F. A. Perret — Lava Eruption of Stromboli. 453 



these recent Stromboli lavas differs markedly, notably in lower 

 silica, from that of the older basalts of San Bartolo (Si0 3 = 

 52 - 25) and still more from that of the andesite flows at the 

 earliest Vancori cone (SiO. J =61 , 78). In this respect, aiso, 

 Stromboli resembles many other volcanoes. 



The writer took pains to coat specimens of the basalt, taken 

 incandescent from the flow, with paraffine as soon as they were 

 sufficiently cool, in order to prevent diffusion of gases from the 

 interior, and it is hoped that this lava, when heated in vacuo, 

 will yield its original gas content for analysis. Direct collec- 

 tion of gas by any other means was not feasible. 



The lava streams have formed upon the surface of the Sciara 

 a very considerable mass, in the form of a fan, with a base 

 line of perhaps six hundred meters. The shape is due to the 

 continual shifting of the course of the flow by the upgrowth of 

 its bed through cooling, with consequent lateral overflow, and 

 to the convexity of the solidified mass which diverted the later 

 material, flowing from above, to one side or the other — in 

 point of fact, the two latest flows are respectively the most 

 easterly and the most westerly of all. 



The total quantity of lava emitted during the eruption cannot 

 be computed by the usual method of cubic measurement, as 

 the larger part is beneath the sea surface, and it is impossible 

 to estimate the quantity on the basis of the rate of flow because 

 it was not under observation during four fifths of its period of 

 eruptive activity. 



One of the most interesting obsciwations in connection with 

 this eruption undoubtedly is the behavior of the lava stream 

 on coming in contact with the water of the sea and during its 

 continued flow as a sub-aqueous stream, and this not only 

 because of the light which may be shed upon the manner of 

 growth of the many volcanoes which have begun their 

 existence at the bottom of the sea, but also on account of the 

 paramount question of the absorption of water by hot lava — 

 whether, that is to say, with the two in contact and under a 

 certain amount of pressure, there will or will not be any actual 

 acquisition of water by the lava which may result in chemical 

 or physical changes therein. 



It goes without saying that rapidly flowing lava — especially 

 if containing; lars-e gas bubbles under high tension — will, on 

 entering water, cause considerable commotion and generate 

 great quantities of steam with some appearance of violence. 

 But when massive, highly incandescent lava, free from large 

 gas bubbles, comes slowly into contact with water, also in mass, 

 there is an absence of explosive phenomena or of serious 

 commotion which at first is very surprising. At Sakurashima * 

 *F. A. Perret, Zeitsch. Vulk., i, p. 143, 1914. 



