Ralemaibmau Lava Lake. 339 



August 11th, the flow of surface lava was from under the bank 

 at the western end of the lake, where it was evidently rising 

 from below, to a number of points at or near the eastern end, 

 including the localities of the large fountains, where it was 

 intermittently or continuously descending.* Now, we know 

 that the basin does not concentrically overlie the conduit, but 

 is greatly extended to one side (the west) as though excavated 

 in that direction. We might deduce this conformation by 

 the localization of the large lava fountains in the eastern half 

 of the lake, these being due to the rapid ascent of great gas 

 bubbles directly from the conduit, but the survey of the empty 

 crater basin by Baldwin in 1902 also showed the eccentric 

 position of the conduit, it standing then, as now, under the 

 eastern half of the crater cavity. But, as the reader will 

 already have perceived, there is here an apparently anomalous 

 state of things, viz.: that it is in the eastern half of the lake, 

 which overlies the conduit and where the large bubbles rise 

 and form the fountains, that the lava is going down, while, per 

 contra, it is rising — quietly and free from large gas bubbles — 

 at the western end of the lake, as far as possible from the con- 

 duit itself. We should, therefore, have already been far 

 astray had we based our study on the behavior of convection 

 currents in a simple liquid. 



Let us now attentively observe the surface material in its 

 passage across the lake. Issuing in clear incandescence from 

 under the western bank the surface is instantly chilled on com- 

 ing into contact with the outer air and a film is formed which, 

 from its appearance, has been generally designated as "scum." 

 The word is inappropriate as regards the nature of the material 

 which is that of a continuous skin, flexible at high tempera- 

 ture and very tough. Being formed quickly it is non-crystal- 

 line and of a high degree of diathermacy which, however, is 

 limited by entangled gas vesicles. Congelation to the under 

 surface will, therefore, continue but with decreasing rapidity 

 until crystalline deposition begins, whereupon the radiation of 

 heat from below will be greatly retarded and the process of 

 thickening will proceed more slowly but with the formation of 

 a true crust of lava rock. As the surface layer spreads out 

 from the place of issue into the body of the lake the skin is 

 pulled apart in irregular lines through which glows the bright 

 lava below as shown in the night view (Hg. 1). It is important 

 to note that, although the lava rises quietly here at the west 

 end of the lake and is free from the great gas bubbles which 

 form the large fountains, it is, nevertheless, highly charged 

 with niinute gas vesicles some of which, coalescing, burst 



* The records show that, for a number of years, this condition has quite 

 generally prevailed. 



