296 !dk. h. s. Washington on the [June 19 14, 



equant subhedral individuals, roughly square in outline. They 

 generally show a line polysynthetic twinning according to the albite 

 law, Carlsbad twins being rare. The extinction-angles are low, and 

 correspond to those of albite. No microperfhite is to be seen. 

 Generally these crystals contain toward the centre some dusty 

 material, which high powers resolve into irregular specks of green 

 aegirite. Albite also forms some small thin laths, often in Carlsbad 

 twins. 



The quartz forms anhedral grains, and contains many minute 

 liquid inclusions, which cany gas-bubbles, the volume of the gas- 

 bubble being about 12 to 15 per cent, of the volume of the cavity 

 (Wright). The form of these cavities is sometimes negative. 

 There are a few small, apparently micrographic, intergrowths of 

 felspar and quartz. In these are to be seen very minute prismoidal 

 crystals of a colourless, highly-refracting, optically-positive, uniaxial 

 mineral, which must be referred to zircon. The total amount of 

 this in the sections cannot amount to more than about one- tenth 

 of 1 per cent. Extremely rare and minute grains of magnetite 

 are seen in places, and there are a few very small prisms of 

 apatite. I could find none of the rare blue hornblende mentioned 

 by Prof. Judd. 



Chemical Composition. 



The chemical analysis was made by me with the greatest care, by 

 the methods advocated by Hillebrand and myself. Zirconia and 

 the rare earths were especially looked for, because of the highly - 

 sodic character of the rock, it being known that these elements are 

 prone to association with sodic magmas. 1 The amounts of both 

 which were found fully justify the search. The ' Ce 2 3 ' of the 

 analysis (II, p. 297) was chiefly composed of cerous oxide, as was 

 shown by colour -reactions, but possibly yttrium, didymimn, or 

 other rare earths are present. Titanium and manganese were 

 determined colorimetrically. Not a trace of chlorine was leached 

 out of the rock by hot water. Nickel was not looked for, owing 

 to the small amount of material. 



It is evident that my analysis fully confirms the general features 

 of Makins's analysis. They both show high silica, feme oxide, 

 and soda, and low alumina, ferrous oxide, magnesia, lime, and potash. 

 The only noteworthy divergences in the determinations — those in 

 silica, alumina, and soda, and possibly iron oxides, may reasonably 

 be ascribed to the piece analysed by Makins containing a little more 

 quartz and a little less albite and pyroxene than mine, a divergence 

 which is perfectly possible when such small fragments only were 

 available. His high manganese oxide is probably in large part 

 alumina, as coprecipitation of this with the manganese, when the 

 usual sodium-acetate method is used, is a frequent cause of high 

 figures for manganese. The very small amounts of potash and 



H. S. Washington, Trans. Am, Inst. Min. Eng. vol. xxxix (1909) p. 755. 



