«2 Proceedings of the Royal Society of Victoria. 



of water upon these heated layers ; 8rd. Tlie escape tlirougli 

 fissures or vents of high pressure steam, caused by the contact 

 of the water with the molten lava. 



Professor Judd considers that the immediate cause of 

 eruptions is due to the escape of high jiressure steam through 

 the lava, in the form of bubbles which rise and burst at or 

 near the surface, much as would be the case if pon-idge weie 

 to be boiled in a tube. Now this does not appear to satisfy 

 the conditions required for explaining the various phases of 

 the eruptions of Mouut Yasur, particularly the normal 

 quiescent regular state. 



With the aid (jf a diagram (plate 8), let me suggest another 

 method, or it might be called a moditication of the above. 



I look upon tlie Tana volcano as a lava geyser or a natural 

 underground steam machine, depending for its peculiarities 

 upon the nature of the arrangements of the Avater supply, 

 steam chamber, and outlet vent. I assume a lake of molten 

 lava (F), which at normal atmosph<3ric pressure maintains in 

 the vent (C) a column at about a constant level. (D) is a 

 space above the lava, filled with steam. The vent is tapped 

 at a point below that at which the lava stan :k by a cave or 

 fissure (B), into which the water supply from the lake (A) finds 

 its way. This watei coming into contact with the heated 

 lava is turned into steam, which on the pressure becoming 

 sufficiently strong to overcome the resistance, is forced under 

 the column in the vent and ejects it with great violence. On 

 the escape of the steam, the lava rises again immediately to 

 its former level in the vent, and the steam making process 

 goes on as before. The heat, the atmospheric pressure, and 

 the water supply being constant, there is no reason why this 

 process should not go on for ever with the most i)erfect 

 regularity. 



I may state that 1 have tested this arrangement practically, 

 and have succeeded in obtaining the result as stated above, 

 i.e., a perfectly regular spasmodic ejection of fluid from a pipe 

 by the steady admission of steam at a point below its 

 surface. 



As the molten lava would also contain steam at considerable 

 pressure, or water ready to turn to steam, absorbed from the 

 ocean, the effect of the escape up the vent of a considerable 

 volume of free steam, and the emptying of what 1 have called 

 the steam chamber, would be to cause a sudden decrease of 

 pressure and an additional quantity of steam to be formed 

 which would aid in the ejection of the lava. 



