The Emperor Mahomed, in the 14th century, made an attempt to establish the capital of his empire at Deogire; and to that end almost ruined Delhi, in order to drive the inhabitants to his new capital, about 750 miles from their ancient habitations. This scheme, however, did not succeed: and was, if possible, the more absurd, as at that time but a small progrefs had been made in the conquest of the Deccan.

The pagodas of Elora are in the neighbourhood of Dowlatabad, and are mostly cut out of the natural rock. M. Thevenot, who particularly describes them, says, that for two leagues together, nothing is to be seen but pagodas, in which there are some thousands of figures. He does not, however, greatly commend the sculpture of them: and, I apprehend, they are of early Hindoo origin. We must remember that Deogire, which stood in this neighbourhood, was the greatest and richest principality in the Deccan; and that the fame of its riches, incited Alla to attack it, in 1293: and these elaborate monuments of superstition, were probably the offspring of that abundant wealth, under a government purely Hindoo.

M. Bufsy's line includes within it, the positions of Hydrabad, Golconda and Beder. When the line is corrected as above, to Aurungabad, Hydrabad will be found in lat. 17° 24'; which I conceive to be too far to the northward, considering its reputed distance from Nagpour and Cuddapah. M. D'Anville too (in his Eclaircissemens) says that the latitude of Hydrabad is 17° 12′. How he came by his information, I know not; but I believe it to be nearly right: and this is the parallel it is generally placed in. A third circumstance tending to confirm this opinion, is, that the of Col. Peach's march from Ellore to Warangole (in 1767), in which the distances were measured, and the angles of position taken by Major Gardner, places the latter only 37 G. miles from the position in which Hydrabad stands by M. Bufsy's line. It can hardly be deemed an impeachment of the general truth of a line

map

of 360 G. miles, measured after an army, that a position, in or near that line, should be ten or twelve miles out of the supposed line of direction. It is conformable to my idea of the distances of Nagpour, Cuddapah, and Warangole, that Hydrabad should be in 17° 12′, rather than in 17° 24'; and I have accordingly followed M. D'Anville: giving the lines between it and Aurungabad on the one side, and Condapilly on the other, a new direction accordingly.

Although by proportioning M. Bufsy's march from Bezoara, Hydrabad is placed in 78° 51' longitude; or only 114 G. miles from Bezoara, yet the different reports of the distance between these places, is much greater than the construction allows. For 114 miles will produce only 87 cofses, according to the proportion of 46 to a degree (which is the result of the calculation made on the road between Aurungabad and Masupalitam, page 5); whereas, one account from General Joseph Smith, states the number of cofses at 98; and another from a native, at 103. Again, Col. Upton reckoned 118 cofses between Hydrabad and Ellore, which the construction allows to be only 138 G. miles; or 105 cofses, according to the same proportion of 46 to a degree. So that I have either mistaken the longitude of Hydrabad, which is improbable, all circumstances considered; or the cofs is even smaller than I have supposed. Or the road leading through a hilly and woody country, is more crooked than ordinary:* and the journals remark its being very woody, and thinly inhabited, between Condapilly and Hydrabad. Until we have the latitude and longitude of Hydrabad, or some place very near it, we cannot be satisfied with its present position; for M. Bufsy's line is too long, to be exact, without the aids of latitude to check it. The reputed distance between it and Nagpour, 169 cofses, agrees perfectly with its corrected parallel of 17° 12′.

* General Smith's proportion of cofses to a degree, is 51; Col. Upton's 52; and the map by the native 55.

Hydrabad, or Bagnagur, is the present capital of the Nizams of the Deccan; who since the dismemberment of their empire, have left Aurungabad, the ancient capital; which is not only in a corner of their dominions, but in that corner which lies near their hereditary enemy, the Poonah Mahrattas; and which is also the least defensible. About 5 or 6 miles to the WNW of Hydrabad, and joined to it by a wall of communication, is the celebrated fortress of Golconda,* occupying the summit of a hill of a conical form, and deemed impregnable. When Aurungzebe conquered the kingdom of Golconda, in 1687, this fortrefs was taken pofsession of by treachery.

The next primary point or station, and one of the most important, as being the farthest removed from any other given point, in the whole construction, is Nagpour; the capital of the eastern division of the Mahratta empire, and nearly in the centre of India. This last consideration, and the number of roads issuing from it to the circumjacent cities, most of which roads had their distances given by computation only, made the determination of this point a grand desideratum in Indian geography. Mr. Hastings therefore, with that regard to useful science and improvements of every kind, which has ever distinguished his character, directed a survey to be made of the roads leading to it from the western frontier of Bahar; and also from the side of Allahabad. This was executed in 1782 and 1783, by Lieut. Ewart, under the direction of Col. Call, the Surveyor General. The result of this expedition was perfectly satisfactory. He began his measured line at Chittra, or Chetra, in Bahar, placed in 85° of longitude, and in lat. 24° 12′, in my map of Bengal and Bahar; and his difference of longitude from thence to Nagpour, in lat. 21° 8' 30", was 5° 16' west: by which Nagpour would be in 79° 44′. And from Nagpour back to Benares, in lon.

The termination, conda, or kond, signifies fortrefs, and often occurs in the south part of India; as cotta, and cote, which have the same signification, do in the north. Gur is used in the same sense occasionally in every part.

83° 13′, in the same map, he made 3° 25′ 10", difference of longitude, east; which placed Nagpour in lon. 79° 47′ 50′′; or 3′ 50′′ only, different from the other account; and this I suspect to arise partly from the error of his needle. If we close the account back again to Chittra, the place he set out from, he made only 4 minutes difference, in the distance out and home: and the road distance, was 600 B. miles from Chittra to Nagpour, only.

Taking the mean of the two accounts, the longitude of Nagpour will be 79° 45′ 55′′, or 79° 46'. The observations for determining the longitude at this place, by Lieut. Ewart, do not accord with the above account, by a considerable number of minutes: therefore I have not inserted them here, in expectation that they may be compared with corresponding ones, taken at places whose situations are already ascertained.

As Mr. Ewart's route to Nagpour, was by way of Burwah, Surgoojah, and Ruttunpour: and from thence to Banares, by Gurry, the capital of Mundella, he ascertained the positions of those places, satisfactorily; and by that means added to the number of primary stations. The latitudes were constantly taken, in order to correct the route, in detail. Nor did his work end here: for his inquiries at Nagpour, furnish a number of estimated or computed routes from that capital to Burhanpour, Ellichpour, Aurungabad, Neermul, Mahur, Chanda, &c. that is, in every direction except the SE; whence we may infer the state of that tract to be wild, uncultivated, and little frequented. And it appears by his intelligence, that the way to Cattack is unsafe, in any direction farther south than Sumbulpour.

Nagpour, the capital of Moodajee Boonslah, the chief of the eastern Mahratta state, is a city of modern date; and though very extensive and populous, is meanly built, and is open and defencelefs, save only by a small citadel; and that of little strength. The city is said, by Golam Mohamed, to be twice as large as Patna; but Mr. Ewart's account makes it but of a moderate size. Moodajee's

Ff

principal fortress, the depository of his treasures and valuables, is Gyalgur (called also Gawile) situated on a steep mountain, about 103 G. miles to the W by N of Nagpour. Each of the native princes in India, has a depository of this kind, and commonly at a distance from his place of residence: the unsettled state of the country making it necessary. The country round Nagpour is fertile and well cultivated, interspersed with hills of a moderate height: but the general appearance of the country at large, and particularly between Nagpour and Bahar, is that of a forest, thinly set with villages and towns. It is the western and northern parts of Moodajee's country, that produce the largest part of his revenue; together with the chout, or proportion of the revenues of Ellichpour, &c. held by the Nizam.

Ruttunpour is a city lying in the road from Bahar to Nagpour, and is the capital, and residence of Bembajee, who holds the government of the eastern part of the Nagpour territories, under his brother Moodajee. This place, also, has its position fixed very accurately by Mr. Ewart, in lat. 22° 16', lon. 82° 36'. It is a primary station of great use, as it regulates all the positions between Cattack and Gurry-Mundella; and between Bahar and Nagpour. As its corrected position differs only 3 miles from the former estimated one, collected from Col. Camac's observations and inquiries: it serves as an abditional proof, how much may be effected, by a careful examination and register, of the estimated distances on the roads: and this mode of improving the geography of India, may be adopted when all others fail. An intelligent person should be employed in collecting such sort of information, as Mr. Ewart collected at Nagpour; from the principal cities in the least known parts of Hindoostan; at the same time determining the position of such cities, by celestial observations: by which means, a number of fixed points would be established, from whence the computed distances might at once be laid off, and corrected. More could be done in this way, in a short time, towards completing the geography,