and clothe themselves in dresses made of the bark of trees. tants of Kharkhand are Mohammedans.

The present inhabi

On the Physical Geography of Colorado and adjacent Regions.
By Governor GILPIN.

On Lines for a Ship-Canal across the American Isthmus.
By General W. HEINE, U.S.Ą.

The author visited the Atlantic side of the isthmus early in the present year on a mission intended to examine the correctness of the statements of M. Lacharme, an engineer who explored the interval from the Tuyra River on the Pacific side and the Cacarica branch of the Atrato on the Atlantic, and declared there were no great obstacles to the construction of a canal at that point, the length to be cut being only 52 miles, and the greatest elevation only 186 feet. The author was not able to ascend the Atrato, but all he saw went to confirm M. Lacharme's statements. These he gave in detail, showing that they were founded on a conscientious survey, with all necessary scientific instruments. From his own observations and those of the recent United States Survey, the author demonstrated the utter impracticability of any other part of the isthmus for the purposes of an interoceanic canal, and insisted upon the necessity of a further survey by the line of the Atrato and Cacarica.

On the Great Currents of the Atmosphere.

By JOHN K. LAUGHTON, M.A., F.R.A.S., F.R.G.S.

The author pointed out several geographical facts which were opposed to the received theory of the trade-winds, known as Hadley's theory. Heat does not cause a wind towards any of the principal areas of greatest temperature; either towards the Desert of Sahara, the Arabian Desert, the interior of Australia, the Red Sea, the Persian Gulf, or even, when carefully traced, towards the Great Prairie of North America. The effect attributed to the rotation of the earth also is not consistent with numerous observed facts, such as the S.E. wind in the Gulf of Mexico, the N.W. wind on the coast of North Africa, between Cape Verde and Cape Palmas, the N.W. gales in the North Atlantic, the S.W. wind on the south coast of Australia, and very many others; the idea, indeed, appeared to have been formed in a temporary forgetfulness of the power of friction, which in the case of air is very intense. Winds which, in accordance with Hadley's theory, have been very generally divided into polar and equatorial, seem more naturally to divide themselves into easterly and westerly. As our experience grows larger, we learn that the westerly winds have an extent and a power incompatible with the idea of their secondary nature. They extend from 60° N. to 60° S., interrupted only by the trade-winds. The trade-winds are small in comparison, and of very limited height, the westerly winds blowing above them as strongly as they do both above and near the surface in temperate zones. The westerly are really the primary winds, whilst the equable trade-winds, of very limited volume, are reflex streams of air caused by the impact of the great westerly winds on the continental barriers, whether against mountain-ranges or against the more sluggish air which lies over the land. In the Atlantic we see the main westerly stream of air dividing, on about the parallel of 45° N., and turning north as a S.W. wind on our coasts, or south as a N.W. and N. wind on the coast of Portugal. On the other hand, at the extreme west, the westerly wind continually dragging away the air from the eastern side of the Rocky Mountains, causes such a tendency towards a vacuum, that the air from the south and north is induced towards it. It was impossible to say definitely why the atmosphere should have this prevailing westerly motion, but the author was inclined to seek its cause in the attraction of the heavenly bodies.

The Landfall of Columbus. By R. H. MAJOR, F.S.A., Hon. Sec. R.G.S. It is surprising that after the lapse of nearly four centuries there should be any doubt as to the spot in the New World which was first lighted on by Christopher Columbus. In this paper the author set himself to show, not only that the name which Columbus gave to that spot was for nearly two centuries applied to an island to which it never belonged, but that among the advocates of various islands for the honour of being the true landfall the very latest had to be confuted, while the one who had adduced the best arguments in favour of the correct island had been greatly at fault with respect to the point of anchorage. Columbus gave the name of San Salvador to the island which he first discovered. Its Indian name was Guanahani. In 1793 Juan Bautista Muñoz, in his 'Historia del Nuevo Mundo,' declared his belief that Guanahani was Watling's Island, in contravention of the maps which from the beginning of the eighteenth century had given the name of San Salvador to Cat Island. In 1825 Navarrete, in his 'Coleccion de los Viages y descubrimientos que hicieron por mar los Españoles desde fin del siglo xv.," believed it to be Turk's Island. In 1828 Washington Irving, in his 'Life of Columbus,' decided in favour of Cat Island, relying mainly on Captain Slidell Mackenzie's interpretation of the Diary of Columbus;' and in 1837 this conclusion received the weighty approval of the Baron Alexander von Humboldt in his Examen Critique de l'Histoire de la Géographie du nouveau Continent.' In 1856 the claims of Watling's Island found a fresh champion in Capt. Becher, of the Hydrographic Office, in his work entitled the 'Landfall of Columbus;' but in 1864, and again so late as 1869, Senhor de Varnhagen, in his 'Verdadera Guanahani dé Colon,' has put in a claim for the island of Mayaguana. The author of the paper first examined these respective claims by the light of the Diary of Columbus 'himself, the real fountain-head of information upon the subject; and having shown therefrom that the arguments in favour of Cat Island, Turk's Island, and Mayaguana were untenable, proceeded to fix the identity of Guanahani with Watling's Island by a process which reduced the chances of error to a minimum. He produced a facsimile diagram of a map of the Bahamas published in 1601 by Herrera, the official historiographer of the Indies in Spain, and laid down by him from the original documents in the handwriting of Columbus and his contemporaries, which, in his official position, he had under his special charge. The value of this authoritative map was all the greater that it was constructed before any question was raised on the point in dispute; it was new enough to contain all the islands in their approximately correct position, but old enough to contain, not only the name of Guanahani, but a large proportion of ancient names identical with those at present existing. Side by side with this was a diagram made from the Admiralty Survey, showing these islands as now known, and with their modern names. Out of twenty-four islands thus brought under comparison, ten retained, in the modern map, the same names as they held in the old, thus affording valuable stations for accurate comparison. Of these ten, one was Senhor de Varnhagen's Mayaguana itself, which was represented, together with the island of Guanahani, on Herrera's map; so that His Excellency's claim was completely neutralized, since by no possibility could two islands be made identical which were so markedly distinct as to have several other islands lying between them. The comparison between the two diagrams plainly showed that Guanahani was Watling's Island. But while thus demonstrating the correctness of the conclusions of Muñoz and Capt. Becher on this head, the author entirely disagreed with the latter as to the point where the Admiral first anchored off that island, and also as to his movements while there. Capt. Becher makes Columbus anchor a little to the south of the N.E. point of the island; and when he tripped his anchor, makes him sail round the northern end of the island. He also makes Columbus's ship follow the boats in their reconnaissance. Not one of these statements or conclusions is in accordance with the 'Diary,' nor would such a movement lead to the topographical discoveries recorded. The 'Diary' says that Columbus took the ship's boat and the caravel's barges and went along the island in a N.N.E. direction to see the other part of the island to the eastward; and as the trending of the southern part of the east side of the island is itself N.N.E., it is clear that such a movement necessitates starting from a point on the S.E. of the

island. This very manifest fact is in accordance with the discoveries made by Columbus on the island, and also with what took place when Columbus left his moorings for the second island, for he then saw several islands, and was doubtful which he should visit first. This would really be the case when starting from the S.E. point of Watling's Island, but would not hold good if he started from the anchorage assigned by Capt. Becher; so that the author has here, for the first time, demonstrated that the first anchorage of Columbus in the New World was off the south-east point of Watling's Island.

On Railway Routes across North America and the Physical Aspects of the Country. By Lord MILTON, M.P., F.R.G.S.

Journey into the Interior of Hadramaut. By WERNER Munzinger. The author, after recovering from wounds received in Abyssinia, accompanied Capt. Miles on an excursion eastward from Aden towards the interior of Arabia. The region traversed abounded in Himyaritic inscriptions and in other vestiges of remote civilization. The travellers went by sea as far as Bir Ali, and travelled thence into the interior for a distance of about 300 miles, their furthest point being a place called Habban, 3000 feet above the level of the sea. Their route was laid down by compass bearings, and they took barometric observations for height. From Bir Ali the country formed a plain with a gentle slope inland, nearly covered by isolated hills and ridges of sandstone with flat tops, all of the same height, about 1500 feet above the plain, and quite destitute of vegetation: the very narrow strips of alluvial soil in the ravines, not one-tenth of the whole, are alone capable of cultivation; but these are generally well cared for, and yield three and even four crops in the year, being irrigated by wells. These patches form a number of oases with a dense population and towns of several thousand inhabitants. The people cultivate dates, millet, wheat, and the Abyssinian grain called tef. Water is generally met with, in boring, about fifty feet from the surface. Beyond this region, and further inland, they came to what M. Munzinger called a granitic and metamorphic land, with rounded hills bounding several wide plains. Here there was more vegetation, with some fine trees; and wild hogs, gazelles and herds of cattle were met with. The people belonged to different races, and the Himyaritic language was not entirely lost or forgotten, in spite of 1200 years of Islam. But all spoke Arabic, though in a very strange dialect; there was an absence of religious feeling and of regular government, and civilization was at a very low ebb, the only sign of it being the very large houses with several stories, each of them a castle in itself. The travellers met with little hospitality, but were not actually ill-treated. At Ghorab they were near the Desert of Akhaf, described by Wrede, and the Bahr el Saffi, or Sea of Saffi, so called from King Saffi, who, in an attempt to cross the desert, disappeared with his entire army. The desert was described as an immense sandy plain, covered with numberless undulating hills, which gave it the appearance of a moving sea, and as lying 1000 feet below the level of the granitic land. On the desert are white patches formed of impalpable powder, into which if a plummet with a sixty-fathom line is thrown, the whole slowly disappears. In one of these quicksands the ill-omened King Saffi and his host found a tomb. The whole region of Hadramaut and Yemen is full of curious legends, and abounds in geographical and historical questions of the deepest interest. The author confessed that the excursion of Capt. Miles and himself formed but a small contribution to our knowledge of Arabian geography, but hoped it would incite other travellers to explore this nearly unknown land.

Notes on the Site of the Terrestrial Paradise. By Major-Gen. Sir HENRY RAWLINSON, K.C.B., LL.D., D.C.L., F.R.S.*

In this paper the author propounded, as the result of his investigations of Semitic antiquities and of the Cuneiform Tablets of Babylonia, a new hypothesis * This paper will be printed in extenso in the Journal of the Royal Geographical Society.

regarding the site of the traditional Garden of Eden of the Hebrews. He remarked, in the first place, that on examining the early traditions of nations, we invariably found the Heaven-land, the abode of the gods, the connecting-link between divinity and humanity, to lie in that region of the earth from which the recording race took its intellectual origin. In illustration of this he need only refer to the Olympus of the Greeks and the Merú of the Aryans, which latter had three sites, according to the habitat of the three branches of the Aryan race; the Persians, or Western Aryans, placing their Irán-víj in the Paropamisus, while the Merú of the Central Aryans was in Pamír, and of the Eastern about the Sacred Lakes in Thibet, and in each of these there were supposed to be four rivers flowing from a common centre. There was ground, then, for supposing the Paradise of the Hebrews to lie in that region which was the cradle of the nation, namely, near Ur of the Chaldees, which the author had been able to demonstrate, from cuneiform inscriptions, to have been situated on the lower Euphrates, at the place now called Muzheir. The name of "Hebrew" was also derived from the same locality, the zone, or belt, of alluvial land between the river and the tertiary formation having the specific title among the Arabian geographers of Ibr, or "the bank;" so that Ibri was a perfectly correct ethnic title for the Abrahamic emigrants. Further, the author suggested that Gan-eden, which we translate "Garden of Eden," was nothing more than the Hebrew rendering of one of the old vernacular names of Babylonia, which was Gana-duni (or, with the case-ending, Gana-duniyas), Gana signifying apparently "an enclosure," while Duni or Aduni was one of the earliest gods worshipped in the country. Without, however, insisting on this identification of the name of the country, he would rely mainly on the names and attributes of the four rivers which watered the garden, and which were evidently intended, as Kalish has remarked, to furnish an exact geographical description of Eden. These rivers, as it is well known, were Pison, Gihon, Hiddekel, and Euphrates. Now the land of Babylonia was constantly illustrated in the cuneiform inscriptions by the names of four rivers, two of which answer to the Tigris and the Euphrates, and the other two were named the Surrapi and the Ukni. The latter two were Assyrian terms, and their Babylonian equivalents had not yet been identified. The Surrapi seemed, however, to answer partly to the Biblical Gihon, and Ukni to the Pison, and they represented respectively the left hand, or eastern arm of the Tigris, and the right hand, or western arm of the Euphrates.

Regarding the Pison, it is said in Genesis, "The name of the first is Pison, that is it which compasseth the whole land of Havilah, where there is gold; and the gold of that land is good: there is bdellium and the onyx stone." The name Pisun, coming from the Hebrew verb "to disperse," signifies "the overflow;" and as in all ages there has been an outlet to the Euphrates above Babylon, where the flood drains off to the south-east, varying constantly in its course and name, and as Ukni had been shown on independent grounds to mean "the onyx," or "the onyxriver" (though probably the term really refers to alabaster, quarries of which existed just outside the Euphrates alluvium), the author considered there was good reason for identifying the stream with the Pison of Genesis. Bdellium he considered to mean "pearls" (Bedolat), which were obtained at the mouth of the river, from the banks in the Persian Gulf; and the land of Havilah he believed to be the strip of sandy desert which skirts the Arabian upland; Haul signifying simply "sand.” With regard to the Gihun, or the river "which compasseth the whole land of Cush," his theory was, that in very early times the left branch of the Euphrates, which left the main river just above Babylon and ran due east to the Tigris, was considered to be the same as the left arm of the Tigris itself, that arm being prolonged in the same line to the eastward, while the right arm of the Tigris was considered to be the true continuation of the upper course of the river following the same general direction of south-east. In a rough way, it might be said that the left arm of the Euphrates thus crossed the Tigris and formed the Gihun. He justified this theory on philological grounds, showing that the left arm of the Tigris had retained the name of Guhá, absolutely identical with the Hebrew reading of Gihon, almost to the present day, and discussed the whole subject in some detail. As to the description of the Gihun as "encompassing the whole land of Cush" (which, by a very bold guess, our translators had rendered "Ethiopia"), " Cush," or

66

"Kish," was one of the primitive capitals of Babylonia, and it gave its name apparently to the whole country along the river. Kusiya" was mentioned in that quarter among the possessions of Darius Hystaspes. Various other reasons for this identification were adduced.

The third river offered fewer difficulties, as no one had ever doubted that the Heddekel was the Tigris. The fourth river was Perat, or the Euphrates. In the inscriptions the word is often represented by the sign for "water," in the same way as it is called in Scripture "the great river;" but usually the upper river has the name of Purat, where we have probably a very ancient root, signifying "to abound" or "fructify," common to both the Aryan and Semitic tongues; the lower river, below the Pison branch, is called in the inscriptions the river of Sippara, from the town of that name.

Early Traditions regarding the River Oxus.

By Major-Gen. Sir H. RAWLINSON, K. C.B., LL.D., D.C.L., F.R.S., &c.: Whether Bournouf was right or not in regarding the term Pámir (the region in which the Oxus takes its rise) as a contraction of Upá Méru, “the country above Mount Meru," and in thus associating the name directly with the holiest spot in the Brahmanical Cosmogony, the author of the paper thought it was certain that the geographical indications of the Puránas all pointed to this quarter of Central Asia as the site of the primeval Aryan Paradise. We were not, however, limited to Sanscrit authorities in studying this subject; the Puránas were supplemented by the traditions and travels of the Buddhists, and in these later sources of information we often find evidence of so direct a nature as almost to meet the requirements of modern science. Thus, in regard to the four rivers of the Aryan Paradise, which were named by the Brahmans, 1, the Sita; 2, the Alakananda; 3, the Vakhshu; and 4, the Bhadra, the Buddhists varied both the order and the nomenclature, classing the four rivers as, 1, the Ganges; 2, the Indus; 3, the Oxus; and 4, the Sita; and, further, deriving them from a great central lake, which was named A-neou-ta, and one of the representatives of which was either the Kara-kul or the Sarik-kul Lake of Pamir. The Buddhist traveller Hiouen-thsang in A.D. 644 recognized in the Tsung-ling, or Pamir chain, the Sú-merú of his national cosmography. Capt. Wood, in the account of his journey to the sources of the Oxus, had furnished us with an explanation of the origin of the old legend of a fourrivered Paradise. He observes that "the hills and mountains which encircle Siri-kul, give rise to some of the principal rivers in Asia. From the ridge at its eastern end flows a branch of the Yarkand river, one of the largest streams that waters Chinese Tartary, while from the low hills on the northern side rises the Sirr, or River of Kokan, and from the snowy chain opposite, both forks of the Oxus as well as a branch of the River Kuner are supplied." Although the position of these various streams is now known to be incorrect, we had a right to infer that the many-rivered wealth of Pamir had so impressed the imagination of the primitive Aryan colonists that in their subsequent migrations towards the south, and with a more extended geographical knowledge, they transferred the physical features of the fatherland to the abode of Brahma and the gods, precisely in the same way as the Semitic Jews, after being transplanted to the coast of Syria, preserved in their delineation of the terrestrial Paradise the memory traditionally handed down of their old habitat in Babylonia between the Tigris and the Euphrates. Another Aryan legend confirmed this presumed connexion between the headstreams of the Oxus and the several rivers of Asia which were fabled to fall from heaven upon Mount Meru, thence to flow to the surrounding world. One version of the Puranic legend described the rivers flowing from Mount Meru as seven; and this had its parallel in the popular geography of Pamir, for the region of the Upper Oxus was known to the Iranian division of the Aryan race by the name of the Country of the Seven Rivers. The passage in the Vendidad to this effect was confirmed by Abu Rihan El-Biruni, a very competent authority. The author believed that a critical examination of the geography of the Puránas might lead to some curious results as to the period and track of the various Aryan migrations.