750 H. S. WASHINGTON 



probable in many cases. Indeed, in view of the experimental 

 formation of analcite at temperatures of 500° and higher, the known 

 presence of water in rock magmas, and the great lowering of the 

 melting-point through its presence, the primary character of the 

 analcite is the only rational view in such cases as these and many 

 others, where the rock is unquestionably fresh and unaltered. 

 Such an explanation is far simpler and more intelligible than the 

 invocation of the entire, or almost entire, replacement of potash by 

 soda in a single mineral of the rock through assumed soda-bearing 

 solutions of quite hypothetical origin, with no alteration of the 

 other minerals. At the same time, of course, there is no question 

 that in some cases, where the rock has undergone alteration, the 

 analcite present is secondary, as it is in that from Rathjordan. 



It may be mentioned that the occurrence of primary analcite in 

 monchiquites and analcite basalts and other such rocks has a bearing 

 decidedly adverse to Brun's hypothesis that lavas are water-free, 

 as does also the fact, well known to all petrographers, that many 

 unquestionably fresh and recent pitchstones and other glassy rocks 

 carry up to 10 per cent of H2O. 



With my three analyses are given others of closely similar rocks, 

 which likewise contain small, rounded crystals of a mineral which 

 looks like leucite and which in all cases is entirely isotropic and 

 without twinning. Indeed, those from Colorado and Ireland were 

 considered to be leucite basalts until chemical analysis showed the 

 true character of the isotropic minerals. These two carry no 

 biotite. In the case of the Bohemian rock, which, like those of 

 Binzale Prunu, contains large plates of biotite, Hibsch apparently 

 overlooked the significance of the dominance of soda over potash. 

 This has been pointed out by Prior in the paper cited. Chemically, 

 they are much like the Sardinian lavas, but are higher in CaO and 

 also in Ti02. The last two are also much lower in soda. 



Classification.— -The norms of the three Sardinian basalts ana- 

 lyzed by me are given in Table III. From this it appears that the 

 Scano rock (A) falls in the subrang represented by the symbol 

 III. (5) 6. 2. 5, there being just sufficient normative nephelite pres- 

 ent to be taken into consideration. This subrang is as yet only rep- 

 resented by the Scano rock and a basalt glass from the Val di Noto, 



