296 A. GEIKEE ON THE SUPPOSED 



I was inclined to regard the paler varieties, both in the form of fine 

 tuffs and of breccias, as having probably resulted mainly from the 

 destruction of more siliceous lavas, probably of fine-grained felsites 

 or other acid rocks. These inferences in the field have now been 

 confirmed by chemical analysis and microscopic examination. 



That many of the tuffs are due to the destruction of diabase lavas 

 may be surmised from their close general external resemblance to 

 these rocks, and from the way in which they are associated with the 

 contemporaneous sheets of diabase. Some of the dull dark-purple 

 tuffs below the crag of Ehosson, and again to the north of Clegyr 

 Eoig, might almost at first sight be mistaken for truly eruptive rocks. 

 Typical specimens taken from different parts of the district were 

 analyzed for me by my colleague, Mr. J. S. Grant Wilson, in the 

 laboratory of the Geological Survey, Edinburgh, and by my friend 

 M. Eenard, of the Eoyal Museum, Brussels, with the following 

 results. 



Analysis of Basic Tuffs from the St. David's District. 



I. Purplish-red shaly tuff from below the olivine-diabase crag, 

 Ehosson. (Analyzed by Mr. "Wilson.) 



II. Dull purple and green tuff from the lowest group of tuffs 

 between Pen-maen-melyn and Pen-y-foel. (Analyzed by Mr. 

 Wilson.) 



III. Greenish, shaly, finely granular tuff from the road-side, north 

 of Board Schools, St. David's. (Analyzed by M. Eenard.) This 

 specimen was chosen as one of the intermediate varieties between 

 the basic and acid types ; and the position thus assigned to it is 

 confirmed by the analysis. 



I. II. III. 



Silica 5125 4811 6154 



Alumina * 2041 13-30 1630 



Ferric oxide 3.02 370 440 



Ferrous oxide 3-91 810 • 3*66 



Manganous oxide 021 1*43 0-32 



Lime 4-53 8-48 308 



Magnesia 7'22 9-51 2-99 



Potash 2-93 1-57 1'62 



Soda 1-82 1-96 281 



Loss on ignition and water 502 4*21 2-99 



100-32 100-37 99-71 



Specific gravity 2'84 2-92 



Comparing the composition of JNos. I. and II. of these analyses 

 with those of the diabases given on p. 303, we readily perceive how 

 the tuffs might have been derived from lavas like those of Ehosson 

 and Clegyr Foig. The differences between them are not greater than 

 might be expected, if such were the source of the tuffs. There 

 would be a partial decomposition of the volcanic dust and lapilli by 

 the water into which they fell ; and there might also be an inter- 

 mingling with the ordinary non-volcanic sediment that happened to 

 be in course of transit and deposit in the locality at the time of the 



