ANALYSES OF LAVA 587 



with our present limited knowledge of gas relations at these temperatures, 

 can be only partly inferred. It is, therefore, a matter of the first im- 

 portance to collect the gases directly from the liquid lava or the explosive 

 vents before' contact with the air has given opportunity for these altera- 

 tions to occur. It may very well be that the physical difficulties attending 

 the collection of volcanic exhalations, particularly from volcanoes of the 

 explosive type, will often make it impossible to obtain unaltered mag- 

 matic gases for laboratory study, in which case burned gases, or even 

 very dilute mixtures of these with air, may prove to be the only products 

 available for study. In this event the student must perforce bow to the 

 necessities of the case. 



Something of the same cautious attitude requires to be maintained 

 toward the study of the flame spectra of burning volcanic gases. The 

 pocket spectroscope is primarily an instrument of preliminary reconnais- 

 sance in the field and is sometimes of value, but the pale blue flames of 

 sulphur and hydrogen are extremely difficult to analyze with the pocket 

 spectroscope, and can not be distinguished at all against a bright back- 

 ground of solid or liquid lava. For this reason a more elaborate spectro- 

 scopic equipment would not help the matter. Moreover, the gases are 

 much altered or are in process of active alteration before any opportunity 

 for identification is offered, and no estimate of the relative quantity of 

 the various participating gases is possible by this means. Inferences from 

 the chemical study of gases which have been burned by contact with the 

 air while still hot, and inferences from the spectroscopic study of the 

 gases while burning, therefore suffer alike from limitations of principle 

 and should be resorted to only when the difficulty of collecting unaltered 

 gas is insuperable. 



These reasons may serve to show why this somewhat elaborate effort 

 was made to collect unaltered gases for laboratory study and why we are 

 inclined to give greater weight to the results obtained from the study of 

 such gases than to many of the earlier studies 12 of volcanic emanations, 

 in which the gases had become altered through contact with the air or 

 otherwise. 



The domes from which these gases were collected were built up by the 

 lava itself on the floor of the crater (Halemaumau) and wore both chemi- 

 cally and physically ideal gas collectors, being lined with fresh splashes 

 of liquid lava of the same temperature and chemical composition as that 

 from which the gas had just emerged. They formed at the level fcf the 

 lava lake and, as could be plainly seen after the collapse of the domes, 



u e. g., Wm. Libbey: Am. Jour. Sci. (3), vol. 47, 1894, p. 371. 



