In the present chapter I shall invite my reader to make good his advantage over the traveller, and to realise, before proceeding further with the journey, the true meaning and wider connection of those natural features which have composed the landscape day by day. At the same time I shall endeavour to trace the limits of north-eastern or Russian Armenia, extending our view for awhile to comprise the whole of Armenia, and again narrowing it to the area of the Russian provinces.
But at the outset we are prompted to examine the conception so vaguely expressed by the metaphors of tableland and frame of mountain ranges which, with slight variations in the figure, have in the foregoing pages so often been employed. The pursuit of this analysis carries us beyond the sphere of our particular survey, compelling us to consider the structure of Asia as a whole.
From the Mediterranean to the Pacific the Asiatic continent is traversed by a zone of elevated country, which, flanked on the north and south by great chains of mountains, breaks off on the west to the Ægean Sea and to the lowlands of China on the east. Extensive areas of land with considerably lesser altitude are outspread on either side of this gigantic system: in the north the plains of Russia and Siberia, in the south the peninsulas of Arabia and India. The mountain chains which confine the zone of elevated country have been reared during different geological periods; yet they are subject to common laws. They are disposed in extensive arcs, of greater or lesser curvature, which are festooned across the continent on either side of the plateau region with a general direction from east to west. The plateau region is in general synclinal or, in other words, of slightly hollow surface, and, in comparison with the flanking ranges, is flat.
If we enquire of the geologist the origin of these phenomena, we receive an answer which, while it leaves many points obscure and doubtful, still enables us to trace the operation of fixed principles in the mighty work unfolded before our eyes. Our globe sails through the wan expanse of Æther, diffusing the heat with which it is charged. The cooling crust shrinks and gathers inwards towards the centre; but the material of which it consists is inelastic and is thrown into gigantic wrinkles or folds. Radial contraction induces tangential stresses at the surface, colossal forces which bend over and invert the folds, and even thrust the strata one beneath another, causing them to be disposed like the tiles upon a roof. This lateral tension finds most relief where the crust is weakest; and it is at such points, or along such zones, that the process of mountain-making has been developed on the largest scale. It is the tendency of such folded ranges to form arcs of large curvature, which are drawn inwards, where the lateral pressure meets with most resistance, and expand outwards, where it is withstood in a lesser degree.
In Asia the operation of this process of mountain-making has been accompanied by, or has produced, the elevation in mass of large portions of the earth’s crust. The intensely folded regions, or, in other words, the great chains of mountains, are found along the inner and the outer margins of the elevated mass. Between these zones the stratified rocks have no doubt been subjected to the folding process; yet they have escaped the immense contortions that have taken place on either side.
Throughout the continent the lateral force which has been most operative in mountain-making has proceeded from the north. The fact may perhaps be explained by supposing that this force is the result of the active pressure exerted by the hard, unyielding material of which the steppes of Siberia and the basin of the Arctic Ocean are composed. The great arcs which are described by the mountain ranges are in general convex to the south. Thus in western Asia the chains on the inner and outer margins of the elevated area are disposed in two roughly parallel series of arcs bulging towards the south. Of these series the inner arcs have less curvature than the outer, to which they are roughly parallel.
The inner series may be traced with greatest singleness of feature on the west of Hindu Kush—that natural centre of the mountain systems of Asia which at once supplies the most convenient standpoint for a general survey of the structure of the continent, and is placed at the junction of the two great divisions, western and eastern, into which geographers have partitioned this vast area. The Hindu Kush inclines over into the Paropamisus; and the southern portion of the latter range is continued, on the north of Persia, by the mountains of Khorasan. A sharp bend in the belt, just east of the Caspian, turns southwards into the Elburz range, and the beautiful curve of the chain along the margin of the shore may be admired from the waters of that inland sea. The line of Elburz is protracted across the depression of the Araxes valley into the peaks of Karabagh; while the Karabagh system unites with the bold and lofty ridges which in full face of their gigantic neighbour, the Caucasus, overtower the right bank of the Kur. These ridges again connect with the chain we have ourselves crossed between Kutais and Akhaltsykh—a chain which joins the mountains on the southern shore of the Black Sea. The Pontic range forms a bow of wide span and gentle curvature, ending in the hump of Anatolia, where it meets the arc of the Bithynian border hills.
The parallel series on the outer margin of the elevated area commences with the outer arc of the Hindu Kush system, the severely bent and S-shaped Salt Range. Thence it proceeds into the mountains which flank Persia upon the east and belong to the outer Iranian arc.1 The bold sweep of this arc into the chain of Zagros may be recognised by a glance at the map. We remark the greater protraction of the north-western arm of the bow, a feature which may be traced in the configuration of most of the great Asiatic chains. We admire the clean and uniform outline of the curve, broken only by a slight indent at the straits of Ormuz, which may be answered by the bend in the inner system which we have already noticed on the east of the Caspian Sea. The outer Iranian arc effects a junction with the Tauric ranges along two parallel but fairly distinct orographical lines. Of these the inner line crosses over from the Zagros to the Ararat system, and assumes commanding orographical importance in the western arm of that system, known as the Aghri or Shatin Dagh. It is in the Shatin Dagh that the bend to the west-south-west is effected, which may be followed through a series of volcanoes into the Anti-Taurus and the Mediterranean range. The outer line is formed by the grand half-circle of the Kurdish mountains; from the parched plains about Diarbekr you see them, as from the well of an amphitheatre, covered or capped with gleaming snow. This principal chain of Taurus extends to the coast of Syria, and emerges from the sea in the island of Cyprus and in many a headland and island of the Anatolian coast.
It can scarcely fail to impress the most casual of observers that this double series of arcs, from Hindu Kush to Mediterranean, meet or almost meet at three distinctly traceable and widely separated points. Such approximations occur in Hindu Kush, in Armenia, and in the mountainous districts which border the Ionian seaboard. We can scarcely doubt that they are due to the incidence of a strong opposing force, moving from the south and causing the arcs to be constricted, the ranges to be piled up one behind another, and mountain development to assume its grandest forms. It is probable that the resisting pressure has been furnished in the first two cases by the Indian and Arabian peninsulas. Another feature, less obvious but not less noteworthy, is furnished by the fact that in Armenia and Asia Minor the arcs have been fractured in the process of bending over at or near the points where the approximations between the two series have taken place. The closer the constriction, the sharper, of course, becomes the curve, and the greater the tendency to split. In Asia Minor the union of the series has resulted in complete fracture; the folded area sinks beneath the waters of the Ægean to be represented by the islands which stud the Archipelago, and, further west, by the mountains of the Dalmatian coast.
On the east of Hindu Kush we are as yet in want of sufficient material for so convincing an analysis as the researches of geologists have rendered possible on the west. We know that in eastern Asia a vast area of elevated land is bounded both along the inner and the outer margins by mountain systems of wide extension and great height. Such are the systems of Altai and Tian-shan upon the north, and the mighty bow of the Himalayas on the south. Probably the Kuenlun range carries over the inner series of western Asia, extending eastwards from the Pamirs and serving as a buttress to the immensely elevated plateau of Tibet. If this view be correct, then the Tian-shan and Altai systems may perhaps be regarded as minor earth-waves, following close upon the heels of the Kuenlun, and supporting the highlands of the Tarim basin and the desert of Gobi, the Han-hai or Dry Sea of the Chinese. The plain reader may be content to observe the echelon of mountain ranges which extends from Hindu Kush towards Behring Sea; to note the constant curvature of the arcs towards the south, until, in the Altai group, the eastern arms of the bows are protracted ever further towards the north; to contrast the low-lying plains along the western ends of the echelon with the lofty highlands of Mongolia on the east. The necks of the valleys issue upon the depression of Siberia and the low country through which the Oxus and Jaxartes flow.
In western Asia the elevated area with its flanking ranges is bordered on the north by the northern Paropamisus and further west by the Caucasus chain. The Paropamisus may perhaps be regarded as the most southerly of the many branches which belong to the system of Tian-shan.2 Geologists invite us to connect the Paropamisus with the Caucasus, and trace the links of the broken chain to the mountains of Krasnovodsk on the Caspian, whence a submarine ridge carries the line into the mountains of Caucasus, to be protracted far to the west, through the Crimea, and emerge from the waters of the Black Sea in the Balkans, Carpathians and Alps. In this manner we see described on the north of the Asiatic highlands, with their series of inner arcs, a further arc of immense span and wide curvature, which is represented on the east by the northern Paropamisus and by the Caucasus on the west. Both these ranges may best be viewed as independent of the inner series; but Paropamisus is closely adpressed to the inner arc of Persia, and Caucasus is joined at a single point to the series, namely by the Meschic linking chain. Lines of elevation, similar to that which we have traced from Paropamisus, may be discovered, although with less orographical distinction, proceeding westwards and struggling over towards Europe from the more northerly branches of Tian-shan; they are almost lost in the great depression of the Turanian lowlands, but they follow arcs of increasing width of span.
This interesting study of the structure of Asia, which is due to the researches of recent years, not only serves to explain the pronounced features of Asiatic landscapes, as integral members of a vast design, but also enables us to understand many of the movements of history and many of the phenomena of the human world.3 India is enclosed on all sides by the sea or by the outer mountains, and appears reserved by natural causes for herself. China, with her teeming millions, is separated from western Asia by the whole bulk of the broadest and least hospitable portion of the system of lofty plateaux with peripheral ranges. The echelon of chains, which seam the continent in a north-easterly direction, are the nurseries of the hardiest tribes. The valleys which space these ranges are the arteries of human movement, and they lead from west to east, from east to west. Thus during the period of armed migrations which is represented by the Tartar conquests, one division of the Tartar armies might be fighting in China on the Yellow River while another was laying waste Khorasan. The bend of the arcs towards the south places the framework of Nature in harmony with the migrations of man. The tablelands of Persia, Armenia and Asia Minor are members of a continuous system of elevated plains at a temperate altitude, which extend like some great causeway along the breadth of Asia, giving access from east to west, from west to east. This causeway forms the natural avenue of commerce and of conquest, by which the tide of war or of commercial intercourse ebbs and flows between the remote recesses of Central Asia and the Ionian shore of the Mediterranean Sea. Only on the east is the causeway blocked by Nature to human traffic, by the constriction of the arcs on the north of India, leading over by a gigantic knot of mountains into the impassable plateau of Tibet. The stream is therefore diverted from the highlands to the lowlands; great cities arise on the lowlands, at the mouths of the Tian-shan valleys, Merv, Bokhara, Samarkand. And when we contemplate and contrast the structure of Asia and of Europe—the vast forces which have produced the stately body of eastern Asia dying out towards the west in the insignificant but widely ramified elevations of the European mountain chains—we may readily understand how different has been the influence exercised by structural features upon the peoples of either continent. In Asia such features are a factor of the first importance, determining climate, controlling migrations, setting barriers to intercourse or relentlessly fixing the highways which it must pursue. In Europe, on the other hand, they have done little more than diversify the scenery, and for purposes of peaceful or hostile movements among the nations may with some exceptions be almost left out of account. What are our European mountains but arbitrary wrinkles on the face of the continent? One valley leads over into another of much the same height above sea-level by a pass which is not more lofty than the neighbouring ridges. One plain is succeeded by a companion expanse of similar character, and only some small diversity in the forms of the spires of the churches tells the tale of national distinctions. Differentiation rather than the presence of marked ethnological types is characteristic of the peoples of Europe. But once the narrow strait is passed we may no longer dally with our geography; and the further we proceed towards the east and the inner sanctuaries of Nature the greater grows the necessity of comprehending phenomena which must always exercise a dominant influence upon human affairs. It will not suffice in Asia to observe the latitude of a great plain in order to know beforehand the degree of heat which it will support in summer, the rigour or the suavity of the climate during winter. You will be freezing in Erzerum while Erivan is relaxed in sunshine; yet both cities are placed on the margins of level expanses, and the advantage of latitude is in favour of the temperateness of that first named. Not even the convenient distinction of highlands and lowlands will carry us very far. We must enquire into the nature of the highlands; are the mountains their prevailing feature, or are those mountains, as we see them from the floor of the lowlands, a mere buttress to a sequence of elevated plains? Penetrate the chain, and you rise by successive steps from valley to valley, while each ridge is higher than the last. Follow its extension upon the map and you will see it rising from the Mediterranean and terminating in the knot of mountains north of India. Mark the characteristics of the people who inhabit it, be they Kurds or Lurs or Lazes, they will not offer much divergence from a common standard. Yet what a gulf of human nature between these and the inhabitants of the lowlands—a gulf which is scarcely spanned by the equalising tendencies of a long spell of misgovernment! When at length these alps expand, and you overlook a more level country, everything—climate, the aspect of the sky as well as of the land, people, language, cities, villages are new. And yet our diplomatists who dwell on the Bosphorus, and ruminate Asiatic problems with the aid of indifferent maps which they would not pretend to understand, group the highlands and the lowlands, the shepherds of the mountains and the cultivators of the plains, all together—a strange collection of birds and beasts and fishes—in a single scheme of administrative reforms. The Turk is little wiser; but we may perhaps view with a large indifference his passive resistance to such reforms.
But to return to our plains and mountains—the country which we may still call Armenia takes its place as an integral member of the system of tablelands, buttressed by mountain ranges, which extends from the Hindu Kush to the Mediterranean Sea. It is not separated by any important natural frontier from Persia on the east or from Asia Minor on the west. Moreover most of the characteristics which are found in either of these neighbours are prevalent in Armenia to a greater or a lesser degree. The stratified rocks include the later PalÆozoic, the Cretaceous, Eocene and Miocene series; and these extend across the whole system. The salt deposits of Miocene age which are spread so widely over Persia are not among the least remarkable of the surface features of Armenia; although they have not produced that widespread devastation which attends the extension of the great salt deserts over the Persian plateau.4 In Armenia they are friendly to man, providing him with one of his necessaries; and the various salt works, known in Turkey under the name of tuzla or salt pans, have been exploited from immemorial times. Considerable depressions of the surface of the highlands are phenomena common to all three countries; and the same may be said of the volcanoes which are dominant in Armenian landscapes, but are not wholly absent from the contiguous territories on either side. All participate in the benefits of a southern climate, and are exempted by their elevation above sea-level from the excesses of a southern sun. Slowly-flowing rivers threading vast plains, mountains which determine districts rather than states; a natural penury of vegetation, enhanced by the depredations of countless goats, but perhaps balanced in the eyes of the traveller by the beauty of the land-forms—such are some among the many impressions which may be derived in various degrees from a visit to any of the individual members of the group.
But, if Armenia be closely linked with her neighbours on the west and east, she is divided by some of the most effective of natural barriers and natural distinctions from the countries which lie to the north and south. The zones of mountains which on the one side separate her from the coast of the Black Sea and the Georgian depression, and on the other from the lowlands of Mesopotamia, possess in an equal degree the rugged character due to intense folding and are both of considerable width. Sharp ridges with serrated outlines rising one behind another, narrow valleys in which the shadows lie, hissing rivers and bush-grown rocks, grassy uplands or stretches of forest determine the scenery both of the northern and of the southern zone. The alpine region has a breadth of some fifty miles more or less in the direction of the Black Sea, while the corresponding zone, facing the lowlands about Diarbekr, extends, on the whole, over a smaller span. Both zones are practically unlimited in length. They have been factors of paramount influence in the history of the peoples, not only screening the territories they confine from those which lie outside, but also investing them with distinct climatic conditions. For these parallel belts of peripheral mountains do in fact perform the function of supports or buttresses to a series of elevated plains; the valleys in the alpine region are but the succession of terraces which rise to the margin of a lofty platform. A difference in level of several thousands of feet is productive of marked features in the habits and character of the inhabitants; while the alps themselves must necessarily determine the mode of life of the dwellers within them, constraining them to follow the vocation of shepherds rather than that of agriculturists. Thus along the section between Diarbekr and the Armenian highlands three strongly-contrasted types of people will be met. The nomad Arabs or Arabic-speaking cultivators of the lowlands are succeeded by the pastoral Kurds with their tribal organisation, and these again by the Armenian tillers of the soil.
I have already indicated the intimate connection of these peripheral mountains with the structural system of the Asiatic continent. The northerly belt belongs to the inner series of arcs, and that on the south to the outer series. The compression of these arcs—a phenomenon which has engaged our attention—has been effected in the greatest degree within the section of country between Diarbekr and Trebizond. You see the two opposite arcs, one bent to the south and the other to the north, endeavouring to meet under the stress of contending pressures; while on either side of the section the curves diminish in intensity and the spines of the ranges have been allowed to expand like the spokes of a wheel. The northern boundary of Armenia is constituted by the mountains of the northern peripheral region, which enter the country on the west in the Gumbet Dagh. The line may be followed on the map on the north of Shabin Karahisar through the Giaour Dagh and the Kuseh Dagh to the pass over the Vavuk Dagh, lying to the north-west of the town of Baiburt. From the Vavuk pass the spine of the chain confines the valley of the Chorokh by a well-defined and regular parapet; until just east of the town of Ispir it commences to lose this singleness of feature, and to favour a tendency towards bifurcation and branching out. The ridges stretch across the valley in an east-north-easterly direction, the direction which the spine has so long pursued; and their course may be traced through the mountainous country on the north of Olti until they become buried beneath the volcanic accumulations of the plateau country in the districts of GÖleh and Ardahan. It is most interesting to trace their probable emergence from this canopy on the further side of the tableland, and to recognise in the elevations of Shishtapa (north of Alexandropol) and of Madatapa ridges that have survived the splitting and fracture of the Pontic chain. But this is a feature of greater interest to the geologist than to the geographer; and the latter will follow the Black Sea range through the heights of the Khachkar and Parkhal mountains to the Kukurt Dagh on the west of Artvin. The ridge which stretches thence in a north-north-easterly direction towards the seaboard, giving passage to the Chorokh and determining the Russian frontier, has been deflected by the mass of the Karchkhal mountains, the radial system to the north-east of Artvin. It crosses the river close to the coast behind Batum, and may be traced through the peaks of Taginaura, Gotimeria and Nepiszkaro along the plains of Imeritia to the passage of the Kur through the gorge of Borjom. These last-named peaks belong to the Akhaltsykh-Imeritian border range, which my reader has crossed with me by the pass of Zikar, and of which the direction is almost due east and west.
It is impossible to delimit the northern frontier of Armenia by a slavish insistence upon the boundary of the Black Sea range. That system is the natural boundary for a distance of very many miles, as it extends along the course first of the Kelkid Su, the ancient Lycus, and then along that of the Chorokh. But the fracture of the arc which has taken place in the country watered by the uppermost branches of the Kur and Arpa Chai, and the eating back of the more easterly affluents of the Chorokh, which have carved out the intricate country in the neighbourhood of Olti, have resulted in the interruption of the normal sequence until it is again resumed in the Akhaltsykh-Imeritian range. It is consonant with the natural conditions to take the frontier across the valley of the Chorokh in the vicinity of Ispir, and to lead it by the heights which contain the sources of the Chorokh and the Serchemeh Chai to the DÜmlÜ Dagh, the parent mountain of the Western Euphrates. It will then follow, first in an easterly and then in a north-easterly direction, the elevated water-parting between the basins of the Araxes and the Black Sea; and, after effecting a union through the Chamar Dagh with the volcanoes of the Soghanlu Dagh, will be protracted along the meridional and volcanic elevation which confines the highlands of GÖleh and Ardahan on the west. The junction of these vaulted heights with the Akhaltsykh-Imeritian range may be traced through the ridge of the Sakulaperdi Dagh to the peak of Gotimeria. All the rivers on the northern slopes of this section of the Armenian frontier drain into the Black Sea.
The passes across this zone are of considerable elevation, though a good number are open all the year round. I have been unable to ascertain the height of the pass over the Gumbet Dagh between Karahisar and Kerasun. But the valleys of the Upper Kelkid and the Upper Chorokh may be reached from Trebizond without encountering a greater altitude than something less than 7000 feet. To this figure must be added another 600 to 1000 feet before the traveller will have crossed the block of elevated tableland interposed between those valleys and the great Armenian cities, Erzinjan and Erzerum. East of Baiburt the spine of the Pontic range becomes more lofty: and the track which leads from Rizeh to Ispir in the Chorokh valley surmounts it at a height which has been estimated at 9000 feet above the sea. Where the frontier has become coterminous with the northern border heights of Erzerum and Pasin the roads are taken by passes of over 7000 feet (Erzerum-Bar-Olti) and 8500 feet (Hasan Kala-Olti) into the basin of the Black Sea; while during its protraction northwards through the Soghanlu Dagh to the Sakulaperdi Dagh it may be traversed by well-beaten paths or tolerable roads at elevations which range between 6085 feet (Eshak-Meidan Pass) and about 7000 feet. The principal avenues of communication across the mountainous region are those of Erzinjan-GÜmÜshkhaneh, Baiburt-GÜmÜshkhaneh, Erzerum-Olti, Kars-Olti, Ardahan-Olti and Ardahan-Ardanuch. A road has been constructed from Kutais to Abastuman, and is gaining traffic every year.
Copious rainfall and abundant vegetation are characteristic of the northern peripheral mountains. In some of the valleys the clouds settle for several months in the year, seldom lifting to disclose a view of the sun. It may often happen that during several weeks or even months crests and depressions alike will be shrouded in mist. In summer there is produced the likeness of a succession of forcing houses, the slopes and hollows being covered with a bewildering tangle of trees and creepers and scarcely passable undergrowth. From the branches are festooned the lichens, grey-white streamers like human hair; the crimson stools of a fungus shine out from the gloomy brakes, and the pointed pink petals of the Kolchian crocus clothe each respite of open ground. Such conditions are most prevalent in the narrow valleys near the Pontic coast, while the slopes which face the Rion and the opposite Caucasus are distinguished by magnificent forests. Several peoples, distributed over fairly distinct zones, inhabit these fastnesses. On the west we have the Greeks, inclined to commerce and close to a seaboard; they may be found struggling upwards to the spine of the range and even in a sporadic manner upon its southern slopes. Further east dwell the Lazis, a wild people; and their neighbours, the Ajars, in the mountains behind Batum. These are succeeded by a population of Georgian shepherds and small cultivators, whose picturesque chalets are surrounded with Indian corn.
It remains to follow the extension of the mountains of the northern border during their progress eastwards from the Borjom gorge. The comparative narrowness of the belt in the neighbourhood of that great cleft is explained by the fracture of the arc to the south of this region and the covering up of its more southerly members by volcanic emissions. But this decrease in width is to some extent balanced by the propinquity of the Caucasus. It is in this neighbourhood that the single link connecting the belt with Caucasus stretches across the Georgian depression, dividing the Rion from the Kur; it may be known as the Meschic linking chain. East of this barrier the vegetation diminishes in luxuriance. The Akhaltsykh-Imeritian range is continued beyond the gorge by the latitudinal Trialethian chain—a system of which the backbone is formed by the Arjevan ridge, and which is bounded on three sides by the course of the Kur. A branch of this system is seen to continue the direction of the Pontic range, inclining off at a sharp angle from the principal elevation to form the valley of the Gujaretis. It culminates in the peaks of the Sanislo group at an extreme height of 9350 feet, and sinks beneath the lavas of the plateau region. The Trialethian mountains have undergone a process of uptilt, which has caused them to fall away abruptly towards the north and to form terraces of plateau-like character on the south.
Just as on the west we were constrained to draw the natural frontier inwards from the spine of the Pontic range, so on the east the next successors of the Trialethian ridges lie outside the proper boundary of the Armenian plains. A glance at the map will show that a dislocation of the natural features has taken place in this region. The inner arc, so clearly defined on the one side by the Pontic chain and on the other by the Shah Dagh, overlooking Lake GÖkcheh, has snapped during the process of bending over; and the survivors of the catastrophe, the ridges which obstruct the Khram and the Somketian mountains, are constrained to play a subordinate part. The water-parting and principal elevation is composed of volcanoes, reared in a meridional direction. What an impressive analogy to the phenomena on the side of the Black Sea! These volcanoes pursue two lines, one line close behind the other, and the outer or more easterly far the longer of the two. It is the outer series, known as the Gori Mokri, or wet mountains, that constitute the border of the Armenian highlands on this side. The traveller who journeys westwards from the plateau of Zalka (5000 feet) up the elevated valley of the river Kzia to the little upland plain of the same name (7000 feet)5 will be treading on the dividing line between the folded mountains of the Trialethian system and the meridional volcanic series. On his left hand he will admire the shapely cone of Tawkoteli (9211 feet), which constitutes the most northerly of these volcanic elevations. The barrier is continued southwards through the Samsar Dagh (10,770 feet) to the Daly Dagh; and thence along the eastern shore of the lonely lake of Toporovan (6875 feet) to the dual crown of Agrikar (9765 feet) and to the conical summit of the Emlekli Dagh (10,016 feet). The sequence ends in the heights of Karakach (over 10,000 feet), of which the southerly extension is interrupted by the latitudinal ridges of Aglagan and Shishtapa. But the border is protracted along the parting of the waters into the westerly extremities of the Pambak chain.
We may, perhaps, regard this chain as the most southerly of the latitudinal ridges which begin on the north with the Akhaltsykh-Imeritian and Trialethian systems. It extends the area of the highlands for some distance towards the east, when, after commencing to incline in an east-south-easterly direction, it effects a junction with the Shah Dagh. This last-named ridge takes the frontier along the eastern shore of Lake GÖkcheh to the confines of Karabagh; and the elevation may be traced through the spine of the northern Karabagh mountains across the Kur to the range which faces the Caspian Sea. But Karabagh may be regarded as a separate geographical unit, combining in miniature many of the characteristics of the Armenian highlands—an inner plateau region flanked by peripheral ranges. The immemorial home of Armenian inhabitants, the seat of Tartar immigrants and the happy hunting-ground of nomad Kurds, it constitutes a solid outer buttress to Armenia on the side of the Caspian.6 The true boundary must be taken southwards from the Ginal Dagh (over 11,000 feet) to the Kety Dagh, where it forms a loop towards the west; and, after almost encircling an upland sheet of water, called the Ala GÖl, is protracted through the heights of Sir-er-syrchaly (11,298 feet) and Salvarty (10,422 feet) to the valley of the Araxes at Migry just east of Ordubad. The Karadagh mountains on the southern bank of the river continue the ridges of Karabagh; and the natural frontier is pushed westwards up the course of the Araxes as far as the village of Julfa. From this point you have the choice of two methods of demarcation, both of which repose on geographical facts. The line may be taken south-eastwards along the marginal ridge of the Karadagh to the water-parting between the basin of the Araxes on the one side and that of Lake Urmi on the other. This parting is of little orographical relief, but it would conduct the frontier almost in a straight line to the serried ridges of the southern peripheral zone on the south of Lake Van.7 Or the more pronounced bulwark between the Lake Van and Lower Araxes basins may seem to constitute the true boundary of the Armenian country. In this case an arbitrary line must be drawn from behind Bayazid, leading from the crest of these mountains, which at present constitute the Turko-Persian frontier to our original starting-point, Julfa. My reader will observe that we have left the barrier of the northern peripheral mountains, to explore the less certain limits on the side of Persia.
We have now pursued the northern border of the Armenian highlands from the coast of the Black Sea to that of the Caspian, where the belt passes over into the mountains framing Persia upon the north to be protracted into the Hindu Kush. The corresponding southern zone is much more simple of feature; but it lies outside the province of the present chapter, being included, throughout its entire extension along these highlands, within Turkish territory. Between the northern and southern zones of peripheral mountains several distinct but minor members of the orographical system we have been examining furrow the surface of the tableland. These will receive their proper attention in the companion chapter of the second volume, situated as they are for the most part beyond the limits of our present survey. But one of them may be traced to the commanding elevation which determines the valley of the Araxes during its passage through Chaldiran to the confluence of the Arpa Chai; and it is this range—for it deserves to be described as a range—that not only constitutes the present frontier between the Russian and Turkish Empires, but in fact divides the area of Armenia into two parts. You must either cross the spine of this chain, which describes a symmetrical curve, or follow along the plains at its northern or southern flanks, should you desire to pass from the plateau region on the north and east to the corresponding districts on the south and west. In the preceding chapter we have become familiar with some of its interesting features; and we have been introduced to it under the general name of the Ararat system or Aghri Dagh. Shatin Dagh is another name under which its westerly portion is designated by some writers, and which is scarcely less well qualified to express its ruggedness. This range carries the natural frontier between the two divisions from the Kuseh Dagh (11,262 feet) in the west to Little Ararat (12,840 feet) in the east.
It will thus be seen that the present area of Russian Armenia corresponds in a remarkable manner with the limits assigned by Nature to the more north-easterly of the two extensive regions into which she has parcelled Armenian soil. The Russian frontier is drawn from the coast of the Black Sea along the water-parting of the tributaries to the western bank of the Lower Chorokh through the peripheral region, and west of the town of Olti, to the Armenian border at the Chakhar Dagh. Thence it is taken across the Araxes to the spine of the Aghri or Shatin Dagh just north-west of the dome of Kuseh Dagh. It follows the spine of the range to the neighbourhood of Great Ararat, whose hallowed summit it embraces within the dominions of the Tsar. From the crest of the Little Ararat, whose south-eastern slopes are left to Persia, it reaches across the plain to the right bank of the Araxes a little below the famous monastery of Khor Virap. The Araxes forms the boundary between the Russian and Persian Empires from this point to near its confluence with the Kur.
It is a misleading, nay, a false conception of natural features to distribute the surface of the plateau region into a number of distinct geographical units. That is a method which is favoured by Russian sciolists with political connections in their endeavour to confuse the essential unity of a country which Russia has not yet fully absorbed. Enter this region where you will and with the eyes of any qualified traveller, the same or similar impressive characteristics will at once appeal to the mind. The German scientist Koch has well described these idiosyncrasies as they may be observed from the marginal districts on the west. After a long and laborious climb from the valley of Ardanuch (1800 feet) to the summit of the pass which leads to Ardahan (at least 7000 feet), he was astonished to observe that instead of a rounded ridge, descending with more or less abruptness to lower levels on the further side, the elevation upon which he stood was continued towards the east by the gentle slope of a lofty plateau. “Here was the commencement,” he says, “of the plateau which slopes away from the pass, and which is usually called the Armenian plateau.” The same traveller journeyed back into the Chorokh region from the highlands of GÖleh on the south of Ardahan. On this occasion he crossed the water-parting at the Kanly Dagh between Ardahan and Olti. He tells us that it consists of a narrow ridge with red, porphyritic rocks. He describes the double prospect from the summit, with its contrast of forms and impressions. On the one side, towards the Kur, a scarcely perceptible incline, forming upland valleys after a descent of only some 1500 feet, and leading over to vague and vaulted heights. On the other, in the direction of Olti, rent mountains, gaping ravines—nowhere a gentle, convex shape. Where he was placed the climate was raw, even in early September, and scarcely tempered by a southern sun. Deep down, and far away, they could see the river of Olti, winding like a snake through a maze of sheltered valleys.8 The language in which Herrmann Abich describes his impressions, coming from the side of Georgia up the valley of the Akstafa, and reaching the pass (7355 feet) over the eastern marginal heights between the village of Bekant and the town of Alexandropol, is not dissimilar to that of Koch. He speaks of the strong contrast between the physical characteristics of the plateau region before him and those of the peripheral mountains he was leaving behind. He describes the prevailing horizontality of the land-forms which he overlooked, extending to the limits of sight. In another place he alludes to the lofty, rim-like elevation with which “the Armenian plateau breaks away to the valleys of Ajara.”9 I might multiply the instances in which the most competent observers have at the same time recognised the unity of the plateau region and its sharp distinction from the peripheral mountains.
My reader has journeyed with me from the Zikar Pass to Akhaltsykh and Akhalkalaki; from the caÑon of the Toporovan river and the basin of the Kur to the streams which constitute the most northerly sources of the Araxes. We have crossed the country from Alexandropol to Erivan, from Erivan to Kars, from Kars to Kagyzman. What an impressive unity underlies the pleasing diversity of the landscapes, which melt into one another as you pass! The partings of the waters are formed by slopes which you perceive with difficulty, so gradual has been the rise and the decline. The territories of Akhaltsykh, Akhalkalaki, Alexandropol, Kars and Ardahan are all bound up together in the distribution of the space, and share features in common to a much greater extent than they are distinguished by local idiosyncrasies. The mountains, of which the outlines are never absent from the landscape—soft, long-drawn, convex shapes—stand on the floor of the tableland, like pieces upon a chessboard, which one may move from square to square. Such are the radial mass of Dochus Punar near Akhaltsykh (over 9500 feet), the two considerable elevations which enclose Lake Chaldir (Akhbaba Dagh, 9973 feet; Kisir Dagh, 10,472 feet), and even the colossal AlagÖz (13,436 feet). All are due to volcanic action, quite recent in geological time; and a similar origin belongs to the minor shapes which stud the country like bubbles upon a cooling body. Mountains of this character perform the function of boundary columns between the various districts, great and small. They determine but do not separate. How different in form and function from the folded ridges of the peripheral region, among which a single example of such recent volcanic fabrics could seldom be observed.
If we desire for convenience to partition the plateau region which is Russian Armenia, it falls most naturally into two spheres. The one will comprise a rectangular area, of which the limits on the west and east are the meridional volcanic water-partings from the Soghanlu Dagh to the heights of Sakulaperdi on one side and from the Karakach Dagh to Tawkoteli on the other. The southern boundary of this area will be the caÑon of the Araxes from its entrance into Russian territory to below the confluence of the Arpa Chai. Towards the north it includes the districts as far as the Sanislo extension of the Trialethian mountains and the Akhaltsykh-Imeritian border chain. The vast circumference of AlagÖz is placed on its south-eastern confines, sending out long feelers towards the left bank of the Arpa Chai, pushing back the mountains of the eastern border and, as it were, propping up the highlands on the north-west. This volcano may be said to lead over to the second sphere, which is for a great part an area of considerable depression, and, as compared with the longitudinal axis and symmetrical shape of its companion, is of irregular form with the greatest length from north-west to south-east. These two spheres are distinguished by features which are sufficiently contrasted to suggest a double image to the mind.
I. I have invited attention to the characteristics which Armenia shares in common with her neighbours in the series of the Asiatic tablelands, Persia on the east and Asia Minor on the west. In the brief survey to which I proceed of the plateau region within the Russian frontier it is necessary at the outset to remark upon some of the idiosyncrasies which distinguish Armenia as a whole from the other members of the series. There is in the first place the far greater elevation, investing her territory with the attributes of a roof to the adjacent countries, from which the waters gather to be precipitated in different directions, and to find their way not only to the Black Sea and the Caspian but also by almost endless stages to the Persian Gulf. The prominent part which has been played by recent volcanic action is another and not less impressive phenomenon. Which of her neighbours could compete with her in this respect? Where could one meet with an Ararat, a Sipan and a Nimrud, to say nothing of an AlagÖz and a BingÖl? Both these manifestations are exemplified in a striking manner by the surface features of the rectangular area of the more northerly sphere.
The higher levels of this region are situated at an altitude of some 7000 feet above the sea. I am speaking not of the mountains but of the plains. The uplands which give rise to the Kur in the district of GÖleh must come very near to this level. The parting of the waters of the Kur and Araxes near the village of Shishtapa, in an open landscape which may be compared to rolling downs, lies at about 7000 feet. Lake Chaldir has an elevation of 6522 feet; while of the smaller sheets of water Lake Toporovan, with 6876 feet, and the Arpa GÖl, with 6706 feet, slightly better this already considerable figure. Where the plateau falls away to the abysmal caÑon of the Araxes its edge is nearly 6500 feet high. The town of Ardahan stands at a level of 5840 feet and Kars of 5700 feet. Alexandropol, the principal city, occupies the hollow of a vast basin-like plain; yet it is over 5000 feet above the sea. These elevations are much greater than the average even in Persia, though they are to a certain extent maintained in the frontier province of Azerbaijan and along the edge of the southern peripheral mountains (Tabriz, 4650 feet; but Tehran, 3800 feet; Ispahan, 5070 feet).
The process of gradual uplift of the region by earth movements has been attended by eruptive action, flooding the country with volcanic matter, levelling inequalities of the ground and adding to the height. It has been estimated that the volcanic deposits laid bare in the ravines of the streams which descend from the radial Dochus Punar attain a depth of hundreds of yards.10 A similar phenomenon is made manifest in the caÑon of the Araxes—a cleft which in the neighbourhood of the village of Armutli, west of Kagyzman, has a depth of about 2000 feet and a width on top of at least a mile.11 There the Miocene sedimentary deposits are overlaid with tuffs and lavas in a belt over 300 yards deep.12 The points of emission of volcanic matter are in some cases true volcanoes, in others mere pustules or fissures of varying extent. One or other of these features is never absent from the landscape. But the fires are extinct; the viscous seas have long been solid; not a breath of smoke rises from the stark summits which erewhile were wreathed with vapours reflecting the glow of the flames beneath.
The distribution of such shapes due to volcanic agency may often appear arbitrary to an unpractised traveller. Here a group of stately forms resembling the giants of a forest, there a number of insignificant eminences representing the small fry. All will be found to be subject to definite and ascertainable principles, the nature of which becomes clearer at each step forward of scientific research. Perhaps the most interesting principle which we see operative in this region is the outcrop of volcanoes along meridional lines. Such groups pursue a course at right angles to the strike of the rocks within the area of the peripheral mountains. In this connection we may recall the fact that the plateau region with which we are dealing occupies the apex of the bend over of the inner arc. Lines of fracture have been thrown out at right angles to the folding, and eruptive agency has fastened upon these weakened zones of the earth’s crust. Not only may these lines be traced on the west and east of the plateau, of which, indeed, they have largely determined the shape, but also well inside of the marginal districts. In the west we have the Soghanlu group stretching north to Allah Akbar (10,218 feet), whence the direction is continued through the Ueurli Dagh (9055 feet) and the Arzian Dagh to the Chibukh-Naryn-Bashi Dagh. There the volcanic water-parting effects a junction with the Akhaltsykh-Imeritian chain in the ridge of the Sakulaperdi Dagh. In the east we have already followed the row of marginal volcanoes from Tawkoteli to Karakach. Inside these series we recognise this same north-south direction in the Abul-Samsar system, in the mountains on either side of Lake Chaldir, and, lastly, in the connection which we can scarcely err in assuming between the Kisir Dagh, overlooking the westerly shore of this lake, and its neighbour on the north, the Dochus Punar.
Compared with AlagÖz and Ararat even the absolute height of these mountains may be termed insignificant. The lofty level of the plains from which their slopes gather robs them of several thousand feet. Great Abul, with an altitude of nearly 11,000 feet, rises from a plain which itself lies at an elevation of 5500 feet. The dome-shaped vaultings of the Soghanlu Dagh near some of the sources of the Kars river are almost entirely shorn of their considerable stature by the height of the adjacent downs. In such surroundings the mountains appear to the eye as little more than hills.
The rivers as a rule flow in deep caÑons which they have eroded in the volcanic soil. Their head waters meander over grassy downs. Temperately they thread their way over the uplands or in the caÑons, except where blocks of lava may have tumbled into the trough, causing the stream to wreathe and hiss. You pass from district to district either along such natural avenues, with the towering cliffs, for the most part bare, on either hand; or, emerging from the weird scene within the hollow, over the surface of almost limitless plains. Not a tree in the landscape, and only patches of fallow and stubble, without a boundary, with rarely a village discernible from afar.
From time to time you may obtain a glimpse of the peripheral mountains—serrated summits, bush-grown slopes. These contrast to the soft convexities of the forms about you and the vaultings of the volcanic eminences. The surface of the friable soil is devoid of wood and almost of vegetation; and the volcanic matter of which it is composed produces tints of pink and ochre upon which the shadows lie transparent and thin. The rarefied atmosphere of these high regions braces the faculties and sharpens the senses; and whatever clouds may have climbed the barrier of the peripheral ranges are suspended high in the heaven, seldom obscuring the brilliant sun. During winter the land is covered with snow.
It is a country admirably adapted to grow cereals. The plains through which the Arpa Chai (grain river) eats its way to the Araxes constituted one of the granaries of Armenia in historical times.13 At the present day they have not recovered from the devastations of the Mussulman peoples, and the Russians are jealous of allowing the Armenians a free hand. Extraordinary fertility is induced by the intermixture of the lavas with alluvial or lacustrine deposits. The black earth of the plains about Akhalkalaki is famous14; and the soil in the neighbourhood of Alexandropol derives its richness from the incidence of a peculiar kind of lava side by side with the sediment of a former lake. The southerly extension of these vanished waters is marked by the belt of high ground extending from AlagÖz across the plains to the Arpa Chai. The river has forced its way through this elevation between Ani and Magaspert.15
Other effects of the violent disturbance to which the region has been subjected are manifest on a large scale. Thus all the way from the Soghanlu Dagh on the south to the neighbourhood of the mountains of the Ajars on the north the ground has fallen away to the labyrinth of valleys which feed the Chorokh by what geologists would call an extensive fault. The edge of the plateau region stands up boldly upon that side from the levels adjacent on the west. A still more recent earth movement may be represented by the uptilt towards the north-east of a considerable block of country lying between Kars and the junction of the Arpa with the Araxes. This phenomenon, which recalls a similar occurrence in the Trialethian district, has occasioned the curious course of the stream of Kars, which, rising in close vicinity to the flood of the same river to which ultimately it becomes tributary, pursues a course almost at right angles to that of the Araxes for a distance of thirty miles. To the same cause is in part due the extraordinary elevation of the levels along the left bank of the Araxes between Armutli and the confluence of the Arpa Chai.
Besides the last-named stream this lofty stage of the Armenian tableland gives birth to one of the great rivers of western Asia. The Kur rises from the highlands on the south of Ardahan, between the wall of mountain which overlooks Lake Chaldir on the west and the rim of the plateau region. In Turkish times this district constituted a separate fief, and was governed by a hereditary prince of Georgian origin who resided at Urut. The name of the district, GÖleh, still figures on the Russian maps. It is subject to a rigorous climate, the snow lying during eight months in some years. Only the hardiest of the cereals come to maturity; yet the olive and the pomegranate flourish in the valley of Artvin, but thirty miles distant, and even at this altitude and during winter the rays of a southern sun temper the cold. One of the principal arms of the river comes from the south-west, and is named the river of Ardahan; it is joined by four considerable tributaries, of which the most easterly is said by Koch to have been known to the inhabitants under the name of KyÜrr.16 Even at the present day the Kur is called the river of Ardahan until its entry into the passage of Borjom. The basin from within which these various branches gather has a length which may be computed at eight hours’ journey on horseback and a breadth equivalent to about six hours. It abounds in springs, and marshes cover its floor. Below Ardahan, where it skirts the base of the Dochus Punar system, the Kur threads a narrow valley, deeply buried in the volcanic soil. So it flows past the grottoes of Vardzia and the Devil’s City at Zeda Tmogvi, augmented by small affluents of which the largest is the Karri Chai. At Khertvis it is joined by the Toporovan river, bringing the drainage of the districts on the east, and swirling into the channel with foam-shot waves. The united volume dwells for a short space in wider landscapes, until it pierces the extreme base of the Sanislo branch of the Trialethian mountains, and is again confined in a narrow valley. Thence it issues upon the plains about Akhaltsykh, receives assembled tributaries from the northern border range, and disappears into the gorge of Borjom.
II. A traveller coming from Alexandropol down the stream of the Arpa or along the valley of the Abaran, further east, can scarcely fail to become sensible of an appreciable change in climate and scenery by the time he shall have rounded the colossal pile of AlagÖz. It is not, indeed, a new country or a new clime. The shapes which rise on the skyline are due to the same volcanic agency which has imprinted its character upon the northern landscapes. The shelving away of the ground to the basin-like depression which receives the Araxes recalls similar surface features in the northern districts. The rays of the sun fall from a heaven which remains blue. Clouds are still floating upon the azure, or are suspended upon the higher outlines. What has changed is the scale and intensity of the phenomena. The hills have given place to great mountains, the down-like expanses to one vast area of sloping ground. Into those dreamy spaces sweep the forms of the landscape, circled round them for a visible distance of some sixty miles.
The valley of the Araxes from the neighbourhood of Sardarabad to that of Julfa—a space of over a hundred miles—composes nearly one-half of the more southerly sphere of north-eastern Armenia. We are already so familiar with its overpowering individuality that it would be turning finished ground to describe it anew. For many a mile it is only confined at an immense interval by the fabric of Ararat and the pile of AlagÖz. But, even when the river—a ribbon in the expanse—has already distanced the Little Ararat, the folds of the landscape are ample into which it descends. Volcanoes on such a huge scale as these two Armenian giants could scarcely be expected to rise save on the margins of a great depression, whether subsidence may have been the cause or the effect. To the 7000 feet of the plateau region on the north this basin-like plain opposes a maximum elevation of 3000 feet and a minimum of something over 2000 feet.
The vine flourishes and is cultivated in these plains of the Araxes, and fields of castor-oil plant grace the ground. Such oases with thriving villages soften the lap of the landscape, and diversify the wide stretches of rich but idle soil which the network of trenches with their fertilising waters have not yet reached. Irrigation rather than rainfall is here the productive agency; and, indeed, this valley, with a yearly rainfall of only about six inches, is probably the driest throughout Russian Transcaucasia. The storms of the Pontic region spend themselves before reaching this haven; but they beat against the volcanoes of the meridional water-parting on the easterly margin of the more northerly sphere. Even at Alexandropol the yearly rainfall is almost three times as great as in the neighbourhood of Ararat. And while the climate of the city on the Arpa may compare with St. Lawrence in North America, that of Erivan resembles Palermo or Barcelona.17
On the north of this most extensive depression of the surface of Armenia lies the plateau region supporting Lake GÖkcheh. The axis or greatest length of that expanse of sweet water lies about parallel to the course of the Araxes, to which it sends a tributary varying in volume with the season of the year through a trench-like passage at its south-westerly extremity.18 On the north the lake is confined by a long ridge of the peripheral mountains, and its lofty level (6340 feet) is held up by the volcanic plateau of Akhmangan, acting as a dam on the side of the low-lying plains. The Akhmangan region consists of a gently vaulted platform, interrupted by a series of volcanic eminences extending over a distance of nearly thirty miles. Several of their cone-shaped summits attain a height of nearly 11,000 feet, and one, the Akh Dagh, of close upon 12,000 feet above sea-level. An absence of springs, due to the nature of the volcanic rock, is characteristic not only of this region but also of that part of the neighbouring Karabagh country which lies within the embrace of the two mountainous zones.19 In this respect it contrasts to the well-watered and wooded retreats of the district of Darachichak to the west of the lake. The wealthier citizens of Erivan take refuge in those pleasant upland valleys when the plain of the Araxes has become a furnace under the rays of a midsummer sun.
The area of the country comprised within the two spheres of which I have been speaking is about 20,587 square miles. With the exception of a narrow strip on the right bank of the Araxes, measuring 1518 square miles, the entire territory—more than commensurate with that of Servia—lies within the dominions of the Tsar.