3. J COSMOS. 



take place Lit more complicated periods than those found by 

 Saussure and myself), and its variations in the different 

 seasons of the year, at different distances from the equator, 

 and in the diflerent relations of continental or oceanic sur 

 face. 



The electric equilibrium is less frequently disturbed where 

 the aenal ocean rests on a liquid base than where it impends 

 over the land ; and it is very striking to observe how, in ex- 

 tensive seas, small insular groups affect the condition of the 

 atmosphere, and occasion the formation of storms. In fogs, 

 and in the commencement of falls of snow, I have seen, in a 

 long series of observations, the previously permanent positive 

 electricity rapidly pass into the negative condition, both on 

 the plains of the colder zones, and in the Paramos of the Cor- 

 dilleras, at elevations varying from 11,000 to 15,000 feet. 

 The alternate transition was precisely similar to that indica- 

 ted by the electrometer shortly before and during a storm.* 

 When the vesicles of vapor have become condensed into clouds, 

 having definite outlines, the electric tension of the external 

 surface will be increased in proportion to the amount of elec- 

 tricity which passes over to it from the separate vesicles of 

 vapor. t Slate-gray clouds are charged, according to Peltier's 

 experiments at Paris, with negative, and white, red, and or- 

 ange-colored clouds ^viih positive electricity. Thunder clouds 

 not only envelop the highest summits of the chain of the An- 

 des (I have myself seen the electric effect of lightning on one 

 of the rocky pinnacles which project upward of 15,000 feet 

 above the crater of the volcano of Toluca), but they have also 

 been observed at a vertical height of 26,650 feet over the low 



*■ Humboldt, R6lalion Historiqtte, t. iii., p. 318. I here only refer 

 to those of my experiments in which the three-foot metallic conductor 

 of Saussure's electrometer was neither moved upward nor downward, 

 nor, according to Volta's proposal, armed with burning sponge. Those 

 of my readers who are well acquainted with the qucEsilones vexata of 

 atmospheric electricity will understand the grounds for this limitation. 

 Respecting the formation of storms in the tropics, see my Ril. Hist., t, 

 ii., p. 45 and 202-209. 



t Gay-Lussac, in the Annales de Chimie et de Physique, t. viii., p. 1G7. 

 In consequence of the discordant views of Lame, Becquerel, and Pel- 

 tier, it is difficult to come to a conclusion regarding the cause of the 

 specific distribution of electricity in clouds, some of which have a pos- 

 itive, and others a negative tension. The negative electricity of tha 

 air, which near high water-falls is caused by a disintegration of the 

 drops of water — a fact originally noticed by Tralles, and confirmed by 

 myself in various latitudes — is veiy remarkable,' and is sufficiently iu- 

 lanse to produce an appreciable effect on ^ dolicate electrometer at a 

 distance of 300 or 400 feet. 



