Lieut.- General C. A. McMahon — AUanite in Granite. 195 



a good pocket lens. The mineral possesses considerable lustre, and 

 bright facets catch the light and facilitate its isolation. Crystalline 

 faces are numerous and often suggest those shown at fig. 4, p. 522 

 of Dana's " System of Mineralogy," 6th edition. The fracture of the 

 mineral is subconchoidal. 



The specific gravity of the Lairg mineral is over 3-32, and con- 

 sequently it sinks readily in methylene iodide. It would be difiicult 

 to ascertain the exact specific gravity of the Lairg mineral, as, owing 

 to the microscopic size of the fragments, heavy liquids of viscous 

 character — as, for instance, cadmium borotungstate above a certaia 

 density — are of no use. 



The refraction of the mineral is also high, being well over 1-740 

 that of methylene iodide. On the other hand its double refraction 

 is weak, exhibiting no colour higher than a decidedly dull yellow of 

 the 1st order of Newton's scale. 



The pleochroism of the Lairg mineral is usually distinct and 

 occasionally strong, the change of colour being from a brown-red, 

 or red-brown, to a brown-j^ellow. Sometimes, but rarely, there is 

 a suspicion of green in the brown. 



As regards the optical character of the crystals the Lairg specimens 

 are positive. I have been able to observe the interference figure in 

 convergent polarized light in many of the isolated fragments, and 

 its sign is uniformly positive. Dana, in his " System of Mineralogy," 

 6th edition, 1892, says that allanite is optically negative ; but Michel 

 Levy, in his " Les Mineraux des Eoches," gives the sign as ±. The 

 Lairg specimens therefore support Michel Levy's view. 



Mr. J. J. Harris Teall, F.K.S., has very kindly allowed me to 

 examine thin slices of another Scotch granite, that of Hildesay, in 

 his private collection, in which a precisely similar mineral to that 

 in the Lairg rock occurs, which he had previously suspected to be 

 allanite. The mineral in these two rocks are clearly identical, and 

 the description of one would do for the other. Extinction in one 

 of Mr. Teall's crystals is nearly straight, showing elongation 

 approximately parallel to the orthopinacoid, but in other crystals 

 it varies, being generally about 34°. One of the crystals shows 

 the zonal structure well. 



In the Hildesay granite the allanite is associated with much. 

 epidote, but this mineral does not appear to be present in the 

 Lairg rock. My slice contains none, and I have not observed any 

 among the crushed fragments. 



The allanite of the Lairg hornblende-granite appears to be an 

 original constituent ; it is corroded by the quartz of the groundmass, 

 and it encloses numerous mici'o-crystals which appear to represent 

 two mineral species ; but they are too minute for exact determination. 

 The Lairg mineral fuses easily in the flame of a Bunsen burner, 

 but is very slightly attacked by hydrochloric or by nitric acids. 

 When ground to a fine powder it yields on prolonged heating to 

 aqua regia. 



On all the points noted above, the Lairg mineral agrees with 

 allanite; and as I know of no other mineral which, on all points, 



