias Dagh. The younger movements may find expression in the present trend of the two last-named systems, and, further south, in the KÖshmÜr Dagh, Shaitan Dagh and Javresh Dagh, mountains through which the Kighi Su breaks in a narrow defile after leaving the Khindris Ova or plain. These last extend with impressive orographical distinction to the south-western edge of the BingÖl plateau.

The KÖshmÜr Dagh effects a junction with the mountains of the Dersim; and it would almost seem as if that region had refused to submit to the folding pressure, causing the earth waves to work round it and, like the plateau of Azerbaijan, on the east of Armenia, favouring fracture rather than subordination in any complete sense to the general structural laws.8 Yet I cannot doubt that the Dersim should be included within the limits of the country which forms the subject of the present enquiry. The name appears to be applied more strictly to the mountainous region lying to the east of the upper reaches of the Muzur Su, between that river and the town of Kighi Kasaba. But it may be used to embrace also the country to the south of the Merjan-Muzur Dagh, as far west as the great bend of the Western Euphrates and up to the right bank of the Murad on the south. Separated from the important Turkish military station at Erzinjan by a range of mountains covered with snow during six months in the year, it slopes gradually towards the river on its southern confines, well wooded in many parts, abounding in minerals, but broken and rugged especially in the northern and eastern districts. The original home of an Armenian population, who probably entered their historical seats from the west, it is dotted over with the ruins of Armenian churches, monasteries and villages, and is mainly but sparsely inhabited by Kizilbash Kurds.9 The natural boundary between Armenia and Asia Minor is the course of the Western Euphrates between the town of Kemakh, the burial-place of the Armenian Arsakid kings, and its passage through Taurus below Keban-Maden. North of the Euphrates the line may be drawn in a more or less arbitrary manner from above Egin to the mountains of the northern peripheral region.

The boundary of Taurus is clearly defined from one end of Armenia to the other, describing a symmetrical curve along the threshold of the Armenian highlands, and affording a number of standpoints whence the contrast may be appreciated between the plateau country and the peripheral mountains. A string of great plains extend on its inner or northern side, but plains quite different in character from the lowlands about Diarbekr, and framed in a landscape never wanting in the long-drawn outlines of the loftier levels. The plain of Kharput, with an altitude of something over 3000 feet, commences the series on the west. It is reached from the west and the south by a number of easy approaches, the Tauric barrier being readily surmountable in this neighbourhood. The town is built upon a hill, not far south of the Murad, on the northern confines of the plain; and the old castle overlooks the expanse at a difference in level of about 1000 feet. Various estimates assign a population of from 13,000 to 25,000 souls to this ancient Armenian borough; and, although the Armenians are in great minority in the city, they have a large preponderance among the inhabitants of the surrounding region. It has been estimated that not less than from 130 to 150 villages are situated in the vicinity. The vine flourishes and is cultivated at this moderate elevation; and the dwellings are for the most part constructed of mud and brick with two storeys, in striking contrast to the unhealthy underground burrows in which the peasantry cheat the rigour of an Armenian winter over the greater portion of the area of the tableland. Pear and plum trees grace the outskirts of the settlements, and the mulberry grows in such profusion that the silk crop is often of considerable value. Kharput has become a centre of American missionary effort—on the whole a salutary and civilising influence in these lands. Their educational activities are represented by a well-equipped institution founded in 1876 and bearing the name of Armenia College. Thither flock the Armenian youth from all parts of the country, to grow up beneath the example of the most progressive of Western peoples. Within recent years the value of that example has somewhat diminished in their eyes, owing to the impunity with which the organisers of Palace policy in Constantinople have applied the torch to the property of American citizens and the ban of the censor to the loftiest creations of Western literature. These are little misunderstandings which will disappear.

A fairly level country extends from the territory of Kharput eastwards to the confines of Palu. The Murad wanders in many channels over the expanse, approached at an interval which is always diminishing by the Tauric barrier. The river is forded to the right bank before the castled rock is reached, past which it flows in a single stream. It washes on three sides the steep declivities of the platform upon which the town is built. Palu is described to me as a thriving borough with about 2000 houses, which gives a population of from 10,000 to 12,000 souls. Six hundred are said to belong to Armenian families and the remainder to Kurdish people. A bridge with eight arches and a length of 190 yards connects the place, just to the east of the loop described by the river, with the opposite or left bank. On the north extends a plain in connection with that of Kharput and productive of abundant crops. Rock chambers and a cuneiform inscription of the Vannic king, Menuas, recording his conquests and emblazoned with the name of Khaldis, his supreme god, remind the traveller that he is already approaching the centres of that old civilisation which existed before the Armenians, and was perhaps the highest that these lands have known.10

From Palu a fair track leads through Temran into the Khindris plain, and thence to Erzerum. The passage of the Shaitan Dagh into the plain may be effected at different points, but the pass to Lichig has an elevation of over 8000 feet. The Government are proposing to carry their new carriage-road between Kharput and Erzerum through the gorge of the Kighi Su. Two routes are offered between Palu and the next great plain at the foot of Taurus, comprised within the territory of Chabakchur. The upper route proceeds through Khoshmat to Chevelik in the valley of the Gunek Su, crossing a mountainous region and attaining elevations of 6000 to 7000 feet above the sea.11 The lower follows the gorge of the Murad and is more in use during winter, but it is described in no very favourable terms. The narrows commence just east of Palu and extend to Chabakchur. A waterfall brings to an end the navigation of the river, which is conducted with no small difficulties by means of rafts. When at length the plain is reached communications become better; and the valley of the Gunek Su affords an easy approach to Erzerum, though one which would not be agreeable during winter. Between Chabakchur and the plain of Mush the Murad is again confined in a gorge, and its course is still requiring to be explored. A mule track is forthcoming, which keeps close to the river, passing through the district of Genj. I am informed that it would be impossible to convert into a good road. The more usual route is by Menaskut, entering the plain below Surb Karapet after traversing a mountainous but well-wooded country.

One may say in general terms of the extensive region we are now leaving that the pleasant plains along its southern margin are by no means the dominant feature. The territory lying between the two great rivers, the Western and the Eastern Euphrates, which is bounded on the north by the Merjan-Muzur Dagh with its continuation eastwards in the Baghir Dagh, Girdim Dagh and Khach Dagh; and on the east by the westerly edge of the BingÖl plateau and the water-parting from BingÖl to PalandÖken, the mountain landmark just south of Erzerum—all this area, measuring some 140 miles from west to east and on the average 50 miles from north to south, is intersected by a sea of mountains, threaded, indeed, by considerable streams, but always difficult and in winter almost impossible to cross. The constant acclivity towards the north, the height of the barrier on its northern confines, and the indifference of the approaches from the east combine to shut it off from the stream of human movement, which is diverted into other channels. We can scarcely understand the history of these countries without appreciating this fact. Be the movement from east to west or from north to south, the main current is sure to eddy along the outskirts of this territory, either pursuing the broad avenue of the valley of the Western Euphrates, or turning aside from the plateau region and flooding across the peripheral mountains into the lowlands of Mesopotamia. As might be expected under such conditions, the country is for the most part under little control. Strange people who are classed as Kurds, but speak a dialect called Zaza, and for the most part profess a liberal religion which holds the scales between Christianity and Islam, compose the bulk of the inhabitants in the mountainous parts. The Government works from the upland plains, of which there are many and of ample extent, and from such centres as Kighi Kasaba and PÜlÜmer. If my reader will turn to my sketch from the hill of Gugoghlan (Fig. 195, p. 373), he may realise at a glance the rugged nature of this region and the contrast which it offers to the normal surface features. It comprises the ridges in the west and south-west of the panorama; and the Merghuk Su, which meanders towards them, is the name of the head waters of the Kighi Su, issuing in the plains about Kharput. Travellers praise the woodlands which clothe great parts of the country, though they were not visible from the standpoint of my drawing.

The next great plain at the foot of Taurus derives its name from the town of Mush, built against the wall of the range. It extends from north-west to south-east for a distance of over 40 miles, crossed at its lower end by the wandering stream of the Murad, to which it sends a dull and almost stagnant tributary. How clean the line of Taurus stands out on the southern margin of this flat and almost limitless expanse! Under happier human conditions the plain would soon become a garden and granary, favouring the vine and the luscious growth of the tobacco plant as well as all kinds of cereals. At the present day marshes extend over a great part of the area, and the Armenian peasantry—one of the brawniest and most sturdy in the world—have been reduced by the excesses of the Kurds to abject indigence. Mush is in communication at all seasons of the year with the great grain-growing districts of Bulanik and Khinis, and with the plains of Pasin and Erzerum. It is little more than a step—indeed a step in the literal sense—up to the fertile territories on the north of Lake Van. Ready access is always forthcoming through the Bitlis passage to the Mesopotamian lowlands. The Mush plain represents a considerable subsidence of the plateau region, the average elevation being only 4200 feet. Thence you pass across the dam formed by lavas from Nimrud to the much higher level of Lake Van (5637 feet). That inland sea, with the gulf-like extension of the even area up the valley of the Khoshab, the district of Hayotz-dzor, forms the appropriate termination of the string of level spaces outspread at the base of the chain which comes from the Mediterranean during its passage along Armenian soil.

In the companion chapter of the first volume I have endeavoured to suggest the characteristics of the mountains of the northern peripheral region. The corresponding zone upon the south which is occupied by Taurus is distinguished by many similar features. There are the same sharp peaks, precipitous slopes, narrow valleys and swift streams and rivers, composing a landscape which, except for the greater scale of the phenomena, is essentially and constantly alpine in character. Unlike our Alps but like the barrier on the side of the Black Sea, one valley is ever higher than the trough which lies behind it, each crest more lofty than the last, as you journey towards the edge of the tableland whether from the coast of the northern waters or from the alluvial flats which extend to the Persian Gulf. Of moisture there is less among these southern mountains, and we miss the exuberance of the Pontic vegetation. But forests of dwarf oak relieve the sternness of the scenery, and the knots or whorls on the trunks of the numerous walnut trees sustain an industry which attracts the most adventurous of native traders, causes them to sojourn in these wild districts, and enables them to supply the markets of Europe with excellent material for veneering purposes. The summits attain their greatest elevation in the Jelu Dagh, a group of peaks just east of the valley of the Great Zab which are at least as high as 13,000 feet. But by the time the summer is well advanced the landscape is almost free from snow. West of the Zab a labyrinth of valleys feed the long course of the Bohtan Su across the mountainous belt. The barrier has more than trebled in lateral extension since confining the territories of Kharput and Palu. Even the Tigris, which has been idly spreading over the vast alluvial flats about Diarbekr, is compelled to become a mountain stream. Above the primeval village of Hasan Keif it enters the narrow gorge which pierces the foot of Taurus as he reaches out into the plains in the hill range of Midyat. It is in that gorge that the Bohtan effects the confluence; and well I remember the roar of the tributary and the genuflexions of my companions as the swirling water eddied around our raft. The Tigris is henceforward a noble river at all seasons, and when Jezireh is soon passed its brief activity is over and it luxuriates in open spaces till reaching the Gulf.

All this alpine country between the edge of the tableland and the plains of Mesopotamia, which is watered by the numerous constituents of the Tigris, is the original and natural home of the Kurdish people, the true Kurdistan or Kurd-land. These shepherds love the mountains as the Arabs affect the plains; but they need the warm plains during the winter season when their fastnesses are covered with snow. They descend to the foot of the chain with their numerous flocks and herds, and camp on the lower course of some southward-flowing tributary or even upon the banks of the great river. The winter climate of the lowlands is temperate and delicious; the long Kurd with his loose limbs, hollow cheeks and beak nose meets the neat and nimble Arab. The coarse but perky little highland horse is watered from the same flood to which the Arab leads the graceful creature prized beyond all other possessions, of skin like satin, limbs like ivory, and head which is the supreme embodiment of high courage, docility and intelligence. The noiseless raft surprises a group of Kurdish women bathing quite nude upon the margin of the sandy bed. A man is watching over them, and they seem without concern. Two specks are descried upon the bosom of the waters; the current brings them nearer; they are swimmers from the opposite bank with the chest supported on an inflated skin. Within a few yards of your calek they emerge upon the bank, Arab maidens who would delight a sculptor with their slim forms resembling deer, and who have never learnt the sin of human nakedness. Slowly they free the air from the buoyant skins, unbind the bundle on their heads containing their loose cotton garment, and make their way to an invisible village or encampment.

When summer comes the annual migration to the recesses of the mountains has taken place, and whatever Kurds are not detained in the lowland villages, which are fairly numerous in spite of the aversion of the tribal Kurd to a life within walls, have already ended their brief sojourn in the country of the Arabs and are stretching their goat-hair tents upon the upland pastures. Streams of cattle, sheep, horses and goats obstruct the passes; the shepherds have doffed their felt cloaks and clamber over the boulders, their women beside them, mounted or on foot. The glades and gorges become bright with red and blue cottons, and Kurdish girls with comely faces and white ankles are seen on the mountain paths. Long-drawn shouts are carried far across the hiss of the torrents, and wurra, wurra! or ho, ho! announce the locality of the speaker or awake the attention of callous ears. The Kurd is a picturesque and welcome presence among these solitudes, and it is only when he has been severed from his natural surroundings that he becomes odious and an enemy of the human race.

Of the principal communications across Taurus with the tableland of Armenia I have already glanced at those connecting Diarbekr with Kharput and Erzerum through Arghana, and with Erzerum through Haini. The latter is a direct and, in spite of the great elevation of the country which it traverses between the plain of Altun and the northern capital, nevertheless a promising route. Mush plain may be reached from Diarbekr by way of Kulp and the Gozme Gedik Pass (6645 feet). But between this approach and the Bitlis passage the country is ill-controlled, nor am I aware of any favourable and beaten tracks. The Bitlis passage represents the main avenue between the lowlands and the country about Lake Van (Ch. VI. p. 148); from Diarbekr it is entered by way of Zokh and from Mosul through Sert. I shall not stay to discuss the various more or less direct routes across the mountains between Sert and the city of Van; nor can I speak from personal knowledge or even conjecture of those which conduct to Van from Jezireh-ibn-Omar, or from Mosul by the valley of the Great Zab.12 The entire region to the south and south-east of the great lake—Khizan, Mukus, Shatakh, Nurduz—has been scoured in recent years by various travellers whose experiences have not to my knowledge as yet appeared in print.13 I should now propose to dismiss this part of my subject, dealing with the zones of peripheral mountains and the intermediate lines of elevation upon the surface of the tableland which they enclose; and to bring under review some of the remaining features characteristic of the Armenian highlands in their westerly extension from the spine of the Ararat system to the confines of Asia Minor.

Hitherto our study of the orography of this Tauric Armenia has been mainly occupied—it is interesting to recall the fact—with lines of folding of the earth’s crust. Indeed the country as a whole has not been subjected to recent volcanic action in the same degree as the plateau regions lying to the north of the spinal mountains—the territories of Akhaltsykh, Ardahan, Akhalkalaki, Alexandropol and Kars. At the same time it has not escaped the operation of these agencies; nor have they worked upon a less impressive scale. Be it lavas flooding over the sedimentary deposits and levelling the inequalities of the ground—what more startling manifestation could be offered of the process than the BingÖl plateau with its piled-up layers of lava and tuff? Or if volcanoes in the strict sense be matched against volcanoes, there are Nimrud and Sipan to enter the lists with AlagÖz and Ararat. Several mountains which are due to eruptive action have been added to the map in the course of my own journeys. Such are Bilejan and Kartevin. The roll will be increased as our knowledge is carried further of the districts on the west of BingÖl and PalandÖken.

A striking analogy in some respects to the Russian territories which I have just specified is provided by the surface features of the BingÖl plateau, with its continuation northwards in the shape of a deeply eroded block of land to the confines of the plains of Erzerum and Pasin. This extensive region lies about south-west of the corresponding area of rectangular shape within the Russian frontier. It performs the same function of a roof to the adjacent countries; and just as the one stage gives birth to the Kur and the Arpa Chai, so the other feeds with countless channels the earliest course of the Araxes and contributes the largest proportion of the waters of the Murad. The streams which decline from its north-westerly extremities swell the volume of the Western Euphrates. Built up on the south with lavas and tuffs to the extent of thousands of feet, it has throughout been flooded with volcanic matter. Taken in relation with the general structure of Tauric or Turkish Armenia, we may apply to this elevated stage of the plateau country the designation of the Central Tableland.

My reader is already familiar with the characteristics of the region—the basin-like appearance, the long parapets on the northern and southern edges, in the one case culminating in the volcanic peaks of PalandÖken (10,694 feet) and Eyerli, in the other distinguished by the eminences of BingÖl (nearly 10,800 feet). The limits of the BingÖl plateau are clearly defined on three sides, and may readily be recognised on our map. On the north it merges insensibly into the Shushar and Tekman districts, though at some points, as, for example, the cliffs just south of Kherbesor, lines of demarcation may be laid down. How the waters of this plateau converge together in the shape of two fans, as they are precipitated from the highest levels towards the north and towards the east, burying themselves ever deeper into the volcanic soil! The one group is collected in the plain of Khinis, and the other by the course of the Araxes between the plain of Altun and the narrows on the north of Kulli. There in the hollow of the basin the levels are still lofty—the Altun plain with about 7000 feet and Kulli with about 6000. Ascend to the table surface from the beds of the rivers, and you register heights which range between 7000 and at least 9000 feet above the sea. A country with down-like outlines, composed of limestones with intrusive serpentines and Pliocene lake deposits capped by sheets of the ubiquitous lava—an expanse sterile and vast at all seasons, and in winter covered with snow—a softly billowing surface dappled by the shadows of cumulus clouds and shot with colour from a network of blue streams—such, I think, are the most permanent impressions of our journeys across the Central Tableland.

Volcanic action is largely responsible for the configuration of this tract of country, filling up hollows, preserving the sedimentary deposits with overlying sheets of lava. The extent of the operation may best be gauged on the south-western extremities of the BingÖl plateau. There the ridges in the west are seen stepping up, one after another, almost to the margin of the elevated platform where your tents are spread. The setting sun invests them with an added glamour of gold and purple; yet how futile this fretful array against the solid land about you, dimly spread in horizontal spaces beyond sight! The yellow mullein which scents the air springs from the ruin of all those ridges, growing upon the tomb of their deeply-buried remains. But further north, where the sway of the lavas has already become feeble, the same phenomenon, a little modified, may be observed. Survey the scene as it is unfolded northwards from the western summit of BingÖl or from the hill of Gugoghlan (Ch. XXII. p. 373, Figs. 194 and 195). What a contrast between the landscape of the west and that of the east! All those ridges in the west are dying by themselves into the down-like spaces of the Central Tableland. Here the lavas have been a contributing but not the principal cause.

The truth is that we should here be standing quite near the point of greatest constriction between the inner and outer arcs. In other words, it is just west of this region that the greatest compression of the Armenian highlands by earth movements may be supposed to have taken place. A natural consequence of the process would be the ridging up within a narrow space of the normal surface elevations. East of an imaginary line between BingÖl and PalandÖken the area becomes enlarged. Room is given for the ridges to spread; they flatten out and almost disappear. At the same time the change from the Tauric into the Iranian direction soon commences to make itself felt. Mountain and gentle hill, the rocks on the heights and those in the hollows are all imprinted with the stamp of a new-born force. In the most central districts we recorded this change in what geologists call the strike between the villages of Kanjean and Alkhes in the region called Elmali Dere or Vale of Apples. There the stratified rocks have been flooded with sheets of lava, which have presumably welled up from fissures. A glance at the map will show that all the outlines are bending over, those on the north-east and those to the south-west of this point. And a little looking brings home the fact that most of the great Armenian volcanoes are situated at or near the bend.

The tendency to a strong-pronounced plateau country is in Armenia, and especially in the south-western territories, independent of volcanic action. Hermann Abich aptly describes the effect of this tendency upon the mountain masses when he speaks of their constant, nearly horizontal summit line.14 Yet the heights which elicited this appreciation belong to the system west of BingÖl, and are mainly composed of stratified rocks. Horizontality is the prevailing characteristic of the outlines on the north of the series of plains from Pasin in the east to Erzinjan in the west. Those outlines belong to a block of elevated land from over 9000 to about 8000 feet above the sea. Lavas have accentuated the feature in the case of the border heights of Pasin (Ch. VIII. p. 193, Fig. 163); but when, further west, the barrier consists of limestones and old igneous rocks, the same appearance of a flat-topped mass, representing a higher stage of the plateau region, is only varied by some beautiful shapes emerging upon the sky-line, such as the Cretaceous peaks of Akhbaba and Jejen. If you draw a section between the western extremity of the plain of Mush against Taurus and the maze of valleys which feed the Chorokh on the north of Erzerum, the true character of the land will be exhibited in a striking manner. You will commence with a level plain of immense extent from west to east and with an average elevation of 4200 feet. Proceeding northwards, you scale a wall of 8000 feet, only to find yourself upon a platform almost as flat as a billiard-table, over which the track leads without much change in level for a distance of many miles. This stage breaks off upon the north to a little plain even as water, lying in the lap of an extensive depression of not more than 5000 feet. You cross the depression with a parapet of 8400 to over 9000 feet closing the landscape with gigantic cliffs before your eyes. It is the edge of the Central Tableland. The journey is long from this, its southern margin, to the corresponding rim upon the north—water-worn downs with an average altitude of over 7000 feet. After registering heights, always on the level, of about 9000 feet, a descent is made to the vast expanse of the Erzerum plain (5700 feet). The mass which rises on the north of that plain contains the sources of the Western Euphrates and leads over to the deep valleys which sustain the Chorokh. It is flat-topped, and attains a level of about 9000 feet.

The most fertile and agricultural districts lie to the east of this section; they are generally separated one from another by mountains of recent volcanic origin, upon which, however, with the possible exception of the Tendurek Dagh, a wreath of smoke is never seen. The plain of Khinis (5500 feet) is screened by Khamur from the plains of Bulanik and Melazkert (5000 feet), where some of the finest grain in the world is grown. Bulanik is divided into a western and an eastern territory by the radial volcanic mass of Bilejan. The line of heights which are interposed between Western Bulanik and Mush plain are probably partly due to lavas which have welled up from fissures, and are easily crossed almost at any point. The plain of Mush (4200 feet) and the level country of almost endless extent between Sipan and the Murad are shut off from the cornfields and orchards of the basin of Lake Van (5637 feet) by the immense circumference of the Nimrud crater and by the block of limestones and lake deposits upon which Sipan is built up. The region between Lake Van and the hills of the Persian border is parcelled out into a number of districts by such volcanic eminences as Varag Dagh, Pir Reshid Dagh,15 and Tendurek Dagh, which last-named mountain has sent its lavas a great distance south into the Abagha Plain.16 All the way from Tendurek to the plain of Khinis eruptive agencies have fastened upon the land on a considerable scale. A large area is occupied by the radial volcanic system known as the Ala Dagh, but very scantily explored. It is succeeded further west by the Kartevin Dagh. The extensive territories between Kartevin on the south, the plain of Khinis on the west, and the Sharian-Mergemir Dagh barrier on the north, are for the most part covered with sheets of lava. But the plains of Alashkert (5500 feet) and Pasin (over 5000 feet) are worthy to rank with the most favoured regions; and this sequence is continued westwards by the plains of the Western Euphrates, commencing with that of Erzerum (from 5750 to about 3800 feet). North again of this series one may specially instance the plain of Baiburt (5000 feet), which is a typical Armenian plain.

As you travel from plain to plain, from one basin to another, the horizon is most often filled by some shapely volcanic outline, slowly rising from the floor of the expanse. Yet the stratified rocks are seldom absent, emerging from the volcanic layers or only capped by a thin sheet of lava. Dominant among them are the limestones of various geological periods, from the Cretaceous and probably earlier, to the Pliocene deposits, when the greater part of the country must have been covered by a lake of fresh or brackish water. Intrusive in the earlier limestones are found a variety of old igneous rocks, such as diabase, gabbro and serpentine. The serpentines combine with the limestones to form rounded hills or downs with soft outlines. Sometimes a cap of lava has preserved a particular piece of limestone, and the result has been a summit with a point like that of a needle overtopping adjacent and undulating forms. Where the old igneous rock occurs in a zone, a sombre landscape is forthcoming, as for instance above the northern shore of Lake Van between Akhlat and Adeljivas. Or when the highly marmorised older limestones have the upper hand, there ensue sterility and glaring light. These latter rocks have a fairly wide extension and compose prominent lines of mountain. For example, they have bestowed upon the plain of Khinis its northern boundary; and nowhere are they seen to greater advantage than in that shining and richly modelled barrier appropriately named the Akh Dagh or White Mountain. During the journey from Gopal to Tutakh on the Upper Murad they were constantly emerging from the sheets of lava; and in the south we found them in the vicinity of the southern peripheral mountains. They alternate with mica-schist in the Elmali Dere and GÜzel Dere, districts at the south-western extremity of Lake Van. And they stretch across the water to form the promontory of Tadvan.

A rather later series of limestones would appear to be represented by the slopes over which we climbed to the Vavuk Pass between GÜmÜshkhaneh and Baiburt. There they are placed on the very threshold of the Armenian tableland; and they are distributed in a wide zone over the northern districts of Armenia, extending all the way from the Merjan-Muzur Dagh in the west to be represented by many a summit of the deeply eroded Chorokh region. The block of heights on the north of the Western Euphrates is composed to a great extent of such limestones; and both in the neighbourhood of the Kop Pass, and during the descent northwards from the pass of Khoshab Punar, we have been able to identify them by the evidence of fossils as belonging to the Cretaceous period. The several startling eminences from the surface of this elevated stage—a surface which is characterised by prevailing flatness and horizontality of the summit-line—are mostly due to upstanding masses of limestone, such as Akhbaba and Jejen. In the south we recognised the fossils of this same series of rocks upon the line of hills which border upon the north the great depression of the plain of Mush, where these give passage to the Murad.

Later still in date, and of almost constant prominence in the landscapes both of the plateau region and of the peripheral mountains, are the limestones of Eocene age. They are, perhaps, more usually associated with softer features, especially when they are interbedded with shales. Writing from memory, one may best recall the incidence of their impressive features at such widely distant points as the PalandÖken line of heights, on the south of Erzerum and Pasin, and where they whiten the waters of Lake Van in the neighbourhood of Adeljivas. This pretty town with sweet-sounding name lies at the foot of a lofty cliff composed exclusively of white chalk. As you lunch in one of the caves along the road from Akhlat, numerous corals are observed imbedded in the rock. Even where volcanic action has fastened upon such heights with greatest persistency, the white face of this rock or of the softer Pliocene deposit is seldom absent from the scene. Eocene limestones and Pliocene deposits are prominent over the area of the Central Tableland; and the limestone emerges on the further side of the plain of Khinis to compose the Zirnek Dagh, continuing the outline of Khamur. The almost limitless expanse through which the Murad winds between Tutakh and Melazkert reveals most clearly its essential character as a country of rolling chalk downs beneath the covering of a cloak of lava. The southern limit of that expanse would seem to the eye to be volcanic, misled by the precedent of the immense extension of the train of Ararat. But when the barrier is at length reached it is found to consist of Eocene and Pliocene limestones, forming a pedestal for the fabric of Sipan.

Scarcely a less prominent surface feature are the Pliocene lacustrine deposits,17 crumbling in the hand with masses of freshwater shells. There can be no doubt that at an epoch contemporaneous with the outpouring of lavas a lake or lakes extended from Erzinjan, Erzerum and Pasin across the region now occupied by the Central Tableland, and through Khinis to the plains of the Murad and Sipan. The interior of Asia Minor and the tableland of Persia were covered with lakes at the same date; but that these were salt in the case of Persia is proved by the melancholy saline deserts which disfigure immense tracts of the soil of Iran. In Armenia they have been productive of the greatest fertility, their wholesome sediments having mingled with volcanic matter and become constituent of rich brown loams. It seems likely that the purple sandstones and conglomerates along the northern shore of Lake Van are the representatives of similar conditions within that basin. One is justified in supposing that the waters became gradually more shallow, until they remained only on the surface of the numerous greater and smaller depressions, which still bear their imprint to a degree which must be convincing even to an unpractised eye. A chain of separate lakes was formed, spread broadcast over the land, and washing the promontories of the heights. Such lakes appear to have existed at Alexandropol and in the plain of Erivan; over Pasin, the plain of Erzerum, and that of Erzinjan; in the districts of Khinis, Alashkert, Bulanik and probably Mush, to say nothing of the smaller sheets of water. They were drained away as a result of the increasing elevation of the land as a whole; and, probably, in some cases the process was accelerated by uptilt, causing erosion of the adjacent barriers to be accelerated. The lakes which exist at the present day are almost exclusively due to lavas filling up the mouths of valleys and forming dams on an immense scale.

The relation of geology to geography must always be intimate; and in such a country as Armenia it is scarcely possible to travel without becoming absorbed in the open book of that fascinating study, as day by day the eye is greeted by a new page. The architectural quality of the structural features is perhaps the main incentive, stimulating the curiosity to comprehend the underlying design. But the absence of wood and the sparseness even of vegetation permit and invite the interest to centre in the forms and hues and texture of the material which has been the vehicle of the large idea. Nature is revealed in her sculpturesque rather than picturesque beauties; nor will her admirer regret the nakedness of his love. But the climate suffers from the prevailing treelessness of the landscapes, being deficient in moisture for the most favourable development of the human race. One feels the skin growing contracted as in most Eastern countries, and the native sappiness of the flesh becoming impaired. There is no reason why this country should not be strewn with woodlands, and her plains verdant with a kinder rainfall and extended irrigation. Patches of forest, but thin and miserable, still struggle towards the interior from the luscious zone in the north. They are seen on the sides of the passes at a distance from the villages. But with the exception of the very thinly populated districts of Kighi and the Dersim, and the slopes of the Soghanlu mountains south-west of Kars, the land has been denuded of any covering as a result of progressive economical decline. Centuries of unchecked licence on the part of tribal shepherds—Tartars, Turkomans, Kurds—have brought about the destruction of a source of salubriousness and wealth which under any circumstances would require careful husbanding.

So the clouds are little tempted to descend upon the earth, and the sky lowers without bringing rain. The country streams with light, and the pavements of ubiquitous lava burn like an oven beneath the untempered rays of the sun. In winter the glare is blinding; for the ground is covered with snow, though not generally to any great depth. These are disadvantages which are not entirely without remedy; and there is nothing needed but less perversity on the part of the human animal to convert Armenia into an almost ideal nursery of his race. The strong highland air, the rigorous but bracing winters, and the summers when the nights are always cool; a southern sun, great rivers, immense tracts of agricultural soil, an abundance of minerals—such blessings and subtle properties are calculated to develop the fibre in man, foster with material sufficiency the growth of his winged mind and cause it to expand like a flower in a generous light. One feels that for various reasons quite outside inherent qualities this land has never enjoyed at any period of history the fulness of opportunity. And one awaits her future with an expectant interest.

Both branches of the Euphrates wind their way by immense stages at the foot of these mountains, in the lap of these plains. The eastern branch, called Murad, contains the greater volume, rising in the neighbourhood of Diadin near the base of the Ararat system and traversing Armenia almost from one extremity to the other. The principal affluents are the BingÖl Su, bringing the drainage of the plain of Khinis; the Gunek Su, and the combined waters of the Kighi and Muzur rivers. The more westerly channel is composed in its infancy by two streams of almost equal size, one descending from the DÜmlÜ Dagh and flowing sluggishly through the plain of Erzerum; the other, and perhaps the greater, springing in the neighbourhood of the sources of the Chorokh in the elevated district of Ovajik. The Kelkid and Chorokh are both in their upper courses typical Armenian rivers. The Araxes takes its birth upon the Central Tableland, and its true source is probably represented by the little lake which appears in my drawing from the western summit of BingÖl (Ch. XXII. Fig. 194, p. 373). What a contrast between this wealth of waters, many of which might be rendered navigable, and the hopeless sterility of great parts of the interior of Persia, from which no river finds its way to the ocean!

All these rivers wind slowly and silently over the surface of the tableland, threading landscapes which most often expand beyond the range of sight. They find a tardy issue through the zones of peripheral mountains, where they meet the hiss of torrents and the spray of waterfalls. When one reflects with closed eyes upon the experiences of travel it is not the dividing heights that fill the mind. What are these for the most part but the higher stages of the plateau country? It is the plains, great and small, with their lake-like or sea-like surfaces; and it is the ever-present feature of the volcanic outlines, spaced at large intervals. The streams part on their course to widely distant oceans from a scarcely perceptible rise in the ground. Earth is spread about you, nude and quite unconscious of the restless presence of man. A variety of delicate and transparent tints are shed over the modelling, due to the atmosphere and the volcanic nature of the soil. The hues deepen in the blue ribands of the flowing waters, in the gem-like appearance of those that are still. And when the vision has nearly faded there remain the shapes of Ararat and Sipan, the campagna of Erivan, the ineffable beauty of the lake of Van.... The area of the country which has been delimited within the Turkish frontier measures 35,599 square miles. If we add this figure to the Russian territory (Vol. I. p. 445) we may conceive a geographical unity nearly equal in extent to England and Wales.


1 I must not omit to record the assistance which I have received from the map of H. Kiepert, Provinces Asiatiques de l’Empire Ottoman. The sheets which cover the Armenian country embody the results of my predecessors, which have been compiled with great judgment. I have also had access to two Russian maps embracing portions of the country, (1) scale 10 versts = one inch, 1889, (2) scale 20 versts = one inch, 1899. But the map of Kiepert with all its merit is necessarily sketchy; and the last Russian map is flagrantly incorrect.?

2 See the map of Loftus in Quarterly Journal of the Geological Society, vol. xi. London, 1855, p. 247.?

3 See the map of Abich in Geologische Forschungen in den kaukasischen LÄndern, Vienna, 1882, Atlas, Karte I.; and part ii. p. 141.?

4 A fine view of the range at this point is displayed by Abich, op. cit., Atlas table iii.?

5 Abich, op. cit. part ii. p. 155.?

6 Abich, op. cit. part ii. p. 160.?

7 E. Clayton, The Mountains of Kurdistan, in the Alpine Journal, 1887.?

8 For some account of the geology of Azerbaijan see C. Grewingk, Die geognostischen und orographischen VerhÄltnisse des nÖrdlichen Persiens, St. Petersburg, 1853.?

9 The best account of this country is that of J. G. Taylor, J.R.G.S. vol. xxxviii. 1868. I may also refer my reader to two articles by Dr. Butyka (Mitt. der K. K. geographischen Gesellschaft in Wien, vol. xxxv. 1892, pp. 99–126 and 194–210), who has collected the scanty notices of his predecessors and added his own experiences. I have made use of some unpublished material in the preparation of this part of my map; but it is far from satisfactory.?

10 See the inscription and translation by Professor Sayce in the Journal of the Royal Asiatic Society, vol. xiv. 1882, No. XXXIII. p. 558.?

11 Tozer, Turkish Armenia and Eastern Asia Minor, London, 1881, pp. 256 seq.?

12 H. Binder describes the route from Mosul through Amadia, Julamerik, Kochannes and Mervanen to Van (Au Kurdistan, en MÉsopotamie et en Perse, Paris, 1887). See also W. F. Ainsworth, Travels in Asia Minor, etc., London, 1842, vol. ii. pp. 179 seq., with geological section from Mosul to Lake Urmi.?

13 For the geology of the Taurus between Diarbekr and Kharput an article by W. Warington Smyth in the Quarterly Journal of the Geological Society of London, 1845, vol. i. pp. 330 seq., may be consulted. I do not understand his statement that the breadth of the main ridge of Taurus between Arghana and Kharput is nearly 50 miles. Loftus (op. cit. p. 344) has drawn a geological section from Bitlis through Sert to Jezireh-ibn-Omar. Like the Zagros, the range may be said to consist in the main of nummulitic limestone.?

14 Abich, op. cit. part ii. p. 119 note.?

15 I cannot speak with certainty as to the geological nature of the Pir Reshid Dagh.?

16 For Tendurek, which appears to be in a solfataric condition, see Abich’s article in the Bulletin of the French Geological Society, 2nd series, xxi. pp. 213 seq., and Letter from T. K. Lynch in P.R.G.S. xiii. pp. 243, 244.?

17 The Miocene deposits are found in the valleys, e.g. in those of the Frat (Western Euphrates) and Araxes.

An interesting fact has been brought to my notice by Mr. F. Oswald, my friend and companion during my last journey. There may be seen in the Tiflis Museum the remains of a mammoth which was discovered in the lacustrine deposits of the Alexandropol district. Similar remains had already been found in deposits of similar character and age in the neighbourhood of Khinis by Colonel J. Shiel. These are in the British Museum, where they have been christened Elephas armeniacus.?

                                                                                                                                                                                                                                                                                                           

Clyx.com


Top of Page
Top of Page