the attempt, especially as the two fubjects feem on all occafions to be intimately related, or rather infeparably connected.

Clouds are generally believed to be formed by vapours raised by folar or fubterraneous heat from moisture in the earth, or in greater quantities from water itfelf, and when fo raised they are kept fufpended in the middle regions of the atmosphere in the form of clouds, until by fome means, not indifputably afcertained, the water is again precipitated to the earth in rain.

It may be neceffary to premife, that in treating of this fubject I fhall generally make ufe of the word vapour for that which arifes from water or any other fluid, and of the term exhalation for that which comes from the land.

By fome authors it is fuppofed, that both vapours and exhalations are fmall veficulæ detached, as be fore obferved, from the earth or water by heat, and which must be fpecifically lighter than the air, or they could not afcend. When they have paffed through the denfer medium near the earth, attracted by the dry air above them, they continue to afcend till they arrive at a cold region, where they become con. denfed, and remain fufpended, as before obferved, in the form of clouds. In this ftate they continue floating, till by fome new agent they are converted into rain, hail, fnow, mift, &c. Others again, who equally admit that the clouds are formed by thefe veficulæ, think that they coalefce in the upper regions of the atmosphere, forming into little maffes, until they become too heavy to be any longer fuf. pended, and then defcend in rain. But this hypothefis cannot be well founded, for the vapours are perpetually afcending into the upper

regions of the atmosphere, which are always cold, and confequently, according to this theory, they would again be precipitated in rain as foon as they have arrived at a certain height, which would almoft conftantly produce regular fhowers," The fame objection applies to the fyftem of Dr. Derham, who accounts for rain by fuppofing the veficula to be full of air, which (he fays) becoming contracted in the colder regions, the watery fhell, thus thickened, becomes heavier than the air, and is precipitated in rain by its comparative weight. But Dr. Franklin, in his Obfervations on Electricity, feems to account moft rationally for the formation of the clouds and precipitation of rain." The fun fupplies," he fays, " (or feems to fupply} common fire to all vapours raised from the fea, or exhalations from the land. Thofe vapours which have both common and electrical fire in them, are better fupported than thofe which contain only com. mon fire; for, when vapours rife into the coldeft region above the earth, the cold will not diminish the electrical fire, if it doth the common. Hence clouds formed by vapours raised from fresh waters, within land, from growing ve getables, moift carth, &c. more fpeedily and easily depofit their water, having but little electric fire to repel and keep the particles feparate. So that the greateft part of the water raifed from the land is let fall on the land again; and winds blowing from the land to the fea are dry, there being little ufe for rain on the fea, and to rob the land therefore of its moisture, in order to rain on the fea, would be contrary to the unerring diftribu tions of nature.

"But clouds formed by vapours

rain which falls in all thefe different countries, is nearly in proportion to the extent of the ocean near which they are fituated, and to the height of their different mountains. The temperature of the fea in all of them is fuppofed to be nearly the fame.

The Gulf of Sind is much more fpacious than the Gulf of Bengal. The coaft of Malabar is much more mountainous than that of Coromandel; and therefore, although the wind blow nearly an equal length of time on either coaft, the hills on the former being more numerous, and the fea more fpacious thar, the latter, the rains on the Malabar coaft will be more abundant than thofe on the coaft of Coromandel.

The South Atlantic Ocean is infinitely more extenfive than the Gulf of Sind. The American mountains are, beyond all comparison, higher than any in the peninfula of India, and the N.E. and S.W. winds alternately prevail the whole year on the east coast of America; whereas, on either coaft of India, the monfoon is never faid to blow above fix months, and, to speak ftrictly, does not continue above half, that time. For all these reasons, therefore, the rains in America are probably more abun. dant than on either coaft of the peninfula of India. I fay probably more abundant, as I am not in poffeffion of any meteorological tables kept in Brazil; and during an accidental refidence of ten weeks at the Bay of All Saints, it was not in my power to obtain much information from the Portugueze on thofe fubjects. For the prefent, therefore, we muft form our opinion of the quantity of rain which falls, from the number and mag

nitude of the rivers. This mode of judging will, without doubt,

be deemed exceptionable; for in all probability the Amazon and La Plata, the two principal rivers of America, are confiderably augmented by the fnows which are melted among the Cordilleras, where their fources are fuppofed to lie. But if we admit this objection to be well founded, it ftill may be contended, that the numberless inferior contributary ftreams, which in any other fituation would be deemed very confiderable rivers, and are, in fact, much larger than thofe on the fouthern part of the peninfula of India, are most of them formed by vapours conveyed by the ef winds from the fouthern Atlantic Ocean to the inferior branches of the Andes, where they are com. pelled to precipitate their contents in the manner defcribed by Dr. Franklin. It may, perhaps, be thought that this comparison, and all the inferences deduced from it, are equally applicable to other countries, and particularly to Africa. But the geographer will foon find, that although the continents of Afia and America fomewhat refemble each other, both in fhape and fitu ation, yet Africa, which alfo terminates in a cape to the fouthward, having no ridge of lofty mountains running from N. to S. like the two former continents, the rivers of that country are few in number, and thofe comparatively fmall. But to return to a further account of the monfoon on the coaft of Coromandel.

The island of Ceylon, which lies to the fouthward of the Coromandel coast, and where the peninfula becomes extremely narrow, partakes of both monfoons, but principally of the S. W. The wind immediately on the coaft, at the commencement of the monfoon, takes nearly the fame direction as

the

the coaft itfelf. From the latitude of 9 to 13 degrees the coaft lies N.N.E. and S.S. W. and from the laritude of 15 degrees to the head of the gulf, calied Balafore Roads, it runs almoft N.E. and S.W. The S.W. monfoon, therefore, on this coaft blows at firft along thore, from which caufe it is called the long fhore wind. The nature of the foil on the coaft probably contributes to give it this direction; for the foil being, in fome refpects, like the Gulf of Guinea, on the coaft of Africa, low and fandy, the air near the earth must confequently be much rarefied under almott a vertical fun, and the deafer air coming across the Indian Ocean or the Gulf of Sind, will follow that direction on the coat to fill up the But thefe winds continue only to the end of May or the begiving of June, when the fun being near the fummer folftice, the hot land wind on the coaft of Coromandel commences, and continues about fix weeks. To underftand the caufes of this fudden change, We nut again advert to the geography of the country, and confider the ftate of the atmosphere at this period on the two coafts.

vacuum.

The fouthern part of the peninfula, from the latitude of 16 de. grees to Cape Comorin, may be divided longitudinally into three parts, beginning at Madras, which is fituated in the longitude of so 28′ 45" E. About two degrees to the weftward of that meridian is a range of mountains, forming the eastern boundary of the Valley of Baramaul, where the high land of Myfore commences, commonly called the Ballagat, or country above the Paffes. This high or table land of Myfore rifes at leaft 2000 feet above the coaft of Coromandel, and runs through the pe,

ninfula from N. to S. nearly in the longitude of 78 degrees. Two degrees farther to the weftward is another range of mountains, which may be confidered as the boundary of the Malabar coaft; and the country fituated between these two meridians, from 76 to 78 degrees, is properly the country of Myfore. With this sketch of the map of the country before us, and with a recollection of the firft principle of this hypothefis, it will not be diffi cult to account for the hot land wind prevailing in the Carnatic during the months of May and June.

The fun's declination in the month of May is between 15 and 22 degrees north; he will therefore, before the end of this month, have been vertical over all there countries, and confequently have produced a confiderable degree of heat in the Carnatic; but at the fame time the double range of mountains to the weftward will have arrefted the clouds brought thither by the S. W. monfoon, and made them precipitate their contents both on the Malabar coaft and in the Myfore country. The principal point of rarefaction then, at this feafon, will be the Carnatic, which may, as ufual, be confidered as the heated room, and the nearest cold body of air will come from the table land of Myfore to rcftore the equilibrium.

In the Carnatic, during the months of May and June, the thermonter of Farenheit, in the fhade, is generally at 90 or even 100 degrees and upwards; whilft near the mountains the fame kind of thermometer will not be more than 70 or 80 degrees at the utmoft. The current of air then will move from the mountains across the Carnatic towards the coaft of Coromandel,