478 DOVE ON THE PERIODICAL VARIATIONS 
the contrary, that the air becomes cooler when the barometer is 
rising. But no notice is taken of the fact that in the daily period 
the atmospheric pressure attains two maxima and two minima, 
while the temperature, on the contrary, uniformly rises and falls 
within twenty-four hours; and that something similar is ob- 
served during the annual period, at least in Europe, where more- 
over the pressure during the decided summer months is much 
higher than in spring. 
These contradictions appear to me so striking, that I at- 
tempted to remove them in two treatises published in Poggen- 
dorff’s Annalen ten years ago. One of them, ‘On the Physical 
Causes of the Daily Changes of the Barometer,’ is printed in 
vol, xxii. pp. 219 and 493; the other, ‘On the Distribution of 
Atmospheric Pressure during the Annual Period, and on Baro- 
metric Levelling of the Plains,’ in vol. xxiv. p. 205. My men- 
tioning here this circumstance again may be justified by the 
view then advanced having received a remarkably evident con- 
firmation in the North Asiatic observations. 
This view is, that although a separation of the vapour-atmo- 
sphere seems not to be essential in the variations of the atmo- 
spheric pressure dependent on the direction of the wind, it be- 
comes indispensable in the explanation of the periodical varia- 
tions. 
The pressure of the dry air, the elasticity of the vapours, and 
the temperature, attain their extremes nearly at the same points 
of the wind-rose. Their distribution within the wind-rose is 
likewise so analogous, that it may be expressed by the same 
function of the direction of the wind. Leaving out of question 
the fact that the vapour-atmosphere alone will not produce a wind 
different to that originating at the same time in the dry air, that 
therefore moist atmosphere, when moved horizontally, will be 
moved as a whole, it is proved by empirical data, from calcu- 
lated barometrical and atmospherical wind-roses, that the total 
pressure of the atmosphere in its changes, dependent on the 
direction of the wind, follows the same principles fixed by the 
law of rotation as the dry air. A separation therefore of the varia- 
tions of the vapour-atmosphere will only be necessary with re- 
spect to the precipitates which originate from the reciprocal 
displacement of winds of unequal temperature. But setting 
aside these precipitates, the total pressure of the atmosphere in 
its so-called irregular variations, may be regarded as a direct 
