362 HAROLD L. ALLING 



lime it is evident that the solubility lines of the more basic plagio- 

 clases maintain a more constant position irrespective of the rate at 

 which they have been cooled than do those of the alkali-rich feld- 

 spars. Thus we can expect to encounter such feldspars as potash- 

 oligoclase more frequently than potash-bytownite. 



It becomes necessary, therefore, to project all plotted data, spe- 

 cific gravities, extinction angles, indices of refraction, optical angles, 

 crystallographic angles, etc., upon a triangular base and employ 

 contours. Figures 4, 5, and 6 represent a revision and extension 

 of the work suggested in Part I. 



SPECIFIC GRAVITIES 



The diagrams of the specific gravities of feldspars given in Part I 

 (Figs. 12-6 and 12-7, p. 250) were only tentative, hence the effort is 

 now made to improve their quaHty as it is believed thay may be of 

 value to petrographers and mineralogists. The writer tabulated 

 all of the available analyses of feldspars of which the specific gravities 

 were given. These analyses were then recast into three components. 

 This information is to be taken with caution in that many of the 

 chemical analyses are undoubtedly inferior to those demanded 

 today. Although the writer wishes he possessed the skill, the facili- 

 ties and the time to secure afresh this needed information, he was 

 forced to make use of these rather uncertain data. 



From these data a "peg model" was constructed which clearly 

 showed two sets of pins, one for the "monoclinic" and the other 

 for the "triclinic series." The results of this study are given in 

 Figure 4. 



INDICES OE REFRACTION 



A similar ''peg model" was constructed from analogous data 

 giving the beta values of the index of refraction of the potash-soda 

 lime feldspars. Here the data is less comprehensive and the diffi- 

 culties of the interpretation are consequently greater. At first it 

 was thought that it was not possible to recognize two sets of pins, 

 representing the two series, and as a result the contoured surface of 

 the plaster model made from the peg model was most irregular. 

 Experience with the specific gravity peg model led the writer to 

 attempt to classify the pins into two sets, each representing a sur- 



