52 TEOF. T. G. EOXNEY AXD THE EEY. E. HILL : [Feb. I9I2,. 



sequence of the action of an aplitic magma on an almost pure horn- 

 blendite/ and I have noted since then other instances of similar 

 change. The streaked red dykes of Castle Cornet afford, as I 

 believe, frequent examples, for in many of them it cannot, I think, 

 be doubted that the small biotites have been produced by the 

 melting-down of the hornblendic constituent of the diorite ; but in 

 the latter rock, elsewhere in Guernsey, biotite sometimes seems to 

 replace part of a hornblende-crystal, either by taking the place of 

 about half of it, or by enclosing a core of it, more or less com- 

 pletely. These cases, like some described in 1898/ suggest occa- 

 sionally that the biotite-molecule has been substituted for the horn- 

 blende-molecule in building crystals of the latter. If, however, there 

 has been a subsequent change, this would require the removal of 

 lim.e and the addition of a moderate amount of alkaline silicate ;. 

 which, perhaps, would not be a very difficult matter if the rock 

 were still at a high temperature.^ 



(y) Structure of the acid dykes. — In the acid dykes the 

 relations of the felspar and the quartz differ from those in an 

 ordinary granite. In the latter, as a rule, the felspar is more or 

 less rectilinear in outline ; in these dykes it is rather irregular, and 

 the rock assumes a granular structure. Full thirty years ago I drew 

 attention to the possible significance of this structure,^ and since 

 then have more than once pointed out that it is sometimes accom- 

 panied by a slight elongation of the grains and indication of a 

 fluxion-structure, being probably the result of a little movement in 

 the later stages of crystallization.^ The occasional presence of an 

 interstitial mosaic of fine-grained quartz and felspar accords with 

 that explanation, for it suggests the more rapid crystallization of a 

 small amount of residual magma, which has acted like a lubricant 

 to the grains in the nearly consolidated mass, and then has been 

 either divided between the nearest crystals of the principal con- 

 stituents, or sometimes, perhaps, has locally approached an eutectic 

 mixture, and thus has rather suddenly solidified. The structure,, 

 as I remarked in 1885, is common in a vein-granite ; it is note- 

 w^oFthy in the ' granulitic ' group of the Lizard, and in a rather 

 similar group in 8ark, both of which are now believed to be 



' Q. J. G. S. vol. xlviii (1892) pp. 129-32. 



2 Ibid. vol. liv (1898) pp. 366 & 369. 



3 If Ave take, for purposes of comparison, the ferroniagnesian constituents of 

 tliese two minerals as vmity, this gives, in an average (ahiminous) hornblende, 

 the proportion of silica as also about 1, of alumina "5 or '6, of lime about '6 ; in 

 a biotite, silica about 1"3, alumina '5, potash (or alkalies) "o. Both biotite and 

 hornblende seem to form rather rapidly as secondary minerals in a rock that 

 l\as been much affected by pressure. See Q. J.G. S. vol. xlix (1893) p. 104 ; and 

 T. Gr. Bonney & 0. A. Raisin, in ' Climbing & Exploration iu the Ivarakoram 

 Himalayas,' "bj W. M. Conway, 1894, vol. ii, p. 54. 



•* I then thought that this irregularly-outlined structure (like that of some 

 quartzites) might distinguish gneiss, which was regarded as the extreme stage 

 of metamorphism in a sedimentary rock, from a normal granite. 



^ Q. J. G. S. vol. xxxvi (1880) p. 97; vol. xli (1885) Proc. p. 64; also 

 'Foundation-Stones of the Earth's Crust' 1888, p. 8. 



