general map and Gurry Mundlah, is placed according to the eftimated bearing and distance from Tetwarra.

Nagpour, the modern capital of Berar, and the refidence of Moodagee Roonflah, is said to be 61 coffes fouthwesterly from Tetwarra. Golam Mohamed reckons 82 coffes between Gurry Mundlah and Nagpour; which, confidering the refpective pofitions of the three places to each other, is 3 coffes more than from Tetwarra. And again Golam Mohamed reckons 166 coffes between Nagpour and Aurungabad; but he remarks on his approach to Aurungabad,. that the coffes of this country are fhort". Placing Nagpour 61 coffes from Tetwarra, an interval of only 161 coffes (of 42 to the degree) will be left between Nagpour and. Aurungabad: which difference of coffes must be accounted for in the fhortness of the measure remarked by Golam Mohamed.

5

Between Burwah in Bahar, and Nagpour, are reckoned 206, coffes; that is 86 to Ruttunpour, and 120 from thence to Nagpour.. But having placed the latter according to its reputed position with refpect to Tetwarra and Gurry Mundlah, that is to say, in latitude 21° 50', longitude 79° 36', I can make no more than 198 coffes between it and Burwah; or, 8 lefs than the computation. It may be that the road winds more than ordinary; or it may be rougher; which, in the idea of the Indian travellers, as well as of all others,. lengthens the absolute distance. Laftly, it is only 77 coffes from Hufnabad Gaut, as I have placed it +, altho' Gen. Goddard reckons it 100. The roads from Burwah to Ruttunpour, and from thence. to Gurry Mundlah, Nagpour, Aurungabad, and Burhanpour; are: all from the Itinerary of Golam Mohamed.

The country between Mirzapour and the head of the Soane, was explored by Capt Bruce, who fo eminently diftinguished himself at the escalade of Gwalior in 1780. During his expedition, he

Shawpour was the ancient capital.

+ It happens that Hufnabad Gaut was placed in its true pofition, previous to my féeing the map of Gen. Goddard's march.

2

verified

verified a fact which had been long doubted, though ftrenuously infifted on by the natives; (viz.) that the Soane and Nerbudda rivers had their common fource from a pond, or lake, on the fouthern confines of the Allahabad province. These rivers do literally flow from the fame lake; making, conjointly with the Ganges, an island of the southern part of Hindooftan: and flowing in oppofite directions 1500 miles. The course of the Nerbudda river is ascertained, only in certain points where it happens to be croffed by any of the great roads here defcribed: excepting only in the neighbourhood of Broach. All the intermediate parts are drawn from report. It is represented to be as wide at Hufnabad Gaut, as the Jumna is at Calpy but fordable in most places, during the dry season.

The Soane is drawn in the fame manner as the Nerbudda, in that part of its courfe above the boundary of the Bahar province: as are alfo the Tapty, Cane, Betwa, Chumbul, Sinda, and all the other rivers that flow within the bounds of the tract fpoken of in this Section.

The road from Agra to Amedabad, has its particulars from D'Anville, who appears to have taken it from Tavernier's Itinerary. I have altered the fcale of it, in order to make it correfpond with the difference of longitude (in my map) between Agra and Amedabad, which is about half a degree less than in M. D'Anville's: occafioned by his having adopted the observation taken at Goa. Throughout this long line (450 B. miles) there is only one cross distance to correct the bearing of it; and that is, Sir Thomas Roe's route from Burhanpour to Agimere. He reckons the distance 222 coffes; of which 66 are between Burhanpour and Mundu; 105 between Mundu and Cheitore *; and the remaining 51 between Cheitore and Agimere.

All Tavernier's distances appear to be too great between Agra and Amedabad, according to the standard of the cofs, in page 18;

The ruins of this famous place are laid down at least 80 miles to the fouth of their true pofition in all the maps that I have seen,

owing

owing probably to the roughness and hilliness of the road, which makes the horizontal distance lefs than in countries that are more fmooth and open.

The upper part of the courfe of the Puddar is likewife from D'Anville but this river certainly does not run into the head of the Gulf of Sinde, as described by him. I have not only the authority of a MS. map for this affertion, but also that of Capt. Jofeph Price, who had the misfortune to be carried up to the head of this gulf by Pirates (after à most gallant and obftinate defence of two days). He affures me that nothing larger than a rivulet falls in there. I therefore conclude that the Puddar (or at least a principal branch of it) falls in at Cutch: as a MS. chart defcribes the mouth of a large river near it, by the name of Mud River ; which name is characteristic of a river of the fize, and length of course of the Puddar.

Capt. Price, who also travelled across the Ifthmus between the Gulfs of Sinde and Cambay, eftimates the width of the Ifthmus at 120 or 130 G. miles. The country is generally very flat, with a few small eminences, fortified. The foil is dry, and unproductive of rice.

The inland part of the peninsula of Guzerat, is from M. D'Anville.

The position of the ancient and famous temple of Sumnaut (Sunnam Sumnaut) is collected from the Ayeneh Acbaree, and from Ferifhta. In the former, it is called Puttan Sumnaut, from its vicinity to the town of Puttan near Diu; which town is ftill in being. And by Ferifhta it is faid to be "within the districts of Deo, or Diu *”, The latitude is given at 22° in the Ayeneh Acbaree; but it is hearer 21°.

The Broach district, and in general all the inland parts of Guzerat, from Surat to Amedabad, and from the fea to Dubhoy; are

Dowe, Vol. 1ft. 66.

from

from surveys taken fince the commencement of the Mahratta war. The tracing of the Nerbudda goes as far up as Bowapier Gaut, that is, about 80 British miles from the fea.

The fituation of Pawengur, is from Mr. Farmer. Those of Narwalla, Champanier, Puttan, Nadout, &c. are either from the Ayeneh Acbaree, or from Ferishta.

Mundu, and the road from thence to Burhanpour, is from Sir Thomas Roe, who places Mundu 66 coffes from the latter, on the road to Agimere. Ougein, and the road from Mundu to Sirong, is from M. D'Anville; and Indoor from the information of Mr. Farmer. The divifions of Malwa and Guzerat, are from the Ayeneh Acbaree.

Hindia alfo is from the fame authority. I take it to be the Andi of Tavernier. It is astonishing how he could so far mistake the course of the Nerbudda river at that city, as to fuppofe that it fell into the Ganges.

The road from Burhanpour to Patris is from D'Anville; those from Noopour to Aurungabad, from Tavernier: and thofe from Aurungabad to Amednagur; from Bejapour towards Calberga; and from Nander to Enegil *; are from Mr. Jefferies's map, published in 1768. He, I understand, took them from fome French MSS. The road from Aurungabad towards Sanore-Bancapour, and from Hydrabad to Edghir, are from M. Buffy's marches.

Sourapour, or Sollapour, Callian, Palkee, &c. are from a MS. lent me by Mr. Townsend.

Meritz and Panchmal are from the Eaft India Company's records; as are also Deckenal and Khoud.

Chunderi, Rintimpour, Toumoun, are inferred from Ferishta and the Ayeneh Acbaree. Gurry-Baundhoo, and a number of places in and about Bundelcund, are from a Hindoo map of that

* This I take to be part of the route of M. Buffy towards Nagpour, mentioned by Mr. Orme, Vol. 1ft. 373. Should it be fo, and the direction of the road in Jefferies's map be right; it corroborates the affumed fituation of Nagpour in my map.

country,

country, the names in which, were obligingly tranflated by Mr. Boughton Roufe.

Bejapour, or Vifiapour, is not fo well afcertained as might be wished. Mandefloe, who travelled the roads himself, fays, that it is 80 leagues from Dabul, on the coaft of Malabar; and 84 from Goa; which, if meant of French leagues of 3000 paces, with an allowance of one in feven for windings, will give 142 G. miles from Dabul, and 149 from Goa: making Bejapour in lat. 17° 36'; lon. 75° 44'. P. du Val, who formed a map of Mandefloe's routes, (Vide British Museum) and probably had lights, befides what are furnished by the travels, to guide him, makes the distance between Dabul and Bejapour, greater than between Bejapour and Goa. And this I think likely to be the cafe, though contrary to what is faid in the travels.

Tavernier reckons 85 coffes from Goa to Bejapour (or rather perhaps from Bicholim, the landing place on the continent) or 8 days journey and fays that from Bejapour to Golconda, the distance is 100 coffes more. This account must be exceedingly vague; becaufe Goa and Golconda are more than 200 coffes afunder, by the directeft route; and Bejapour lies more than 40 coffes out of the line.

The Lettres Edificantes make the latitude of Bejapour 17° 30'. I have endeavoured to take the medium of these difcordant accounts, and have placed it in latitude 17° 21', lon. 75° 52'; that is, 94 coffes from the coaft oppofite to Goa; 111 from Dabul; and 122 from Golconda. At the fame time, I think it probable, that it may not be within 20 miles of its true pofition. The particulars of the roads leading to it from Dabul and Goa, are from P. du Val's map.

Raölconda, afamous diamond mine, is placed in Mr. Montrefor's map, about 15 G. miles to the weft of Ralicotte, and 12 from the north bank of the Kiftna; but I know not on what authority. Tavernier, who vifited both Raölconda and Coloor, gives the dif

L

tance