A table of distances, which is supposed to have received the sanction of official approbation, at a former period, at Delhi; and which is calculated in royal or Shah Jehany cofses, is quite out, in the distance between Delhi and Lahore; for it allows only 105 cofses; whereas 150 would have been nearer the truth.

The latitude of Lahore is given by the Oriental tables, at 31° 50'. Col. Polier's map makes its bearing from Sirhind W 41° 45′ N, which would place Lahore in 32° 12′: but it must be observed, that the map is of too little authority in this part, to be implicitly followed. However, we may perceive what the idea of the author was; and it affords a presumptive proof that Lahore is near the parallel of 32°.* The tables in the Ayin Acbaree (Vol. III. p. 55) place Sealkote in lat. 33°; and by the construction it will be found at 53 minutes north of Lahore. Jhylum and Rotas are 33° 15′ in the same tables; and they appear to be from 66 to 69 minutes N of Lahore. These, indeed, point to a higher parallel than 32° for that of Lahore; but I prefer the proper parallel afsigned to it in the Ayin Acbaree, and in the tables of Ulug Beig and Nasereddin ; that is, 31° 50′.

If then we lay off 1351⁄2 G. miles to that parallel, Lahore will stand in longitude 73° 50′; or 6° 26' W of Canoge. The difference of longitude between the same places being 6o 30', in the same tables, proves that Lahore occupies, in my construction, the place assigned it by the Oriental geographers. It is also by the same tables, 24° 20′ east of Casbin, or Kaswin; and by M. Beauchamp's observation (in 1787), the difference is 24° 16′.

I have been thus particular on the subject of Lahore, because it forms a most capital point in this geography: for on it depend all the positions between Sirhind and Cabul. It is much to be regretted that we are not in possession of a good observation of lati

Mr. Forster, who travelled from Loldong to Jummoo, in 1783, gives a general idea of his course, which was too far northward to allow of Lahore being to the south of the position assigned it. (See Vol. I. p. 226.)

M

tude, taken there; as well as its exact distance from Sirhind, or Delhi; or both.

Lahore was the residence of the first Mahomedan conquerors of Hindoostan, before they established themselves in the central parts of the country. It owed its modern improvements, however, to Humaioon, the father of Acbar, who made it his residence during a part of his troublesome reign. Thevenot says, that including the suburbs, it was 3 leagues in length at that period; and, when he saw it, about the year 1665, the city itself was above a league in extent. Jehanguire, son of Acbar, allowed the Portuguese to build a church there; and some of its furniture remained at the time of Thevenot's visit.

The Rauvee (the ancient Hydraotes) on which it is situated, is a noble river; and by its navigable course, has a communication with the Indus and all its branches. The province, of which Lahore is the capital, is oftener named Panjab, than Lahore: however, Panjab being applied to a natural division of country, is applicable also, to part of Moultan. It is very extensive, and remarkably fertile; affording, in addition to all the necefsaries of life, wine, sugars, and cotton wool; the last of which supplied the manufactories of the province. There are also in the tract between the Indus and Jhylum (or Behut), salt mines, wonderfully productive: and affording fragments of rock salt, hard enough to be formed into vefsels, &c. Gold (according to the Ayin Acbaree) was found in the channels of its rivers; and the same is related of those of Kemaoon, which proceed from the same ridge of mountains. Ice is brought from the northern mountains, to Lahore, and sold there all the year. The famous avenue of shady trees, so much spoken of by the early Indian travellers, began at Lahore, and extended to Agra, near 500 English miles. Lahore is now the capital of the Seiks, a new power, whose name, even as a sect, was hardly known, until the rapid decline of the Mogul's empire, in the present century.

The next place, in point of consequence to the construction of this part of the map, is Attock, a city and fortrefs on the eastern bank of the Indus, and built by Acbar in the year 1581. We shall have occasion to speak on the subject of its historical importance hereafter. The position of this place is collected from various authorities. Unluckily its latitude is omitted in the Ayin Acbaree: and it can only be collected generally from that of the places lying on each side of it; and on the road passing through it from Lahore to Cabul: that is, Paishawur on the one, and Rotas, Jhylum, and Sealkote on the other. From these, as well as the distance of Attock from Jummoo, from Bheerah, and from Cashmere, respectively; it may be clearly inferred, that Attock is above the 33d degree, by several minutes. Paishawur is given at 33° 25′; Rotas and Jhylum at 33° 15′; and Sealkote at 33°; as we have said before and Attock lies in the direct road from Paishawur to the other three places. Another proof is, that the shortest road from Lahore to Cabul, that is, the direct line between the two, is through Bungush, which lies to the south of Attock and the pass of Deenkote, over the Indus, to Bungush, lies above the parallel of 32o and a half, and is supposed to be 30 miles below, or to the S of Attock: of course Attock should be above the parallel of 33°. We shall enter more particularly into this subject, when we have discussed its longitude.

With respect to the longitude of Attock from Lahore, we receive very capital assistance from the invaluable communications of Capt. Kirkpatrick, who has collected from various Persian MSS. and histories, the registers of the actual measured distances, as taken by the orders of the Emperors Acbar, Shah Jehan, and others, on the great roads from the city of Lahore, Cabul, Ghizni, Candahar, and Moultan; and back to Lahore again: as well as those between Cashmere and the cities of Lahore and Attock, respectively; and between Cabul, Balk, and Bamian: besides many other portions of different roads, too tedious to mention.

These road measures, it is true, cannot be applied to geographical purposes with critical exactness; but as an approximation to the horizontal distance may be obtained by an allowance for the inflexions of the roads, founded on experience, and applied with the requisite attention to the nature of the country; which Capt. Kirkpatrick has enabled us to do; such materials must at least be superior to computed distances, resting on the vague report, or on the judgment of an individual.

But these measurements being set forth without any intimation concerning their direction, as it respects the points of the compass, we have been necefsitated to call in other aids; since the latitudes are seldom given; or if given at all, not with the requisite exactnefs. We shall now proceed to set forth those authorities on which the longitude of Attock rests, in respect of Lahore: in order to which, we must first establish the intermediate position of Rotas.

The principal, or most frequented road from Lahore to Attock, is by Rotas; and it makes a deep bend to the north, although the bearing between the two is nearer N W by W. The reason may be, to avoid as much as possible the rugged and hilly tract of Joud, through which the direct road by Bheerah leads. That the Rotas road takes the bend abovementioned, is proved by several circumstances: 1st. The latitudes of Rotas, Jhylum, and Sealkote. 2d. The proximity of Rotas to Bember, well known to lie north of Lahore, and in the direct road to Cashmere, which also lies north of Lahore. 3d. The great length of the road by Rotas, compared with that by Bheerah. 4th. The Persian map of the Panjab places Rotas about N 37° W from Lahore, and Jummoo N 10 E (as nearly as the angles can be measured on such kind of map); that is, the angle subtended by the two, is 47 degrees. Now we shall hereafter prove, by the distance between Sirhind, Bifsooly, and Jummoo, that this last place is really about N by E from Lahore; and its distance, in the map of Panjab, is 50 cofses. And as we collect from Sherefeddin, that Timur marched 33

cofses between Jummoo and the Behut river, the distance of Rotas from that river, added to the other, agrees to an angle of about 43 degrees; and may be deemed, at least, a strong presumptive proof, that Rotas bears about NW by N from Lahore. 5th. Sealkote, which lies 10 cofses from Jummoo, in the road to Lahore, is

44 marches distant from the river Jhylum, near the town of that name. These marches were made by the Emperor Baber, and are reported by Capt. Kirkpatrick: they may be taken at 44 G. miles direct distance, reckoning the mean march at 14 B. miles, by the road; or 11 G. miles direct.

The distance of Rotas from Lahore, is very correctly given by Capt. Kirkpatrick, at 41 royal cofses (or those of Shah Jehan) of 4995 yards each, and allowing 1 in 9 for the winding in ordinary cases,* each cofs, in horizontal or direct distance, will be equal to 2,0214 G. miles; or 29,68 such cofses will be equal to a degree. Forty-one cofses then, are equal to nearly 83 G. miles, when the general direction of the road does not undergo a change: but here we have evidence that it changes at the crossing-place of the Chunaub river, 23 cofses from Lahore; and the lofs of distance, by the angle, may reduce that of Rotas to 81: and Rotas will then stand in lat. 32° 58', instead of 33° 15′ of the tables: provided Lahore be in 31° 50′, as we have reason to believe.

The town of Gujerat lies on the Rotas road, between the Chunaub and Jhylum, at four or five miles from the former. Purchas (Vol. I. p. 434) says, that Bember is 16 cofses N or NE from Gujerat; and as Attock is reckoned 138 such cofses from Lahore, 16 would produce 19 G. miles: and Bember being 33 Acbáree cofses, on a bearing of N, a little W from Lahore, this appears consistent, and regulates the connexion of the two roads which lead to Attock and Cashmere respectively. It must be noted, that

* Or, in other words, the distance by the road is allowed to be one-eighth more than the direct distance.

These are of 4757 yards each, and, with an allowance for the windings of the road, are taken at 1,925 G. miles each: or 31,16 are equal to a degree. [Kirkp. MSS.]