from the greatness of its extent, and other circumstances, incapable of defence; must be considered as a great political evil in a state: it is like a fortrefs that exposes its weakest part to the enemy, and points his attacks: and, to pursue the allegory, there may be some danger of the garrison's sacrificing the interest of the empire at large, in order to preserve their own property, in the hour of assault. The Scythians, who were not chained to the soil, could never be conquered: and those who have no large capitals, stand in the next degree of security; all other circumstances taken into the case. If the question be considered, as it concerns morals, the objections are yet stronger: for the larger the capital, the greater will be the proportion of the population that is corrupted.

Amedabad, the capital of Guzerat, was the extreme point of General Goddard's marches to the northward, in the province of Guzerat. In the first Section, a comparison was made between the scale and bearing of the map of General Goddard's marches in Guzerat, and those of the surveys taken between Surat and the Myhie river; and it was found to agree so well, that the line between Brodera and Amedabad might be very safely adopted. The bearing was N 36° W, and the distance 53,2 G. miles; giving for the position of Amedabad, 22° 58' 30" lat.; and 72° 37' lon.; or 3' west of Bombay. By M. Thevenot's account, the latitude is 23o and some odd minutes: and 23° by the Ayin Acbaree.

Amedabad is a very considerable city, and succeeded Mahmoodabad, as capital of Guzerat. It is one of the best fortified cities of Hindoostan; and made a good defence when taken by General Goddard in 1780. On the peace of 1783, it was restored to its former pofsefsors, the Poonah Mahrattas. Travellers have dwelt much on the beauty, and convenient situation of this city, which is in a level country, and on the banks of a small navigable river, named Sabermatty; and which, together with other confluent streams, falls into the head of the gulf of Cambay, near to the city of that name. Cambay, is indeed, the port of Amedabad, and is E e

distant from it about 56 road miles. It is a large city, and appears to be the Camanes of Ptolemy; although the gulf, which is now denominated from Cambay, had then its name from Barygaza, or the modern Baroach.

Aurungabad is a point of considerable importance to the construction of the western part of the tract in question; and although we have neither its latitude, longitude, nor distance accurately measured, from any one point; yet the sort of coincidence that arises between a number of estimated routes, from six different places, in opposite directions, round it; imprefs a certain conviction of its being placed nearly in its true position. It will be necessary to particularize the principal of these routes. One of them regulates also the positions of Hydrabad, Beder, and Mahur; and is that of M. Bufsy from Masulipatam. The copy from whence I collected my ideas on the subject, is that included in the late Mr. Montresor's map of the southern part of India. As his map goes no farther west than Aurungabad, we may conclude that he has not altered the original bearing and distance, with a view to reconcile its situation to any other place to the north or west.

Masulipatam is already placed in the map, in lat. 16° 8′ 30′′, lon. 81° 12′, on the authorities of Col. Pearse and Capt. Ritchie (see page 12). This is a city and port of trade, near the mouth of the Kistna river; and appears to be situated within the district named Mesolia, by Ptolemy. Between this place and Bezoara (or Buzwara) a fort on the north side of the Kistna river, M. Bufsy's route allows only 36 G. miles; but as there is existing, a map of Major Stevens's, which fixes the said distance at 40,3 G. miles, I have adopted it; and allowed M. Bussy's authorities to commence only at that point. Bezoara, so placed, is in lat: 16° 33'; and lon. 80° 39′. Then from Bezoara, to Aurungabad, the bearing is given. at W 35° 10′ N, 323 G. miles; producing 3° 6' difference of latitude; and 264 of westing; or difference of longitude (in lat. 18°) 4° 38'. This would place Aurungabad in lat. 19° 39′, lon. 76° 1′.

Bufsy's (or rather Montresor's) whole distance from Masulipatam to Aurungabad, was 359.

Let us now examine what data we have to check this long line of M. Bufsy's, from the side of Surat, Poonah, and Burhanpour. The position of Surat has been just accounted for: and Noopour, a city on the road from Surat to Burhanpour, is by Goddard's route 59′ of longitude to the east of Surat; or in lon. 73° 47′ 15′′. And from this place to Aurungabad, Tavernier reckons 105 cofses; which, at 42 to a degree, give 150 G. miles of horizontal distance. Now, Noopour, Aurungabad, and Bezoara, lie as nearly as possible in a right line, whose extreme length is 475 G. miles. Tavernier's 150, added to Bufsy's 323, make up 473; or the whole space, within two miles. But from the nature of a march of an army in a warm climate, great part of which is often made in the night, it must necefsarily require correction; in the bearing at least, and probably in the distance too. Nor can the 105 cofses of Tavernier, be expected to be even so correct as the march: it is therefore a matter of surprise that only so small a difference should have arisen. It should be remembered that 4,3 miles were added to M. Bufsy's original distance, between Masulipatam and Bezoara; so that the whole original error was 6,3; if we do not refer a share of it to Tavernier's estimated distance. It is proved in another instance by Major Gardner, in Peach's march from Ellore towards Warangole, that M. Bussy's geographer has given too little distance. This is probably an error of the compiler, not of the surveyor; it being an error of a different kind from what might have been expected, in the ordinary way of measuring distances with a perambulator.*

The latitude of Aurungabad is inferred from its distance from Burhanpour given by Golam Mohamed † at 66 cofses; and as the

• That long distances may be accurately measured by a perambulator, I need only mention, that during the Bengal survey, I measured a meridian line of three degrees, with a perambulator; and found it to agree minutely with the observations of latitude. However, due allowance was made for the irregularities of the ground, whenever they occurred. The country, indeed, was in general, flat the whole way.

A confidential person, sent by Col. Camac, in 1774, to explore the roads and country of the Deccan, and to gain intelligence concerning the Mahratta powers.

bearing is not far from meridional, we may state the difference of latitude at 1° 34'; which taken from 21° 19′, the latitude of Burhanpour, leaves 19° 45', for that of Aurungabad.* Now, M. Bufsy's line, gives only 19° 39′; which is 6' too far southwardly, by this account. If 19° 45′ be adopted, some further addition must be made to the line of distance from Bezoara; but it is too trifling à matter to require discussion. In effect, the longitude of Aurungabad by these data, will be 76° 2′ 30′′; lat. 19° 45'.

Two more lines of distance are given from Nimderrah Gaut and Bahbelgong; two points in Mr. Smith's route, on the west and SW of Aurungabad. Nimderrah is in lat. 19° 12′ 45′′, lon. 74° 54′ 30′′: and Bahbelgong in lat. 20° 45′, lon. 74° 51′ 30′′. M. Anquetil du Perron furnishes these distances. That from Nimderrah to Aurungabad, he reckons 32 cofses; and that from Bahbelgong 34. Now, as the distance between Poonah and Nimderrah, is known, it furnishes a scale for the rest of his route. He makes this distance 34 cofses; but it is clear that he reckoned by some other standard than the common cofs (pofsibly he reckons leagues and cofses the same, as we shall have occasion to remark in his route from Goa to Poonah) for the distance being 69,7 G. miles of horizontal distance between Poonah and Nimderrah, it should rather be 48 cofses, than 34. However, taking his distance for a scale, whatever the denomination may be, the distance between Nimderrah and Aurungabad, will be 64,7 C. miles; and that from Bahbelgong, 70,2. And the mean of these accounts give also, 76° 2′ 30′′ for the longitude of Aurungabad.

There is yet another line of distance to Aurungabad, and that is from Nagpour; whose position is ascertained with precision. Two accounts of the estimated distance between them, collected by Lieut. Ewart, are, 162, and 165 cofses: the mean of which, 163, at 42 cofses to a degree, is 233 G. miles of horizontal distance. This

M. D'Anville reckons the same difference of latitude between the two places, but he has placed both of them too far north by 24 minutes; following I apprehend, the latitude of Burhanpour, given in the Ayin Acbaree.

would place Aurungabad, admitting its latitude to be 19° 45', in 75° 53′ 30′′ or 9' to the west of the other accounts. The result of the short distances, are doubtlefs to be preferred to that of the long ones; and I insert this last only to shew the extremes of the different accounts.

Lastly, if the distance from the four nearest points are taken; that is, from Noopour 150 G. miles; Burhanpour 95; Nimderrah 64,7; and Bahbelgong 70,2: the mean point between the intersections of these, will be in lat. 19° 44', lon. 76o.

Although I have taken the latitude of Aurungabad at 19° 45', as its distance from Burhanpour is so nearly meridional: yet the intersections of the other distances, point to its being in a lower latitude, by 4 or 5 minutes: in which case, its position would also be somewhat more westerly.

Upon the whole, I have placed Aurungabad in lat. 19° 45', lon. 76° 2' 30"; and by what has been said, it cannot be much out of its true place: but as it is a point of great importance in the geography of this part of India, it required particular discussion; being the centre of several roads; and the bearing of that long line, between it and Hydrabad, Beder, Calberga, &c. depending on it.

Aurungabad is but a modern city; owing its rise from a small town, to the capital of the province of Dowlatabad, to Aurungzebe; from whom also, it had its name. When the Deccan became a province of the Mogul empire, it became the provincial capital; and continued to be so, after the Nizams became independent of Delhi; and until the encroachments of the Poonah Mahrattas, of late years, made it an uncomfortable residence to the Nizam. When the Deccan was first invaded by the Patan emperors of Delhi, Deogire was the capital of the province of Dowlatabad, and was situated near the fortrefs of the same name; which is built on a mountain about 4 or 5 cofses to the NW of Aurungabad; and is deemed impregnable by the people of the country.