Vol. 54.] CHLORITOID IN KINCARDINESHIRE. 151 



or base the colour is pale green or greenish-yellow, varying con- 

 siderably with the thickness of the section ; when parallel to the 

 twin-face the colour is the very characteristic, clear indigo-blue. 

 The double refraction is not high ; it seems to be distinctly lower 

 than that of hornblende, though, according to the tables of minerals, 

 it should be slightly higher. On mounting some of the isolated and 

 broken-up crystals in balsam, the high refractive index is well seen. 

 Besides this, we now note the rapidity with which the strength of 

 the pleochroism diminishes as the fragments become thinner — an 

 important point, as it explains the very different aspect, especially 

 so far as colour is concerned, of the minute crystals seen in the 

 slate and the larger ones of the schist. That the mineral possesses 

 a fairly well-marked basal cleavage is at once obvious, for almost 

 all the fragments are lying on their basal plane. Here and there 

 we find traces of a prismatic cleavage, the angle being a little over 

 60°. It is never distinct enough to permit of accurate measurement. 

 Indeed, the comparative absence of any such cleavage is really of 

 more value for purposes of identification than the fact that obscure 

 traces of it may be met with occasionally. Examining suitable 

 flakes, we may note : — 



a = a = olive-green. 



b = (3 = deep indigo-blue. 



c =: y = pale green-yellow. 



The mineral is positive, and shows the oblique emergence of a 

 positive bisectrix. 



The above characters show that, optically, it is a typical 

 chloritoid. 



Its hardness may be expressed by the statement that it will just 

 scratch glass, and will not scratch fresh microcline. An ordinary 

 knife will not scratch it, but, owing to the inclusions being often 

 chlorite and white mica, the knife will frequently appear to do 

 so, when in reality it is scratching only the softer material. 



Chemical Composition. 



The preparation of a sample sufficiently pure for chemical analysis 

 has given great trouble, and taken much time. As the sections 

 show, not only do the crystals contain many inclusions, but the 

 edges are very jagged, and the enveloping material, largely chlorite 

 and white mica, is felted into these jagged edges. Further, it was 

 found that the material of the white spots was heavier than the 

 densest solution either of cadmium borotungstate or of methylene 

 iodide. After repeated trials, a large quantity nearly pure was 

 obtained by the following process : — Several pounds of the rock 

 were powdered up sufficiently small to pass through fiue muslin. 

 The powder was placed in a large enamelled-iron developing-dish, and 

 water from a tap was allowed to fall slowly onto it. By this means 

 more than two-thirds of the material was got rid of, without any 

 appreciable loss of the mineral sought. The residue was placed in- 

 a weak solution of hydrochloric acid and heated over a steam-bath 



m2 



