298 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1913. 



which, when combined with water, readily changes to sulphuric acid 

 and would tend to drive the chlorine out of any combinations which 

 it might form. It constitutes, nevertheless, a plain indication that 

 the amount of chlorine actually evolved is insignificant. 



DISCUSSION OF RESULTS. 

 GENEEAL CONCLUSIONS. 



In so far, then, as this reconnaissance yields final results, it shows 

 that the gases evolved from the hot lava at the Halcmaumau Crater 

 are N„ H^O, CO^, CO, SO,, free H, and free S, with CI, F, and per- 

 haps NH.5, in comparatively insignificant quantity. No argon or 

 other rare gases and no hydrocarbons were found. 



THE EFFECT OF THE REACTIONS BETWEEN THE GASES. 



The first plain conclusion which follows from the discovery of this 

 particular group of gases associated together at a temperature of 

 1,000° or more is that they can not possibly be in equilibrium there, 

 and that chemical action between them is still going on. '\\liatever 

 may have been the previous opportunities for chemical readjustment 

 among the gases as they rose in solution with the magma and were 

 gradually set free with the diminishing pressure, they are still in 

 process of active reaction when discharged into the air. Free sul- 

 phur, for example, could not have remained in permanently stable 

 association with COo ; neither could free hydrogen be found in stable 

 association with CO., and SO, at 1,000^ 



THE EFFECT OF THE EXPANSION OF THE GASES. 



Moreover, as the pressure continued to diminish during the progress 

 of the upward movement, the quantity of gas released from solution, 

 and therefore free to enter into new relations, must have been con- 

 stantly and rapidly increasing up to the moment of discharge into 

 the air. 



Two consequences follow from the continuation of this operation, 

 which are thermally opposite in sense. First is the rapid expansion 

 of the gases with the release of pressure, which is a cooling phenome- 

 non, and which, if the expansion takes place suddenly from a high 

 pressure into the air, might finally be extremely rapid. 



AN EXPLANATION OF THE FORMATION OF AA LAVA. 



Parenthetically, it may be noted in passing that such rapid expan- 

 sion and consequent cooling when occurring suddenly at the surface 

 may very well be the suiEcient cause of the Aa lava formations. 

 Great blocks appear to have cooled in this way so rapidly that no op- 

 portunity was given for the suddenly projected and rapidly ex- 

 panding lava outbursts to " heal " and resume liquid flow. The pro- 



