IV. Libbey, Jr. — Gases in Kilauea. 371 



their variations with temperature throughout an interval of 

 nearly 2000° centigrade remain strictly linear functions* of 

 each other, both may be reasonably looked upon as evidencing 

 a clearly pronounced thermoelectric law. However, in the 

 absence of a set of tables similar to Kulik'sf for the catenary, 

 the labor of computation becomes so great as to be almost pro- 

 hibitory. In the mean time I have therefore contented myself 

 by suitably stretching a chain across the chart of observed re- 

 sults (electromotive force in terms of temperature) for iridio- 

 platinum, and noting the essential similarity of the two curves 

 in question. 



The Smithsonian Institution, Washington, D. C. 



Art. XLI. — Gases in Kilauea; by William Libbey, Jr., 

 Professor of Physical Geography, Princeton, ]N~. J. 



In 1865 Mr. W. T. Brigham called attention to the exist- 

 ence of certain bluish-green flames which broke from the 

 crust of Halemaumau during a disturbance of its slag-like 

 surface. 



In 1887 Mr. Emerson of the Geological Survey of the 

 Islands and several others identified these same flames. They 

 are referred to in Prof. Dana's book upon " The characteristics 

 of Volcanoes" (p. 119) as being pale in color and of a slightly 

 greenish color rather than bluish. This corresponds with my 

 own observations. In order to test this matter I took with me 

 upon my recent visit to the volcano (Sept. 14th to 25th, 1893) 

 a pocket spectroscope. I did not imagine that I should have 

 such a splendid opportunity or I should have taken a better 

 instrument ; however the little spectroscope did good service 

 and the observations may possibly lead to something better 

 later on. 



I spent the greater part of three evenings down on the edge 

 of the boiling cauldron in Halemaumau and the observations I 

 made were repeated many times so that I have every confi- 

 dence in them. As to the conditions I may say that they were 

 exceptionally favorable. The surface of the molten lava was 

 so high during the whole of our visit that it could be reached 

 with an ordinary walking stick. The volcano seemed to be in 

 a very active state as several overflows occurred, and the ebul- 

 lition from what appeared to be three centers of weakness or 

 disturbance were very remarkable. The lava was thrown into 

 the air for a distance of 40 to 50 feet and as we could approach 



*Barus: Phil. Mag., xxxiv, p. 376, 1892. 



f Kulik : Theorie und Tafeln der Kettenlinie, Prag, 1832. 



Am. Jour. Sci.— Third Series, Vol. XLVII, No. 281.— May, 1894. 

 25 



