26 Mr J. D. Dana on the Volcanoes of the Moon. 



as the fjict may be witnessed in any boiling cauldron ; and 

 the lavas of Kilauea are a visible example of it. Tlie ebul- 

 lition in lavas on the earth, proceeds principally from the 

 vapours of water and sulphur, which are constantly rising 

 through them, inflating them more and more as they ascend, 

 and finally escaping in bubbles at the surface. Now the 

 feldspar being the less fusible part of the lavas, would 

 thicken somewhat, wherever the temperature became too 

 low for complete fusion ; the more liquid portion would then 

 ascend tnost easily, being carried along by the inflating va- 

 pours, and much of the feldspar would thus be left behind, 

 and it might be in a nearly pure state. The centre of the 

 volcano under this action becomes necessarily feldspathic. 

 The summit might, therefore, eject either basaltic or feldspa- 

 thic rocks from the material of the vent ; though when the ac- 

 tion was violent and deep, it would eject feldspathic rocks alone. 



At the same time the basaltic lavas, descending laterally 

 in this system of circulation along the sides of the great cen- 

 tral conduit, may pass out as flank eruptions through fis- 

 sures. Besides, there will also be basaltic ejections from 

 sources of lavas at a distance from the central conduit, where 

 they have not been subjected to the separating process de- 

 scribed ; and this may be the more common source. 



Mountains with a feldspathic centre, and basaltic layers 

 forming the circumference, are therefore quite intelligible 

 without supposing the feldspar to have been first thrown up, 

 or appealing to a different system of fissures for their origin, 

 and the examples which the moon presents, are more exten- 

 sive than is necessary to explain the widest facts on the 

 earth.* 



* It is common to say that certain domes of tracliyte were thrown up in a 

 pastij or imperfectly fluid state, in order to account for the fact, that there is 

 no appearance of the rock's having flowed in streams. Without intending to 

 refer the origin of these domes to any particular cause, I woiild suggest the 

 query, whether, if their formation was subaerial, this past^j state does not neces- 

 sarily imply that the ascending vapours would have found some difficulty in 

 escaping, and would have broken through with explosions, as explained in the 

 foregoing pages ; and, consequently, that there would have been scoria and cin- 

 ders accompanying the ejections ? or may we believe it probable, that the paste 

 was so dense that water would not make its way up and escape as vapour ? Is 



