872 IP. Lihbey, Jr. — Gases in Kilauea. 



these fountains quite closely in two instances, we could form 

 very good estimates of their size and the amount of lava in- 

 volved in such an outbreak. The surface for a considerable 

 area near one of these centers (for they seemed very constant 

 in position) would sway up and down, at first gently and then 

 more violently, particularly in the neighborhood of the inter- 

 section of the fissures which were constantly forming in the 

 crust of slag, then there would be two or three small explosions 

 accompanied by a bumping sound, as though gas had escaped 

 from the mouth of an uncorked bottle ; after this there w r ould 

 he a great rush of lava into the air, to be repeated several 

 times; when the surface would become quiet once more. 



At the edge of this cauldron, 1000 feet in diameter, I 

 placed myself after dark and for a couple of hours in each 

 instance had opportunities nearly every five minutes to ob- 

 serve the flames which almost invariably accompanied these 

 explosions. At times they would make their appearance along 

 the fissures, in the crust when they would be very short lived: 

 and I failed to get much information concerning them, aside 

 from the fact that they seemed the exact counterpart of the 

 hydrogen flame from a Bunsen burner. 



It was the study of the flames which accompanied the larger 

 outbreaks which appeared to promise most return. Getting as 

 near as possible to one of the giant fountains (probably 50 

 yards distant) I watched for the premonitory symptoms of an 

 outbreak and got all ready for it, and seldom failed to catch 

 indications of the presence of gases. 



For a good part of the time there was a continuous spec- 

 trum while the spectroscope was directed across the lava. The 

 first thing that impressed me was the sudden appearance and 

 disappearance of broad hands of bright light, showing conclu- 

 sively the presence of gas burning under high pressure. The 

 location of these bands became the next problem and then I 

 longed for a better instrument with a micrometer eyepiece. 

 The first which appeared with constancy was a band in the 

 green, indicating the presence of carbonic oxide in all proba- 

 bility. Then I found on other occasions bands of lighter 

 intensity in the red and blue, and the red and purple portions 

 of the spectrum, thus apparently marking the presence of the 

 hydro-carbons. There were also occasionally noticed upon a 

 full spectrum a large series of dark lines in the yellow and 

 orange, sometimes completely blotting these colors out of the 

 spectrum altogether. These need more careful study before 

 assigning them definitely to any substance or substances. 



It is hoped that the above may serve to indicate a point 

 upon which valuable work can be done and that some one will 

 take advantage of it and follow the subject up. 



