SUMATRA. 
The periodical winds which are fuppofed to blow during fis months 
from the N- W- and as many from the S, E. rarely obferve this regula- 
rity, except in the very heart of the monfoon ; inclining, almoft at all 
times, ieveral points, to. feaward, and not unfrequently blowing from 
the S. W. or in a line perpendicular to the coaft. This muft be attri- 
buted to the influence of that principle which caufcs the land and fea 
windsj proving on thefe occafions more powerful than the principle of 
the periodical winds ; which two always adl at right angles with each 
other. It thefe were of equal power, the current of air would take a 
middle dire^bion, and conftantly blow, on Sumatra, from the W, point, 
during one monfoon, and from the S. point during the other : — ^and as 
the influence of eithet is prevalent, the winds approach to a courfe per- 
pendicular to, or parallel with the line of the Coafl:. The tendency of 
the land wind at n-igKi-, hn% ainiofl: ever a correfpondence with the fea 
wind of the preceding or following day ; (except when a fquall or other 
fudden alteration of weather, to which thefe climates are particularly 
liable, produces an irregularity) ; not blowing in a diredion immediately 
oppofite to it ; which would be the cafe* if the former were, as fome 
writers have fuppofed, merely the efledt of the accumulation and redun- 
dance. of the latter, without any pofitive caufe; but forming an equal 
and contiguous angle, of which the coaft is the common fide. Thus, 
if the coaft be conceived to run N. and S, the lame influence, or com- 
bination of influences, which produce a fea wind at N* W. produce 
a land wind at N, E. or adapting the cafe to Sumatra, which lies 
N. W. and E,, a fea wind at S. is preceded or followed by a land 
wind at E. This remark muft not be taken in too Itridt a fenfe, but 
only as the refult of general obfervation. If the land wind, in the courfe 
of the night, lliould draw round from Er to N. it would be looked u^xin 
as an infallible prognoftic of a W. or N. W. wind the next day. On 
this principle it is, that the natives foretell the dire<ftion of the wind, 
by the noiic of the furf at night, which if heard from the northward^ 
is cflieemed the forerunner of a northerly wind, and vice verfd. The 
c^uarter from which the noife is heard, depends upon the courle of the 
land wind^ which brings the found with it, and drovyns it to lee- 
F ward-- 
