290 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1913. 



reactions to have occurred as described, it is then a matter of straight- 

 forward computation to show that if the known weight of iron which 

 was dissolved away by the gases, both those which entered the pipe 

 and those which merely played on its outside wall,^ had reacted in 

 this manner; and supposing further that all the products of the 

 reaction, botli outside and inside, entered the collecting tubes (which 

 is obviously impossible), it would have involved pumping through 

 the system some 225 liters of pure hydrogen as an equivalent for the 

 iron consumed, and this still falls short of the quantity required to 

 account for all the water collected by more than 40 per cent. More- 

 over, if the attempt is to be made to account for all the water col- 

 lected in our tubes through reactions requiring free hydrogen, it 

 is altogether inconceivable that any such quantity of uncombined 

 hydrogen is available in the emanation from the volcano. For if we 

 were to assume that as much as 1,000 liters of volcanic gas (which 

 is a very liberal estimate) passed into the collecting train in the 15 

 minutes during which the pumping was continued, such a quantity 

 of free hydrogen (375 liters) would be equivalent to 40 per cent of 

 the total composition, a quantity sufficient to form an explosive mix- 

 ture on contact with air of such extreme violence as to change the 

 entire character of the volcanic activity at Halemaumau. It is a 

 fact of general observation that the bubbles of gas which come up 

 through the liquid lava, even when they reach the enormous size 

 of 30 feet in diameter, give no explosion whatsoever. 



We may therefore fairly conclude, both from the character of the 

 reactions in which the iron might have a part and from the quantity 

 of water collected, that the presence of the iron tube has no consider- 

 able significance in relation either to the character or to the amount 

 of volatile material collected. 



THE REACTION BETWEEN H; AND SO, OR COo. 



To this reaction assumed to be going on between H, and FeO may 

 be added another and much more important one in which the iron 

 has no part. The free hydrogen set free by the volcano reacts with 

 sulphur dioxide at 1,000° to give water and free sulphur directly. 

 It will also be recalled that carbon dioxide and hydrogen undergo 

 similar reaction at this temperature. This is the familiar water- 

 gas reaction 

 ^ H^+CO, tz; CO-f H,0 



which has been thoroughly studied by Haber^ and others throughout 

 the entire range of temperatures found to prevail at this volcano, and 

 may be accepted without limitation as an important factor in the 



iThe scale on tho Iron pipe after exposure was found to contain over 85 per cent of 

 ferrous oxide and about 8 per cent of sulpliur. 



2 See, for example, F. Haber : " Thermodynamik teclinlscher Gasreactionen," p. 138. 

 Munchen, 1905. 



