142 CARNEGIE INSTITUTION OF WASHINGTON. 



this ratio becomes identical with that of the siUcate richer in silica, the melt 

 is free from carbon dioxide. In each case the compound containing more 

 silica is the poorest in silica which can be prepared pure at the particular 

 temperature merely by putting together carbonate and silica. Moreover, 

 rise of temperature, the pressure remaining constant, favors the lower 

 silicate. 



The study of systems of this type is important because their behavior 

 serves as a simple prototype of that of the magma, which is a highly com- 

 plex system containing both volatile and non-volatile components. The 

 so-called " mineralizers " are merely volatile components; the effects of 

 their presence differ only in degree from that of the other components. 

 The main difference is due to the vastly greater effect of changes of pressure 

 and of temperature on the concentration (in the melt) of the volatile com- 

 ponent, by reason of the close relation of the concentration to that of the 

 gas phase itself. In consequence of this, there is a "mobility" of equi- 

 librium which is characteristic of the magma and without doubt very closely 

 associated with many aspects of its behavior, e. g., with the likelihood of 

 eruption, differentiation, etc. 



(28) Gleichgewichtserscheinungen zwischen Alkalikarbonaten und Kieselsaure. Paul 



Niggli. Z. anorg. Chem., 84, 229-272. 1913. 



A German translation of "The phenomena of equilibria between silica 

 and the alkali carbonates" (J. Am. Chem. Soc, 35, 1693-1727. 1913). 

 Reviewed under No. 27 above. 



(29) The electrolytic reduction of iron for analysis. J. C. Hostetter. J. Wash. Acad. 



Sci., 3, 429-432. 1913. 



The conditions for the complete electrolytic reduction of ferric sulphate 

 to ferrous are discussed and some results are presented to show the con- 

 venient appHcation of the method to the determination of iron. 



(30) The volcanoes and rocks of Pantelleria. Henry S. Washington. J. Geol., 21, 



653-670, 683-713. 1913. 



The island of Pantelleria was studied in September 1905, under Grant No. 

 95, and the paper describes both the volcanic structure and the petrography 

 of the very interesting lavas. 



Pantelleria (which lies about half-way between Tunis and Malta) is 

 entirely volcanic. It consists of an early, large cone, composed of trachytes 

 and pantellerites. After a large explosive caldera was formed in this, a 

 second volcanic phase began, consisting of the building up of a trachyte core 

 within the caldera. Later this was faulted and a large block tilted down, 

 smaller cones and jflows of glassy pantellerite being poured out about this. 

 Volcanism ceased with the formation of many small basaltic cones on the 

 flanks of the earliest cone. 



The important lavas are described in great detail, 17 complete chemical 

 analyses having been made. The trachytes and pantellerites are interesting 

 because they carry abundantly well-developed crystals of soda-microcline, 

 an unusual feldspar, the crystals of which are to be investigated optically 

 and chemically later. The latter rocks are also noteworthy for their con- 

 tent in the rare triclinic, sodic amphibole, cossyrite. The basalts are of a 

 common type, but their occurrence here in connection with such highly 

 sodic rocks is of interest. 



