1. In the Arabian and Indian seas are winds, which blow six months one way, and six months the other, and are called Monsoons; by the accidental dispositions of land and sea it happens, that in some places the air near the tropic is supposed to become warmer when the sun is vertical over it, than at the line. The air in these places consequently ascends pressed upon one side by the north-east regions of air, and on the other side by the south-west regions of air. For as the air brought from the south has previously obtained the velocity of the earth's surface at the line, it moves faster than the earth's surface near the tropic where it now arrives, and becomes a south-west wind, while the air from the north becomes a north-east wind as before explained. These two winds do not so quietly join and ascend as the north-east and south-east winds, which meet at the line with equal warmth and velocity and form the trade-winds; but as they meet in contrary directions before they ascend, and cannot be supposed accurately to balance each other, a rotatory motion will be produced as they ascend like water falling through a hole, and an horizontal or spiral eddy is the consequence; these eddies are more or less rapid, and are called Tornadoes in their most violent state, raising water from the ocean in the west or sand from the deserts of the east, in less violent degrees they only mix together the two currents of north-east and south- west air, and produce by this means incessant rains, as the air of the north-east acquires some of the heat from the south-west wind, as explained in Note XXV. This circumstance of the eddies produced by the monsoon-winds was seen by Mr. Bruce in Abyssinia; he relates that for many successive mornings at the commencement of the rainy monsoon, he observed a cloud of apparently small dimensions whirling round with great rapidity, and in few minutes the heavens became covered with dark clouds with consequent great rains. See Note on Canto III. l. 129.