Until Mr. Dalrymple furnished me with materials for fixing the longitude of Cape Comorin, by means of the measured lines abovementioned between Devicotta and Madura, the difference of longitude between Madras and the Cape, was variously, and in all cafes, erroneously described. For inftance,

fo that the nearest of the above calculations, is 27 from the truth.

The particulars of the coaft between Madras and Cape Comorin are from various hands. The furvey of the Company's lands (or Jaghire) extends beyond Sadras. From thence to Point Calymere is from M. D'Apres; and from thence to Tondi, chiefly from my own obfervations in 1763 and 1764. Tondi itself is laid down by a furvey from Tritchinopoly and Madura. From Tondi to Goodwater island is from actual furveys of Major Stevens's: thence to Tutacorin from Mr. Delafield's curfory furvey: and the remainder to Cape Comorin, is from Col. Call's map of Tinevelly, inferted in Mr. Orme's History.

Between Cape Comorin and Cochin, in lat. 9° 58', M. D'Anville's, and other maps, allow 1° 25′ difference of longitude, west; which, taken from 77° 53′, leaves 76° 28′ for the longitude of Cochin. The Rev. Mr. Smith's obfervation in 1776, places Cochin in 76° 26' 30", which is only 1' 30" more wefterly than what is given by the bearing.

After the coincidence of these facts, it can hardly be doubted that the longitude of Cape Comorin, taken at 77° 53', is right; for whether we take it by the interfection of the two bearing lines, drawn from the places of observation, Pondicherry and Cochin; or fimply by either of the bearing lines, corrected by the latitude;

the

the refult will be the fame to two minutes of a degree, or lefs; an approximation seldom to be attained, in cafes of this kind.

In the map, Cochin (or at least the strand of it) is placed in 76° 27. Between it and Bombay, an extent of near 600 G. miles, we have neither furveys, nor obfervations of longitude to correct any of the intermediate points: for the obfervation at Goa can hardly be deemed an exception, when it is known to differ 32′ from the longitude deduced from the observations at Bombay: and this deduction is founded on the bearings of a coaft of about 120. G. miles in extent, nearly straight, and in a direction only 25 degrees from the meridian. Moreover, the bearing is corroborated by many authorities, amongst which are thofe of Meffrs. D'Anville and D'Apres:

Taking Cochin at 76° 27′, and Bombay at 72° 40′ *, the difference of longitude by obfervation will be 3° 47′ weft. Now, M. D'Anville, without any attempt towards correcting the longitude (as he was not furnished with any observation on that fide of India, save the one at Goa) makes it 3° 50′ 30′′, or only 3′ 30′′ different from the observation. M. D'Apres' makes it 4° 12′, or 25′ greater than by the observations. It must be observed that there is nearly a degree and a half of wefterly variation + on this coaft; which, if allowed, brings M. D'Apres' difference of longitude to the fame as that by observation. Whether or not, he allowed it, I cannot determine.

• The hon. T. Howe took two obfervations of Eclipfes of Jupiter's firft fatellite in 1762, at Bombay, by which longitude was found to be 72 36 72 40 72 45

The Rev. Mr. Smith in 1776 made it

And

Medium 72 40

But by a mistake of the engraver, Bombay is placed in 72° 38′ 40′′, or 1' 20" farther to the weft than was intended. The latitude is

+ Variation at Goa in 1763

at Tellicherry

at Cochin

18° 57′
50'

at C. Comorin
1777 at Anjenga

M. D'Anville comes nearer than could be expected in fo great an extent of space: and I think, upon the whole, we may conclude that there is very little difcordance between the observations, furveys, and charts of the coafts, between Madras and Bombay.

The observation of longitude at Goa, recorded in the connoiffance de Temps, (73° 45′) has been made ufe of by all the late Geographers, in fixing the longitudes on the west of India; it being the only one that had been made on the west of Pondicherry, till the year 1762. By whom it was taken, or in what manner, I am totally ignorant. And to allow it any degree of credit, we must reject, not only the observations at Bombay and Cochin ; but also the measured lines between Divicottah and Cape Comorin.

I have followed M. D'Apres' general outline of the coast of Malabar; but have corrected it in many particulars by the aid of MSS. furnished by my friend Mr. Dalrymple, and others.

:

It is very much to be lamented that there are no obfervations of longitude, to correct the bearing of a coast of so large an extent as that between Cochin and Bombay; or that a measured line was not drawn across some part of the peninsula; as for instance, between Madras and Mangalore. I understand, the distance between Madras and Ooscotta, which is more than half way over, is already measured so that there remains only about 150 G. miles between Oofcotta and Mangalore. This, however, can hardly be expected, whilft Hyder Ally continues mafter of Mysore and its dependencies. But the longitude on the Malabar coast, it is hoped, will foon be ascertained by means of the time keepers on board the East India Company's fhips, as they are now coming much into use; and as the nature of the navigation on that coaft, affords frequent opportunities of taking regular feries of obfervations.

For my own part, I have little doubt of the truth of the intermediate fituations between Cochin and Bombay, yet pofitive proofs are always to be defired, and fought after, although we may be in poffeffion of the strongest presumptive ones, or the best grounded deductions. By

5

By the observations of longitude at Madras, Pondicherry, Cochin, and Bombay, it appears that the peninsula has hitherto been represented about 46 miles of longitude wider than it really is *, and of this excefs 39' are already accounted for between Madras and Cape Comorin; M. D'Anville's chart making 3° 15′, and mine only 2° 36′ difference of longitude. See page 24.

Bombay is the westermoft point in the map, that is fixed by an observation of longitude. A furvey of the Gulf of Cambay, and coast of Guzerat, from St. John's Point to Diu Point, has been taken by Lieut. Skynner, by order of the Presidency of Bombay; and the fame has been done of Salfette Ifland, including Baffeen; but unfortunately the interval between Baffeen and St. John's Point (about 42 G. miles) has been omitted. I have therefore been under the neceffity of taking the bearing from M. D'Apres, who makes it about N. by W. The difference of latitude is known to be about 38 minutes; and the bearing being so nearly meridional, there cannot be much error, if any, in the difference of longitude. The longitude of this Point, thus deduced from Bombay, is 72° 23': and that of Surat, by the furvey from the Point, 72° 45'.

Diu Point is placed according to the longitude given by Lieut. Skynner's Survey from St. John's Point, which places it in 69° 47′

From Diu Point to Cape Mons, beyond the mouth of the Sinde, the bearing is taken from the medium of three charts, furnished by Mr. Dalrymple; and appears to be N°. 41° 20' W +; and the diftance, corrected by the latitudes of Diu Point and Cape Mons, the former of which is 20° 44', and the latter ‡ 25° 7′ 30′′, gives

3° 58' difference of longitude, weft; and confequently that of Cape Mons 65° 49'. M. D'Anville places this Cape 1° 1' more eastwardly, or in 66° 50′, and Bombay in 71° 47′; making the difference of longitude between them 4° 57', inftead of 6° 51', as given in my map and this difference of near two degrees in feven, makes a great alteration in the figure of the coaft between Surat and the mouth of the Sinde, or Indus: the peninfula of Guzerat being much larger than was formerly supposed, the gulph of Sinde (or Cutch) much smaller; and the Delta of the Indus projecting into the fea, instead of receding from it.

All the particulars of the western coast of Guzerat, and the mouth of the Sinde, are copied from the abovementioned MS. charts of Mr. Dalrymple's: and confequently the whole coast from St. John's Point to Cape Mons, is defcribed from new materials. I now return to Balafore.

From Balafore, eastward to Chittigong, the distance has been determined by the inland furvey; and the figure of the coafts and inlets by Captain Ritchie's Sea Survey. The difference of longitude between the towns of Balafore and Chittigong (or Islamabad) is 4° 53′ caft; and, it is worthy of remark, that the distance by Capt. Ritchie's marine furvey, agreed with the measurement on fhore, to within two miles and a half.

The charts as late as the year 1752, reprefented the difference of longitude between these two places, to be only 3° 48'; that is, 1° 5' less than the truth. And this diminution of the longitude, whilft the difference of latitude continued right, gave the fea coast between the mouths of the Ganges, a direction of two points, or 221 degrees more northwardly than the truth; which doubtless occafioned the lofs of many fhips, who trufted to the information.

From Islamabad, in longitude 91° 55′ latitude 22° 20′, the coafts of Aracan and Pegu take a S. S. E. courfe to Cape Negrais, the extreme point of Pegu to the fouth weft; the latitude of which is

under