224 REV. J. MILNE CURRAN. 



we cannot consider the felspars the source of the soda that 

 crystallises as sodium chloride. 



Nepheline if present would certainly answer to the tests just 

 stated, but a microscopic examination of the rock shows that the 

 mineral so readily soluble is isotropic. Certain minute clear six- 

 sided forms can be found under high powers in every slide, but 

 this mineral too is isotropic, and so far as I could observe no trace 

 of the uniaxial figure that nepheline would show, can be seen. 



Hauyne is excluded, as barium chloride gives no reaction in a 

 hydrochloric acid solution of the soluble portion of the rock. 



We have then an isotropic, and an easily soluble soda-bearing 

 mineral in a rock that gives a reaction for chlorine and contains 

 no sulphuric acid, evidence that points to the presence of sodalite 

 in this rock. 



Some of the powdered basalt was gently warmed in weak 

 hydrochloric acid (one part of acid to ten of water) for two hours. 

 The solution was then evaporated to dryness, and repeatedly 

 moistened and almost dried alternately for two days. Well 

 formed crystals of sodium chloride were then visible. The whole 

 was transferred to a glass slide, the surplus acid removed by intro- 

 ducing the edge of a filter-paper. The crystals were then dried on 

 the slide moistened with turpentine and mounted in balsam. 

 Plate ii, fig. 1, is a micro-photograph of the slide. Fig. 2 shows 

 cubes of Na CI obtained by etching a square inch of surface of 

 polished basalt. The face of the slab was just covered with a few 

 drops of strong hydrochloric acid. This was moistened again 

 with acid as it had gelatinised in three hours. The whole was 

 transferred to a glass slide, and after standing two days was 

 photographed. The figure shows the Na CI cubes just as they 

 appeared in the strong acid, x 30 diameters. 



A complete analysis of the basalt was next made in duplicate, 

 and three separate determinations of the soda were made, all 

 giving practically the same result. 



