260 ANNUAL EEPORT SMITHSONIAN INSTITUTION, 1925 



plainly indicates that (heir relation can not be a constant or a 

 stable one. 



Our knowledge of the reactions which must be taking place 

 between these gases is adequate to establisli the fact that heat is 

 being contributed to the laA^a mass in conse(juence of these reactions 

 going on within it. We have, tlierefore, certainly hit upon one of 

 the sources of energy which serves to maintain the particular type 

 of volcanism which is familiar to all at Kilauea. Very probably 

 we have done somcAvhat more than this. If we have established the 

 fact that continual reaction is going on during the passage of the 

 gases through the liquid lava, whereby heat is contributed to the 

 lava, it is a probal)le consequence that the temperature reaches its 

 maximum at or near the surface. The consequences of this decision 

 ai'e s(»mowhat fai"-reacliing if one is minded to ])ursue them to their 

 logical conclusion. For example, it appears to follow that the 

 temperature in the basin will be lower as we go down into the lava, 

 rather than higher, whence it is reasonal)le to suj^pose that the en- 

 tire phenomenon is local and superficial rather than deep-seated. 



Dr. T. A. Jaggar, director of the Hawaiian Research Ivaboratory, 

 has already confirmed this deduction regarding (lie teuipera(ure 

 gradient by actual u)easuremen(s of the lava ((Muperature.'" lie used 

 for this purpose a stout iron pipe within which groups of Seger 

 cones were fastened at regular intervals and thrust it into (he lava 

 as far as he could (20 feet). Temperatures were found to be some- 

 Avhat variable but lower by 100" or more than at the surface. The 

 existence and behavior of the islands and the heavy semirigid masses 

 below the surface upon which they rest, also indicate a lower temper- 

 ature below, for the circulation of gases in these masses will be con- 

 siderably harhpered. Very likely they are themselves losing gas 

 more or less continually and thereby losing their fluidity. In any 

 event such a temperature distribution does not aflf'ord support for 

 the old hypothesis that volcanoes are ""safety valves" to insure the 

 stability of a molten interior, indeed they all'ord not the slightest 

 indication of the existence of a generally molten interior. 



In a book which appeared quite recently Mr. Harold Jeffreys, 

 of Cambridge University, offered a careful analysis of existing evi- 

 dence of the manner of formation of the earth ® and the probable 

 state of its interior. His reasoning has seemed to me to include more 

 of modern physics and chemistry than the reasoning hitherto applied 

 to the constitution of the earth. He figures that the gaseous mass, 

 presumably split off from the sun by tidal rupture due to another 



» T. A. Jagger, Volcanologic Investigations at Klhuica, Am. J. Sci. (4), 44, p. 208 et 

 spq.. 1017. 



» Harold Jeffreys. The Earth, its Origin, History and I'hysical Constitution. Cam- 

 bridge, 1924. 



