Istakhri says that Rasak is two days' journey from Fahalfahra (which there is good reason for believing to be Bahu Kalat), but Idrisi makes it a three-days' journey from that place, and three days from Darak, so that it should be about half-way between them. Now, Darak can hardly be other than Dizak, which is described by the same authority as three days' journey from Firabuz (i.e. Kanazbun). It is also said to have been a populous town, and south-west of it was "a high mountain called the Mountain of Salt." South-west of Dizak are the highest mountains in Makran, called the Bampusht Koh, and there is enough salt in the neighbourhood to justify the geographer's description. It may also be said to be three days' journey from Kanazbun. Somewhere about half-way between Dizak and Bahu Kalat is the important town of Sarbaz, and from a description of contiguous ruins which has been given by Mr. E. A. Wainwright, of the Survey Department (to whom I am indebted for most of the Makran identifications), I am inclined to place the ancient Rasak at Sarbaz rather than in the position which the modern name would apply to it. It is rather significant that Ibn Haukal omits Rasak altogether from his map. Its importance may be estimated from Idrisi's description of it taken from the translation given by Elliott in the first volume of his History of India: "The inhabitants of Rasak are schismatics. Their territory is divided into two districts, one called Al Kharij, and the other Kir" (or Kiz) "Kaian. Sugar-cane is much cultivated, and a considerable trade is carried on in a sweetmeat called 'faniz,' which is made here.... The territory of Maskan joins that of Kirman." Maskan is probably represented by Mashkel at the present day, Mashkel being the best date-growing district in Southern Baluchistan. It adjoins Kirman, and produces dates of such excellent quality that they compare favourably with the best products of the Euphrates. Idrisi's description of this part of Western Makran continues thus: "The inhabitants have a great reputation for courage. They have date-trees, camels, cereals, and the fruit of cold countries." He then gives a table of distances, from which we can roughly estimate the meaning of "a day's journey." After stating that Fahalfahra, Asfaka, Band, and Kasrkand are dependencies of Makran which resemble each other in point of size and extent of their trade, he goes on to say, "Fahalfahra to Rasak two days." (Istakhri makes it three days, the distance from Bahu Kalat to Sarbaz being about 80 miles.) "From Fahalfahra to Asfaka two days." (This is almost impossible, the distance being about 160 miles, and the route passing through several large towns.) "From Asfaka to Band one day towards the west." (This is about 45 miles south-west rather than west.) "From Asfaka to Darak three days." (150 to 160 miles according to the route taken.) "From Band to Kasrkand one day." (About 70 miles, passing through Bih or Geh, which is not mentioned.) "From Kasrkand to Kiz four days." This is not much over 150 miles, and is the most probable estimate of them all. It is possible, of course, that from 70 to 80 miles may have been covered on a good camel within the limits of twenty-four hours. Such distances in Arabia are not uncommon, but we are not here dealing with an absolutely desert district, devoid of water. On the contrary, halting-places must have always been frequent and convenient.

I cannot leave this corner of Makran without a short reference to what lay beyond to the north-west, on the Kirman border, as it appears to me that one or two geographical riddles of mediæval days have recently been cleared up by the results of our explorations. Idrisi says that "Tubaran is near Fahraj, which belongs to Kirman. It is a well-fortified town, and is situated on the banks of a river of the same name, which are cultivated and fertile. From hence to Fardan, a commercial town, the environs of which are well populated, four days. Kir Kaian lies to the west of Fardan, on the road to Tubaran. The country is well populated and very fertile. The vine grows here and various sorts of fruit trees, but the palm is not to be found." Elsewhere he states that "from Mansuria to Tubaran about fifteen days"; and again, "from Tubaran to Multan, on the borders of Sind, ten days." Here there is clearly the confusion which so constantly arises from the repetition of place-names in different localities. Multan and Mansuria are well-known or well-identified localities, and Turan was an equally well-recognized district of Lower Sind, of which Khozdar was the capital. Turan may well be reckoned as ten days from Multan, or fifteen from Mansuria, but hardly the Tubaran, about which such a detailed and precise description is given. There are two places called indifferently Fahraj, Pahrag, Pahra, or Pahura, both of which are in the Kirman district; one, which is shown in St. John's map of Persia, is not very far from Regan, in the Narmashir province, and is surrounded far and wide with ruins. It has been identified by St. John as the Pahra of Arrian, the capital of Gadrosia, where Alexander rested after his retreat through Makran. The other is some 16 miles east of Bampur, to the north-west of Sarbaz. Both are on the banks of a river, "cultivated and fertile"; both are the centres of an area of ruins extending for miles; both must find a place in mediæval geography. For many reasons, into which I cannot fully enter, I am inclined to place the Pahra of Arrian in the site near Bampur. It suits the narrative in many particulars better than does the Pahra identified with Fahraj by St. John. The latter, I have very little doubt, is the Fahraj of Idrisi, and the town of Tubaran was not far from it. Fardan may well have been either Bampur itself (a very ancient town) or Pahra, 16 miles to the east of it; and between Fardan and Fahraj lay the district of Kir (or Kiz) Kaian, which has been stated to be a district of Rasak. "On Tubaran," says Idrisi, "are dependent Mahyak, Kir Kaian, Sura" (? Suza), "Fardan" (? Bampur or Pahra), "Kashran" (? Khasrin), "and Masurjan. Masurjan is a well-peopled commercial town surrounded with villages on the banks of the Tubaran, from which town it is 42 miles distant. Masurjan to Darak Yamuna 141 miles, Darak Yamuna to Firabuz 175 miles." If we take Regan to represent the old city of Masurjan, and Yakmina as the modern representative of Darak Yamuna, we shall find Idrisi's distances most surprisingly in accordance with modern mapping. Regan is about 40 miles from Fahraj, and the other distances, though not accurate of course, are much more approximately correct than could possibly have been expected from the generality of Idrisi's compilation.

I cannot, however, now open up a fresh chapter on mediæval geography in Persia. It is Makran itself to which I wish to draw attention. In our thirst for trans-frontier knowledge farther north and farther west, we have somewhat overlooked this very remarkable country. Idrisi commences his description with the assertion that "Makran is a vast country, mostly desert." We have not altogether found it so. It is true that the voyager who might be condemned to coast his way from the Gulf of Oman to the port of Karachi in the hot weather, might wonder what of beauty, wealth, or even interest, could possibly lie beyond that brazen coast washed by that molten sea; might well recall the agonies of thirst endured during the Greek retreat; might think of the lost armies of Cyrus and Simiramis; and whilst his eye could not fail to be impressed with the grand outlines of those bold headlands which guard the coast, his nose would be far more rudely reminded of the unpleasant proximity of Ichthyophagi than delighted by soft odours of spikenard or myrrh. And yet, for century after century, the key to the golden gate of Indian commerce lay behind those Makran hills. Beyond those square-headed bluffs and precipices, hidden amongst the serrated lines of jagged ridges, was the high-road to wealth and fame, where passed along not only many a rich khafila loaded with precious merchandise, but many a stout array of troops besides. Those citizens of Makran who "loved fair dealing, who were men of their word, and enemies to fraud," who welcomed the lagging khafila, or sped on their way the swift camel-mounted soldiers of Arabia, could have little dreamed that for centuries in the undeveloped future, when trade should pass over the high seas round the southern coast of Africa, and the Western infidel should set his hated foot on Eastern shores, Makran should sink out of sight and into such forgetfulness by the world, that eventually this ancient land of the sun should become something less well known than those mountains of the moon in which lay the far-off sources of the Egyptian Nile.

Yet it is not at all impossible that Makran may once again rise to significance in Indian Councils. Men's eyes have been so much turned to the proximity of Russia and Russian railways to the Indian frontier that they have hardly taken into serious consideration the problems of the future, which deal with the direct connection overland between India and Europe other than those which touch Seistan or Herat. That such connection will finally eventuate either through Seistan or Herat (or through both) no one who has any appreciation of the power of commercial interests to overcome purely military or political objections will doubt; but meanwhile it may be more than interesting to prove that a line through Persia is quite a practicable scheme, although it would not be practicable on any alignment that has as yet been suggested. It would not be practicable by following the coast, for instance. It would be useless to link up Teheran with Mashad, unless the Seistan line were adopted in extension; and the proposal to join Ispahan to Seistan through Central Persia would involve such a lengthening of the route to India as would seriously discount its value. The only solution of the difficulty is through Makran to Karachi. Military nervousness would thus be met by the fact that Russia could make no use of such a line for purposes of invasion, inasmuch as it would be commanded and protected from the sea. Political difficulties with Afghanistan would be absolutely avoided by a Persian line. Whether that would be better than a final agreement with Russia based on mutual interest, which would certainly make strongly for the peace of our borders, is another question. I am only concerned just now in illustrating the geography of Makran and pointing out its facilities as a land of possible routes to India, and in showing how the exploration of Baluchistan and of Western India was secured in mediæval times by means of these routes.

It will, then, be interesting to note that at the eastern extremity of Makran, dovetailed between the Makran hills as they sweep off with a curve westward and our Sind frontier hills as they continue their general strike southwards, is the little state of Las Bela. The mountain conformation which encloses it makes the flat alluvial portion of the state triangular in shape, and from the apex of the triangle to the sea runs a river now known as the Purali, which in ancient times was called the Arabis from the early Arab occupation of the region. There are relics of apparent Arabic origin which, independently of Greek records, testify to a very early interest in this corner of the Indian borderland. Las Bela has a history which is not without interest. It has been a Buddhist centre, and the caves of Gondakahar near by testify to the ascetic fervour of the Buddhist priesthood. The grave of one of the greatest of frontier political leaders, Sir Robert Sandeman, lies near this little capital. Already it forms an object of devotional pilgrimage through all the Sind countryside. Possibly once again it may happen that Las Bela will be a wayside resting-place on the road to India, as it has undoubtedly been in the centuries of the past. It is not difficult to reach Las Bela from Karachi by following the modern telegraph line. There are no great physical obstacles interposed to make the way thorny for the slow-moving train of a khafila, and where camels can take their stately way there the more lively locomotive can follow. Should the railway from Central Persia (let us say Ispahan) ever extend its iron lines to Las Bela, it will make little of the rest of its extension to Karachi. It is the actual physical arrangement of Makran topography only which really matters; and here we are but treading in the footsteps of the ubiquitous Arab when first he made his way south-eastward from Arabia, or from Syria, to the Indian frontier. He could, and he did, pass from the plateau of Persia into the very heart of Makran without encountering the impediment of a single difficult pass.

Although the chief trade route of the Arabs to India was not through Persia, but by way of the sea in coasting vessels, it is probable that both Arabs and Persians before them made good use of the geographical opportunities offered for an approach to the Indus valley and Northern India, and that the central line of Persian approach through Makran had been a world-old route for centuries. It is really a delightful route to follow, full of the interest of magnificent scenery and of varied human existence, and it is the telegraph route from Ispahan to Panjgur in Makran. With the initial process of reaching Ispahan, whether through the Kurdistan hills from Baghdad by way of Kermanshah and the ancient town of Hamadan to Kum (the mountain road selected for the telegraph line), or whether from Teheran to Kum and thence by Kashan (a line not so replete with hills), we have no concern. This part of Persia now falls by agreement under the influence of Russia, and it is only by further agreement with Russia that this link in any European connection could be forged. But from Ispahan to Karachi one may still look over the wide uplands of the Persian plateau and imagine, if we please, that it is for England to take her share in the development of these ancient highways into a modern railway. Ispahan is 5300 feet above sea-level, and from Ispahan one never descends to a lower level than 3000 feet till one enters Makran.

As Ispahan lies in a wide valley separated by a continuous line of flanking hills from the main high road of Central Persia, which connects Teheran and Kashan with Kirman, passing through Yezd, it is necessary to cross this intervening divide in order to reach Yezd. There is a waterway through the hills, near Taft, a little to the south-west of Yezd which meets this difficulty. From Yezd onwards to the south-west of Kirman, Bam, and the populous plains of Narmashir and Regan, the road is never out of sight of mountains, the long lines of the Persian ranges flanking it north and south culminating in the magnificent peak of the Koh-i-Basman, but leaving a wide space between unhindered by passes or rivers. From Narmashir the modern telegraph passes off north-eastward to Seistan, and from there follows the new trade route to Nushki and Quetta. It is probable that through all ages this palpable method of circumventing the Dasht-i-Lut (the Kirman desert) by skirting it on the south was adopted by travellers seeking Seistan and Kandahar. There is, however, the difficulty of a formidable band of mountains skirting the desert Seistan, which would be a difficulty to railway construction. From Regan to Bampur and Panjgur the normal and most convenient mountain conformation (although the ranges close in and the valleys narrow) points an open way, with no obstacle to bar the passage even of a motor; but after leaving Bampur on the east there is a divide (of about 4000 to 5000 feet) to be crossed before dropping into the final system of Mashkhel drainage, which leads straight on to Panjgur, Kalat, and Quetta. Early Arab commercial explorers did not usually make this detour to Quetta in order to reach the Indus delta country, nor should we, if we wished to take the shortest line and the easiest through Persia to Karachi or Bombay. Much depends on the objective in India. Calcutta may be reached from the Indus valley by the north-western lines on the normal Indian gauge, or it may be reached through the Rajputana system on the metre gauge. But for the latter system and for Bombay, Karachi becomes our objective. To reach Karachi via Seistan and Quetta would add at least 500 unnecessary miles to our route from Central Persia, an amount which equals the total distance between the present Russian terminus of the Transcaspian line at Kushk and our own Indian terminus at New Chaman. A direct through line from Panjgur to Karachi by the old Arab caravan route, within striking distance from the sea, would apparently outflank not only all political objections, but would satisfy those military objectors who can only see in a railway the opportunity for invasion of India.

CHAPTER IX

EARLIEST ENGLISH EXPLORATION—CHRISTIE AND POTTINGER

The Arabs of the Mediæval period, whose footsteps we have been endeavouring to trace, were after their fashion true geographers and explorers. True that with them the process of empire-making was usually a savage process in the first instance, followed by the peaceable extension of commercial interests. Trade with them (as with us) followed the flag, and the Semitic instinct for making the most of a newly-acquired property was ever the motive for wider exploration. With the Chinese, during the Buddhist period, the ecstatic bliss of pilgrimage, and the acquirement of special sanctity, were the motive power of extraordinary energies; but with this difference of impulse the result was much the same. Arab trader and Chinese pilgrim alike gave to the world a new record, a record of geographical fact which, simple and unscientific as it might be, was yet a true revelation for the time being. But when Buddhism had become a memory, and Arab domination had ceased to regulate the affairs of the Indus valley; when the devastating hordes of the Mongol swept through Afghanistan to the plains of India, geographical record no longer formed part of the programme, and exploration found no place in the scheme of conquest. The Mongol and the Turk were not geographers, such as were the Chinese pilgrim and the Arab, and one gets little or nothing from either of geographical record, in spite of the abundance of their historical literature and the really high standard of literary attainment enjoyed by many of the Turk leaders. That truly delightful historical personage Babar, for instance, "the adventurer," the founder of the Turk dynasty in India, good-looking, intellectual, possessed of great ability as a soldier, endowed with true artistic temperament as painter, poet, and author, the man who has left to all subsequent ages an autobiography which is almost unique in its power of presenting to the mind of its reader the impression of a "whole, real, live, human being," with all his faults and his fancies, his affections and aspirations, was apparently unimpressed with the value of dull details of geography. He can say much about the human interests of the scenes of his wanderings; he can describe landscape and climate, flowers and fruits (especially melons); but though he doubtless possessed the true bandit's instinct for local topography (which must, indeed, have been very necessary in many of the episodes of his remarkable career) he makes no systematic attempt to place before us a clear notion of the geographical conditions of Afghanistan as they existed in his time. His literary cousin Haidar is far more useful as a geographer. To him we owe something more than a vague outline of the elusive kingdom of Bolar and the limits of Kafiristan, but he merely touches on Afghanistan in its connection with Tibet, and says little of the country with which we are now immediately concerned.