4$ 4 Vapour confifts of hollow Peficks. [Book VI L 



rebounding like a ball from a wall, and others fixing 

 themfelves to it. Small drops of water might alfo be 

 feen to fettle on the tortoife fhell, but they were eafily 

 diftinguiftied from the hollow fpherules, by their 

 tranfparency, their gravity, and their pace. Further, 

 clouds do not form a rainbow ; it is produced by folid 

 drops j when a cloud is not in an a<5lual (late of refo- 

 lution, it does not change the form of the ftars which 

 are feen through it j for infinitely thin menifcufifes do 

 not fenfzbly change the courfe of the rays of-light; but 

 as foon as the cloud begins to refolve itfelf into folid 

 drops, or even without clouds, when folid drops begin 

 to be formed in the air, the ftars feen through them 

 are all defined, furrounded with a diffufe light, circles, 

 and halos ; hence, thefe meteors are the forerunners 

 of rain, for rain is nothing more than thefe drops aug- 

 mented or united. When the veficular vapours are 

 condenfed by cold, the water which formed the bub- 

 ble cryftallizes, fometimes into hoar froft, fometimes 

 into fnow.j when it does not freeze, they unite in 

 dew, or defcend in rain. Many other curious pro- 

 perties concerning the veficular and concrete vapours* 

 are related in M. de Sauflure's EfTai fur 1' Hygro- 

 metric *. 



The precipitation of water from the atmofphere jn 

 rain has given rife to as much fpeculation as evapora- 

 tion j and it muft be confefled, that the caufe of neither 

 has yet been very clearly afcertained. By fome it is 

 fuppofed, that the capacity of the air for fufpending 

 aqueous vapour is in proportion to its heat, and there- 

 fore that any circumftance which cools the atmofphere 

 will produce rain. If, therefore, according to this 

 theory, a ftratum of cold air meets with a warmer ftra- 



* See Adams's Leisures on Natural Philofophy. 



turn, 



