90 SCIENCE PROGRESS 



of hydrogen, nitrogen and carbon dioxide practically making up 

 the rest. Specimens of the gas taken in later stages show a 

 progressive increase in oxygen and in carbon dioxide. The fact 

 that chlorides of magnesium and iron are deposited on cinders 

 around the crater again proves, according to Brun, that the 

 exhalations of volcanoes are anhydrous. The " steam " of 

 volcanoes consists of volatile chlorides, mostly ammonium 

 chloride. If the "steam" had been water-vapour, it would 

 dissolve in air and soon disappear. The white cloud, on the 

 contrary, remains suspended during long periods over the 

 volcano and the wind may carry it many miles to leeward. 

 The most positive evidence Brun advances is his measurement 

 of the humidity of the cloud given off from the pit of Kilauea in 

 eruption relatively to the humidity of the air outside. In a long 

 series of observations, he found that there was less moisture in 

 the cloud than outside it, and consequently he inferred that 

 there was no water-vapour in the exhalation. On the other 

 hand, the cloud of the fumaroles on the north of the pit, in 

 action at the same time as the volcano, contained much water- 

 vapour. Gautier found from 62 to 77 per cent, of water-vapour in 

 the fumaroles of Vesuvius after the eruption of 1906 ; but in view 

 of Brun's work in Hawaii, one is not justified in maintaining that 

 the gases of the central chimney must equally be hydrous. 

 Moissan's(9) analyses of the gases of the Mont Pelee fumaroles, 

 interesting from the fact that considerable quantities of argon 

 were discovered in them, show large amounts of oxygen and 

 water-vapour. It will be remembered, also, that in the 

 beginning water was pumped up into L'£tang Sec and caused 

 the mud-rush which overwhelmed the Usine Guerin. That is 

 to say, when a volcano begins to work after a period of 

 quiescence, the volcanic gases drive before them the water 

 contained in the crevices and pores of the rocks ; then, when the 

 eruption ceases, the same water from the surface seeks to 

 penetrate again into the cracks which it previously occupied. 

 As the pressure of the volcanic vapours grows less and less, 

 the surface water advances more and more into the heated area, 

 till, coming at last into the neighbourhood of the cooling molten 

 rock, it is driven forth in the form of aqueous vapour mixed 

 more or less with volcanic products. 



The elements of water, it is true, are found in the volcanic 

 exhalations, but combined with carbon, chlorine, or nitrogen. 



