Natural Hutm-y. 165 



tube closed by a prepared cork, which had a small hole through 

 it, closed by fat lute. The tube was then retained by weights 

 at a depth of between two and three feet beneath the water, and / 

 then, by means of a steel wire fixed to a long rod, the lute was 

 perforated, and water admitted. The powder immediately in- 

 flamed, and a weight of above 2lb. was thrown out of the vessel 

 containing the water. — Annates de Chim. xxi. 197. 



III. Natural History. 



1 . On the Ascent of Clouds in the Atmosphere, by M. Fresnel. — 

 Among the causes which most effectually contribute to the ascent 

 of clouds in the atmosphere, there is one to which little attention 

 has been given, but without which it appears impossible to give 

 a satisfactory explanation of the phenomenon. It is indepen- 

 dent of the constitutioh of the globules of water, or vesicular 

 vapour composing the cloud ; and is equally applicable to one 

 formed of an assemblage of delicate crystals, such as may 

 actually exist in the high regions of the atmosphere. 



Air, as well as other colourless gases, permits the solar rays 

 to pass without being heated by them ; and to heat them, the 

 contact of a solid or liquid body, heated by the same ray, is re- 

 quired. Consider, then, the case of a cloud formed of minute 

 globules of water, or very fine crystals of snow : from the ex- 

 treme division of the water, a very multiplied contact with the 

 air is obtained, and the water being susceptible of an increase 

 of temperature from the solar and terrestrial rays, the air 

 within the cloud, and near to its surface, will become more 

 dilated than the neighbouring air, and consequently lighter. 

 It equally results from the hypothesis, on the extreme division 

 of the matter of the cloud, that the particles which compose it 

 may be very near each other, so as to leave but small intervals, 

 and nevertheless be very much smaller than the ifttervals ; so 

 that the whole weight of the water in the cloud should be but a 

 a small fraction of the weight of the air containing it, and so 

 small, that the difference between the density of the air in 

 the cloud and the neighbouring air should more than com- 

 pensate it. When the weight of the water and air contain- 

 ing it is less than that of an equal bulk of the surrounding air, 

 it will ascend until it arrives at a region where these two weights 

 are equal; and this height will depend on the fineness of the 

 particles of the cloud, and the intervals which separate them. 

 M 2 



