OF THE WESTERN ISLES OF SCOTLAND. 345 



distiuctioii between the Tertiary hornblende- and augite-andesites 

 and the supposed older dio rites and diabases could no longer be 

 maintained. This conclusion they were able to establish by a careful 

 examination of the extensive materials collected by the officers of 

 the United-States Geological Survey. They proved that the diorites 

 are only deep-seated portions of the rocks, which are poured out at 

 the surface as hornblende- or mica-andesites, and that both alike are 

 of Tertiary age ; the diiferences between them were shown to be 

 due, not to the period of their formation, but to the more perfectly 

 developed crystallization in the deeply seated masses. The " dia- 

 bases " of the district w^ere also proved to be similarly related to 

 the augite-andesites. 



In the second edition of his ' Massige Gesteine,' published in 

 1886 and 1887, Professor Rosenbusch admirably summarizes the 

 results which have been obtained by the study of both the European 

 and the American propylites. Admitting, with von Eichthofen and 

 Zirkel, the extreme modification of the constituent minerals, and 

 the frequently well-developed crystallization, that recall so strik- 

 ingly the characters of the diorites and diorite-porphyrites, he shows 

 that the former characters are unquestionably due to the peculiar 

 kiiul of alteration that the rocks have undergone. The term " pro- 

 pylite " is therefore accepted only as serving to distinguish a well- 

 marked and interesting facies of the andesitic type of rocks — "a 

 pathological variety," employing Kosenbusch's apt designation. 

 Used in this way, the term propylite may still be of great service 

 to petrographers and field-geologists as a descriptive name ; just as 

 the terms shale, melaphyre, and porphyrite are convenient, and 

 even necessary to us, for describing peculiar modifications of clay, 

 basalt, and andesite respectively. It is in this sense that I propose 

 to use the term propylite in the present memoir. 



I shall be able to show that a careful study of the oldest Tertiary 

 igneous masses in the Western Isles of Scotland makes us acquain- 

 ted with a series of rocks presenting all the distinctive features of 

 the " greenstone-trachytes " of Hungary and Transylvania, and of 

 the " propylites " of California and Nevada. 



I shall endeavour to illustrate the exact nature of the processes 

 by which ordinary andesites (and rocks which in their structure 

 and degree of crystallization arc intermediate between andesites and 

 diorites) have been converted into the curious varieties that present 

 all the characters of the " greenstone-trachytes " or " propylites "; 

 and it will be my especial aim to investigate the exact causes to 

 which these peculiar modifications must be assigned. 



III. PuYSrCAL ClIARACTEKS AND ChEMICAL CoMrOSITIOX OF THE 



Scottish Pkopylites. 



In colour these rocks vary from a very dark grey, almost black 

 tint, through many Hgliter shades, to varieties that are nearly white. 

 Usually, however, more or less marked green tints are exhibited by 

 them, and in some cases this colour becomes very pronounced. 



