272 ANNUAL, REPORT SMITHSONIAN INSTITUTION, 1920. 



The igneous magma may be compared, as it usually is, to a com- 

 plex solution of salts in water. This idea, which was first suggested 

 by Bunsen in. 1861, is of great importance, and has been very 

 fruitful in our study of the origin, formation, and characters of 

 igneous rocks. 



Among other things, it may be mentioned here that the magma 

 contains various gases in solution, much as air is present in solution 

 in spring water, or, rather more appropriately, as carbon dioxide 

 is present in the waters of many mineral springs, so that it escapes 

 on relief of pressure. 



Of these gases by far the most important, and generally the most 

 abundant, is water vapor. This forms the major part of the clouds 

 that are given off during volcanic eruptions, and, with other gases, 

 produces the spongelike structure of pumice and the cavities of 

 vesicular lavas through expansion, caused by relief of pressure on 

 reaching the surface. In some glassy lavas water is present to the 

 extent of several per cent, the magma having solidified so rapidly 

 as not to permit of its escape, and inclusions of visible water and 

 liquid carbon dioxide are present in the crystals of many granites 

 and other rocks. The presence of water in volcanic magmas has 

 been doubted by Brim and others following him, but its existence 

 in lavas, especially those of Kilauea, has been shown conclusively by 

 the researches of Day and Shepherd, 3 is shown by practically every 

 rock analysis, and in other ways, so that the existence of water in 

 magmas may be regarded as one of the established truths of the 

 chemistry of igneous rocks. 



Besides water, other gases are often present in volcanic ex- 

 halations, such as carbon dioxide, carbon monoxide, hydrogen chlo- 

 ride, sulphur trioxide and dioxide, hydrogen sulphide, hydrogen 

 fluoride, ammonia, methane and possibly other hydrocarbons, sul- 

 phur vapor, hydrogen, nitrogen, oxygen, argon, helium, and other 

 rare gases. The study of these and the bearing of their interreactions 

 on the maintenance, and possibly the partial production, of volcanic 

 heat, is an interesting subject. 



The presence of these gases in the magma lowers its solidifica- 

 tion point, so that a lava, on coming to the surface, may be, and 

 usually is, liquid at a temperature considerably below the fusing 

 point of the solid rock formed from it, during which solidification 

 much, if not most, of the dissolved gas is lost. Either because of 

 this, or because of the lessened viscosity, or in some other way that 

 is not yet well understood, the gases contained in the magma seem 

 to promote the crystallization of minerals, so that they are often 



"Day and Shepherd, Bull. Geol. Soc. Amer., sxiv, 573, 1913. 



