OBSERVATIONS TAKEN IN INDIA. 
373 
The winds of the S.W. monsoon, however, terminate in July instead of October. 
This is the more remarkable, as Dodabetta lies between Madras, where these winds 
are the prevailing winds of May, June, July, August and September, and Bombay, 
where the same winds prevail in the same months. It is probable, therefore, that 
Dodabetta is situated just above the upper surface of the stratum of wind and 
aqueous vapour which supplies the S.W. monsoon to Western India, and therefore 
has comparatively a small supply of rain from this source. But it is not situated 
(although on the western coast) above the stratum of wind and aqueous vapour 
which supplies the Coromandel coast during the N.E. monsoon, as it has the same 
prevailing winds between the N. and E. points in the same months as at Madras, 
from October to February, when the N.E. ceases at Madras, but continues at Doda- 
betta until late in May. The 'prevalence of winds from points between N. and W. 
in the months of July, August and September, is peculiar to Dodabetta; neither 
Mahabuleshwur, at 4500 feet, nor Madras, Bombay nor Calcutta has similar indica- 
tions. However, as this so-denominated N.W. wind very frequently blows from only 
one or two points to the northward of west, the wind may belong to the monsoon of 
Western India, local physical circumstances having given it a slant. On the whole 
it may be said, that the peninsula of India, in its length and breadth and the atmo- 
sphere over it to the height of nearly 9000 feet, is subject to periodic winds ; but at 
widely separated places, varying a point or two from the apparent normal winds, and 
commencing or terminating a week or two earlier or later. May, June, July, August 
and September have prevailing winds from points between W. and S. ; October 
November, December, January and February have prevailing winds from points be- 
tween N. and E. ; February, March and April, at Madras, have a prevailing wind 
from points between S. and E., as opposed to Bombay, where the wind in those 
months prevails from points between N. and W., with rare instances of a S.E. wind ; 
while at Calcutta the N.W. wind terminates in February ; but a series for years of 
hourly observations, like those at Madras, are necessary to give that confidence which 
cannot be given to deductions from the meteorological data at present available from 
India. 
It is usually understood that very high winds materially depress the barometer, but 
the records at Dodabetta do not support this view. On the 17th and 18th of April 
1847, the wind blew with a mean pressure of 21 lbs., and 14 lbs. respectively upon the 
square foot; but the barometer only fell from 21'955 on the 16th to 21'917 on the 
17th, and rose to 21 '984 on the 18th ; and there was a maximum pressure from the 
wind on the 17th at one time of 35 lbs. 26th of May, maximum wind 28'5 lbs., baro- 
meter not affected more than 0 010 inch ; 12th of June 30 lbs., 26th of July 32 lbs., 
10th of September 35 lbs., and 14th of October 22 lbs. ; but these pressures of the 
wind had little or no effect upon the barometer. 
