OF THE WESTERN ISLES OF SCOTLAND. 369 



such an extent that it is often impossible to distinguish between * 

 lavas and tuffs. 



The characters of the propylites of this particular district, as we 

 now see them, are most remarkable, not one of the minerals of the 

 original rock being present in it. The ground-mass has been com- 

 pletely altered, all traces of glass having disai)peared by secondary 

 devitrification, and many secondary minerals being developed in it. 

 The outlines of the original felspar-crystals can sometimes be made out, 

 bnt this is all. Their substance has been converted into aggregates, 

 among which epidote, zoisite, secondary felspar, and even quartz, 

 play the most important part. The ferro-magnesian silicates usually 

 appear as pseudomorphs in isotropic " viridite " or in some species 

 of chlorite. The resorption-halos of the hornblendes and micas can 

 often be detected by the clustered magnetite grains ; and certain 

 forms of biotite may be seen developed at the expense of the secon- 

 dary chlorites. Lastly, in addition to the original magnetite grains 

 of the rock, we find enormous quantities of the same material pro- 

 duced during the breaking up of the ferro-magnesian silicates, a 

 process which is so frequently attended with the separation of mag- 

 netite grains (see Plate XIV. fig. 7). 



As " epidotization " is the ulterior and most marked change of 

 which the propylite rocks exhibit evidence, it may be well to consider 

 the nature of the mineral species Avhich result from the change. 



From one of the extremely altered augite-andesites of Beinn 

 More, in Mull, Mr. W. B. D. Edwards isolated, by means of Klein's 

 solution, a considerable quantity of the beautiful green epidote, the 

 material proving, on microscopic examination, to be remarkably free 

 from foreign admixture. The specific gravity of the mineral proved 

 to be 3*42, and a partial analysis made by Mr. Edwards showed it 

 to be a lime-iron- epidote or pistacite. I find that all the optical 

 characters confirm the identification of the epidote in question with 

 this variety. 



Besides the beautiful deep green epidote which is most abundant, 

 pale-coloured lime-epidotcs occur, and some w^hich have a pale pink 

 colour, probably due to manganese. The highly coloured withamitc, 

 however, has not yet been detected in these rocks. 



It is a most suggestive circumstance that this solfataric action is 

 found to have been developed around each of the five great centres 

 which I have identified as the sites of the great volcanoes of the 

 Western Isles of Scotland. In most cases where this action can 

 be shown to have taken place intrusive masses of granite and 

 felsite can be shown to be in tolerably close proximity to the altered 

 rocks. 



In a very interesting memoir, M. de Lapparent has insisted on 



* I beheve that there cannot be any reasonable doubt tliat the district lying 

 immediately to the east of the summit, oC Beinn Moi-e must have existed under- 

 neatli what was the great central active crater of the ]Mull volcano, and in this 

 way the excessively altered condition of its rocks and the production of the 

 remarkable volcanic minerals described by me in 1874 is accounted lor. 



