524 abstracts: petrology 



It was also found possible in several instances to determine the com- 

 position of the liquid phase, present at temperatures within the melt- 

 ing interval, by measuring the refractive index of the quenched glass. 

 Thus points on the liquidus were determined by an independent method 

 and were found to be in excellent agreement with the results of the 

 temperature method. 



In the theoretical discussion, equations are developed which express 

 the concentrations on the liquidus and solidus at any temperature in 

 terms of the melting temperatures and latent heats of melting of the 

 two components. The experimental results therefore make possible the 

 calculation of the latent heat of melting of anorthite and of albite. The 

 calculated values are 104.2 calories per gram for anorthite and 48.5 

 calories per gram for albite and these values remain practically constant 

 for all ranges of composition. The calculated latent heat of anorthite 

 is in excellent agreement with the figure found by direct measurement, 

 105 calories per gram. No direct determinations of the latent heat of 

 albite have been made. It is shown that if these values of the latent 

 heats are taken and liquidus and solidus curves calculated, the resulting 

 curves pass very close to the experimentally determined temperatures 

 (within the limits of error of the temperature measurements) . It is also 

 shown that values of the latent heats differing from these by as little as 

 10 per cent will not give a like result. This extreme agreement with 

 the requirements of theory and its bearing on certain theoretical ques- 

 tions is discussed. 



The geological significance of the complete solid solution of the feld- 

 spars is considered, as well as the extent to which zoning may occur 

 under favorable conditions and the consequent great range of tempera- 

 ture thru which plagioclase may crystallize. N. L. B. 



PETROLOGY. — Graphical methods in microscopical petrography. Fred 

 Eugene Wright. Am. Jour. Sci. (4), 36: 509-539. 1913. 

 Experience in microscopiqal petrography has shown that the results 

 furnished by graphical means are, as a rule, sufficiently accurate and 

 in accord with the quality of the data of observation. Graphical 

 methods in petrography serve three purposes: (1) to solve certain equa- 

 tions, (2) to represent data of observation, and (3) to picture certain 

 important crystallographical and optical relations. In all these cases 

 it is essential: (a) that the graphical means employed represent the 

 relations adequately and as free from distortion as possible; (b) that 

 they are easy of application, and (c) that wherever possible the func- 



