Variations of the Barometer under the Tropics. * 829 
In the equatorial zone, there may be admitted, for these four 
periods, 21 J, 16, 10 J, 16; and, in the temperate zone, 20 J, 3J, 
9J, IT ; these numbers expressing the hours counted from 
noon. 
2. In the temperate zone, the periods of the maximum of the 
morning, and of the minimum of the evening are nearer, by 1 
or 2 hours, to the passage of the sun through the meridian in 
winter than in summer. Observations are wanting regarding 
the minimum of the night. M. de Humboldt recommends them 
to be made. 
8. In the torrid zone, the hours of the maxima and minima 
are the same at the level of the sea, and on plains of from 1800 
to 1400 toises in height. This is asserted not to.be the case 
in some parts of the temperate zone. On Mount St Bernard, 
for example, the barometer falls at the same hours at which it 
is rising at Geneva. 
4. Near the maxima and minima , the barometer is almost 
stationary during a more or less considerable period ; this period 
varies from 15 7 to 2 hours. 
5. Between the equator and the parallels of 15° N. and S., 
the strongest winds, tempests, earthquakes, and the quickest 
variations of temperature and humidity, do not interrupt or mo- 
dify the periodicity of the variations. In India, on the contra- 
ry, the rainy season entirely disguises the type of the horary 
variations in the interior of the Continent, on the coasts, and in 
the straits, although in the open sea they remain unaltered. 
6. Between the tropics, a day and a night suffice for know- 
ing the extreme points, and the duration of the variations. In 
the latitudes of 44° and 48°, they are very distinctly manifested 
in means of from 15 to 20 days. 
7. The extent of the diurnal variations, at the same hours, 
and in different months, is not the same. This extent also de- 
creases in proportion as the latitude augments. — (See the an- 
nexed Table). Lastly, The maximum of the morning is a little 
higher than the maximum of the evening. The height of the 
place does not influence these results. 
8. The barometrical means of the months differ among' them- 
O 
mm mm 
selves from 1.2 to 1.5, between the tropics ; and from 7 to 
8 millim. near the tropics, nearly as in the temperate zone. The 
