748 



H. S. WASHINGTON 



to the rocks of Sardinia, as has been aheady pointed out.' The 

 presence of a noteworthy amount of nickel is also of interest. The 

 most striking feature — -when the presence of "leucite" is con- 

 sidered — -is the small amount of alkalies and the dominance of soda 

 over potash. The presence in all three of from 2^ to 4 per cent of 

 water, in spite of the evident freshness of the rocks, is also note- 

 worthy. 



TABLE II 



A. Analcite basalt [III. (5) 6. 2. 5]. Scano, Monte Ferru, Sardinia. H. S. Washington, analyst. 



B. Analcite basalt [IH. 5. (2) 3. (4) 5]. Monte Columbargiu, Monte Ferru. H. S. Washington, 

 analyst. 



C. Analcite basalt [III. 6. i (2). 4]. Bonorva, Sardinia. H. S. Washington, analyst. 



D. Analcite basalt [III. 6. 2' 4 (5)]. The Basin, Cripple Creek, Colorado. Hillebrand, analyst. 

 Cross, Jour. Geol., V (1897), 689. 



E. Leucite basalt [III. 5. 4. 4]. Dobernberg, near Tetschen, Bohemia. R. Pfohl, analyst. J. E. 

 Hibsch, T.M.P.M., XV (1896), 255. 



F. Analcite basalt [III. 5." 4." 5]. Rathjordan, County Limerick, Ireland. G. T. Prior, analyst. 

 G. T. Prior, Min. Mag., XV (1910), 317. 



G. Leucite basalt. Scano, Monte Ferru, Sardinia. C. Doelter, analyst. C.T)oe\.ttr, Denks. Akad., 

 Wiss. Wien, XXXIX (1878), 80. 



Doelter's analysis (G) of the Scano rock is obviously at fault in 

 several particulars. The alumina is much too high because of the 

 non-separation of Ti02 and P2OS, and probably also in part through 

 the incomplete separation of MgO from AI2O3, as his MgO is much 



^ H. S. Washington, Q.J.G.S., LXIII (1907), 69. 



