GENERAL DISCUSSION 277 



though other volcanic vents ma}' be and are of this type. Their number 

 and contiguity on the terrace (2), the lack of S3^nchronism (3), their 

 small size (4), the presumably great depth of their conduits (5), and, 

 above all, their situation near the edge of a scarp (6), all negative such 

 an origin. Any explosive action sufficient to establish connection between 

 the assumed magma reservoir and the surface could not permit of the 

 formation of a number of small conduits close to each other, with non- 

 synchronic activity, and situated close to a scarp, tlirough the face of 

 which, with its many planes of weakness, any such activity would most 

 easily j&nd outlet. 



To account for such a combination of characters as the Stromboli vents 

 present, there must be invoked an agency or mode of action that is at 

 once quiet, but effective in penetrating the overlying masses; with not 

 only preponderatingly, but essentially, vertical direction of action, and 

 with but limited lateral extension ; with the possibility of action at several 

 points in close contiguity and without mutual connection near the sur- 

 face, and, finally, the possibility of the continuance of the conduits and 

 the activity of their vents in the same locations during a considerable 

 period of time. 



The only agency which seems competent to satisfy all these (somewhat 

 complex and difficult) conditions is that suggested by Daly^* in his "gas- 

 fluxing" hypothesis. Indeed, so difficult is it to account for the char- 

 acters and situations of the Stromboli vents, and also for the situation 

 of the vents on the brink of the Val del Bove scarp and Byron's Ledge 

 at Kilauea by any agency that involves extensive or violent explosive 

 action or fracturing at depth, that the formation and existence of such 

 vents must be considered as not only confirmatory of, but as demonstrative 

 of, the validity of such a theory as Daly's in these cases, and presumably 

 in others, though of course not necessarily in all. 



Briefly put, Daly's hypothesis is as follows: Gas bubbles rise in the 

 magma to the top of the assumed reservoir and accumulate in pocket-like 

 irregularities ("cupolas") in the roof. They are supposed to be more 

 highly heated than the magma itself, chiefly through exothermic chemical 

 reactions, and thus melt or "blow-pipe" their way up through the super- 

 jacent rock. This action is sharply localized by l)eginning in greatest 

 intensity at one or several cupolas, will be predominantly vertical in direc- 

 tion, and "the size (of conduit) will be small because the fusing power of 

 emanating gas must be strictly limited." Thus more than one conduit 

 may be formed which will be independent one of another. After a con- 

 duit has been formed, it will tend to persist as long as there is a supply 

 of gas. 



R. A. Daly : Igneous rocks, 1914, pp. 251 ff. 



