WATEE AXD VOLCANIC ACTIVITY DAY AND SHEPHERD. 283 



cent of FeO, the momentary contact of the gases with the oxidized 

 surface of the iron was not accounted a serious disturbing factor. 



The other end of the tube system was connected to a piston pump 

 about 4 inches in diameter, with a displacement of about 2^ liters per 

 stroke to insure a rapid passage of the gases through the tube system. 



The gases entered the pipe line at a temperature of about 1,000°. 

 Their path was through the 12 inches of iron pipe, about 20 feet of 

 glass tubing (pure rubber joints at 4-foot intervals), then through 

 20 collecting tubes and out througli the pump at the back. Tiie 

 pumping was kept up for 15 minutes in order to make sure that the 

 air originally contained in the pipe line and connecting tubes was dis- 

 placed by the gases from the volcano, after which the pump and pipe 

 line were sealed off with pinchcocks and the crates raised to the rim. 

 In this pipe line Avater began condensing with the first stroke of the 

 piunp, and at the end of 15 minutes about 300 cubic centimeters had 

 accumulated in the collecting tubes. It was clouded with free sul- 

 phur, partly from the original emanation and partly from the action 

 of the iron tube on the sulphur dioxide contained in the emanation. 



In arranging this experiment Brun's conclusions were known to us, 

 and accordingly we had provided ourseh'es with apparatus for col- 

 lecting fixed gases onl3^ We were wholly unprepared for any which 

 might condense in passing through the collecting tubes. 'What w^e 

 obtained, therefore, was a quantity of the fixed gases, which may be 

 assumed to be approximately in the proper quantitative relation one 

 to another, and water, the latter in considerable excess from the fact 

 that it was not pumped through the tubes with the fixed gases, but 

 condensed and remained behind, chiefly in the first three or four 

 tubes. There is, therefore, no way to estimate from the results of 

 this experiment the proportion of water to the total quantity of 

 volatile matter discharged from the lava. Perhaps this should be 

 regarded as a fortunate mischance notwithstanding, for we were 

 thereby enabled to gather a quantity of water sufficient to establish 

 its existence among the volatile ingredients exhaled by the volcano 

 beyond the criticism of the most skeptical. Furthermore, the con- 

 densing water by its accumulation in the first tubes served as a kind 

 of wash bottle for the collection of any soluble material contained in 

 the gaseous emanation. 



The next day we began preparations to meet the emergency thus 

 thrust on us by building in the laboratory of the Hawaiian Volcano 

 Research Association^ an extemporized mercury pump of the dis- 



1 The Hawaiian Volcano Research Association is organized under the general super- 

 vision of the Bishop Museum of Honolulu, and is in charge of Prof. T. A. Jaggar, of the 

 Massachusetts Institute of Technology, to whom our most cordial thanks are offered for 

 many courtesies extended to us throughout our work at Kilauea. Mr. F. B. Dodge, an 

 assistant In the association laboratory, accompanied us in the first descent into the 

 crater, and Dr. H. O. Wood, wlio is in charge of the selsmologic work of the station, on 

 the second, both rendering Invaluable assistance in carrying out this difficult and some- 

 what hazardous task. 



