166 PRESSURE OF THE ATMOSPHERE. 



tions, in which the mercury in the barometer is always 

 higher at 9 A.M. and 9 P.M., than at 3 P.M. and«3 A.M. 

 These motions are much more distinct in India, and in 

 tropical regions in general, than in temperate region?. 

 From the observations of Humboldt we learn, that in tropi- 

 cal America these atmospheric tides are independent of 

 changes in the weather and seasons. Thus, it the mercury 

 is falling from nine in the morning until three or four in the 

 afternoon, or if it be rising from four in the afternoon until 

 nine or eleven at night, a storm, an earthquake, or violent 

 tempest of wind does not aflect or alter its course. It appears 

 to be aifected only by true time, or the position of the sun. 

 In the tropical regions, he adds, the moment when the mer- 

 cury begins to fall is so marked, that the barometer indi- 

 cates true tune within a quarter of an hour. Whatever 

 truth may be in the latter observation of Humboldt, there 

 can be no doubt as to the motions themselves. The only 

 obser^•ations made io Ij^dia with which we are acquainted, 

 are the very interesting one's of Dr. Russel at Burhanpoor,in 

 24° north lat., and of Mr. Prinscp at Benares, in 25h° north 

 lat., continued for three 3'ears, and which hannonize in 

 general with those made in other tropical countries. At 

 present we are not in possession of a range of observations 

 sufficiently extensive to enable us to explain these horary 

 motions. The speculations of Humboldt, Leslie, and others 

 on this subject are unsatisfactory. It is indeed evident 

 that these motions are connected, not only with the atmo- 

 spheric temperature, but also with its associated moisture. 

 Until, however, we have a series of hourly-connected ob- 

 servations of the barometer and hygrometer, we cannot at- 

 tempt any explanation likely to be plausible. 



The connexion of the mean monthly heights of the ba- 

 rometer with the south-west and north-east monsoons is 

 also a subject of considerable interest, and is well illus- 

 trated in barometrical observations made at Seringapatam, 

 Bangalore, Calcutta, Benares, Catmandoo, and Madras, for 

 the particulars of .which we refer to the original tables and 

 observations of the obsen^ers. 



2. Composition of the Atmosphere. — It would appear 

 firom experiments made in ditierent countries and at different 

 heights, that the proportions of oxygen and azote, the. prin- 

 cipal constituents of the atmo?phcrc,'do not vary. Carbrmjc 



