(| xvii) 
APP TON DX, } Nowe I I 
RESULTS of METEOROLOGICAL INQUIRIES, made at Manpras, by Jonn 
Gopincuam, Esq., F.R.S., M.R.A.S., M.A.S. Calcutta, &c. &c.—(Communicated 
16th of January 1830.) 
I wave the honour to lay before the Society some results of meteorological inquiries 
made at Madras, between the years 1796 and 1826. The observatory, where the baro- 
meter and the other instruments used were placed, was two miles and three-quarters 
distant from the sea, and about twenty feet above its level. It was known that the 
barometer in settled weather in India had a continued regular rise and fall in the 
twenty-four hours, and that the times of its greatest and least heights were nearly the 
same on different days; but I did not recollect that any observations upon a very 
extended scale had been made with the view of exhibiting such changes: in order 
therefore to obtain the quantities and times of these diurnal variations, I had the 
height of the barometer taken every hour during three days of each month in the year 
1823. ‘These observations are given in detail in the table beginning at page 374 of the 
volume of “ Madras Observatory Papers.” 
The rise and fall of the barometer were very regular in the twenty-four hours during 
the year; at about ten o’clock in the forenoon it was at its greatest height; it then 
began to fall, and continued falling until a little after five in the afternoon, when it 
was lowest; it then commenced rising, and before eleven at night attained its highest 
point; and was again lowest at four o’clock in the morning. Any change of wind or 
weather, of course, will break this chain of regularity, more or less according to the 
suddenness or violence of the change; but as there were no great or sudden changes at 
Madras in the twenty-four hours during the year 1823, the regularity is not much 
broken, scarcely at all indeed in the forenoon, but more at other times. 
The following abstract, made from the hourly diary, shews the times of the greatest 
and least heights of the barometer in twenty-four hours: I have divided this into two 
equal intervals of six months, in order to shew what difference there would be between 
the means of each interval. 
4K 
