THE EARTH. 



233 



part of the ocean tliat lies between Africa and India, the east 

 winds begin at the month of Januaiy, and continue till about 

 the commencement of June. In the month of August or Sep- 

 tember, the contrary direction takes place : and the west wind ; 



winds soutli of it, to the lOih parallel: and finally, the east and south-east 

 trade-winds. In summer, the phenomena are less contradictory: South, 

 west wind^ frotn tlie lOth parallel to the northern limits ; trade-winds Boutli 

 of the 10th parallel. 



These general tendencies arc subject to variations, depending on the figure 

 and elevation of coasts, on straits, and currents of the sea. The north-west 

 and south.wett monsoons am weaker and mure variable in the Bay oJ Een. 

 gaj, more steady and violent in the gulf of Arabia. Both those monsoons 

 grow broader to the west, langing in this direction over the whole tract ot 

 tea that lies between Africa and Madagascar. In the seas extending between 

 China, the kingdom of Siam, Sumatra, and the equator, those monsoons are 

 UM likewise ; but here, excepting local variations, they are almost entirely 

 north and south. Ther extend as far as the Philippine Islands, and though 

 with much inconstancy, even to Japan. Between the equator, the islands of 

 Java, and New Guinea, the monsoons are nearly similar to those of the 

 Chinese Sea, in regard to their direction, which merely varies a little to the 

 north-west in the north munsoon, and a little to the south-west in the south 

 monsoon. But they do not begin till six weeks after those of the Chinese 

 Seas. 



Some other striking circumstances still remain to be noticed. The mon- 

 soons do not change, or, as t-ailors express it, do not break, of a sudden. 

 Their change, which usually takes plac» fifteen days or four weeks after the 

 equinoxes., is announced by the decay of the existing monsoon, by calms and 

 f quails in rapid succession, by storms, waterspouts, tornadoes, and by Indian 

 hurricanes, called taifnuns, particularly terrible from the explosions of elec- 

 trie matter accumulated by the monsoon. The beginnings of the subsequent 

 monsoon are, at first, liable to variations, till finally it establishes an absolute 

 dominion. 



Navigators assert, that (m quitting the region where a monsoon prevails, 

 one is sure, in ordinary circumstances, to fall in with a very strong and im- 

 petuous wind, blowing from a quarter directly opposite.- They must natu- 

 rally have observed this phenomenon with much care, since the calms and 

 whirlwinds it occasions are productive of great danger. It can hardly be ex. 

 plained, except by admitting, with Halley, the exibtence of two currents,— 

 one above, compi)?eii of warm and rarified air ; another below, ' onposed o. 

 the column of cold and condensed air. This hypothesis will become almost 

 a settled truth, if we obfcrve how small is the elevation to which the mon. 

 soou extends — a fart clearly exhibited in the peninsula on this side the Gan- 

 ges, where the mcmsoons are arrested for several months by the mountain 

 chain of the Gauts (not certainly of extraordinary height) ; so that the coast 

 of Coromandel, and that of Malabar, have always their dry and their rainy 

 seasons, at opposite periods of tlie year. 



According to the preceding description, it is the south-west monsoon alone 

 which presents any phenomena directly contrary to the general movement 

 of the atmoEphere ; for the north-east monsoon is in conformity with it, and 



U ii 



