512 BerieicH — The Survey Memoir on the Scottish Uplands. 



reached the distant Girvan region, all the thin-bedded black shnle 

 zones of Moffat have disappeared as such, and are now represented 

 by thick rock sheets of very varied lithological types, rich in most 

 of the classes of invertebrate life. But through all these changes, 

 from end to end of the Uplands, we find at intervals among the 

 arenaceous deposits layers with characteristic Moffat graptolites, 

 "which enable us to parallel with certainty all the expanding 

 divisions of the Moffat succession, and trace them from point to 

 point, and district to district, through all their variations in lithology, 

 thickness, and palaeontology. 



The systematic place and geographical distribution of the zones of 

 the Moffat Series, typical and expanded, being thus established, the 

 remainder of the Upland formations fall naturally into line. The 

 overlying wide-spreading sheet of greywackes and grits (Gala or 

 Queensberry Series) is shown both by superposition and by 

 characteristic graptolites to be the equivalent of the Tarannon. The 

 succeeding formations of Riccarton and the Pentland Hills answer 

 both by systematic position and by fossils to the Wenlock - Lower 

 Ludlow formation ; and finally, resting upon these Wenlock-Ludlow 

 rocks in Lanai'kshire and Midlothian, come the Scottish equivalents 

 of the Passage Beds (Downtonian) of Siluria. 



In Chapter IV Mr. Teall deals with the petrology of the igneous 

 rocks associated with the stratified deposits of the Uplands. Few 

 of the points worked out during the revision of the Upland geology 

 has e?;ceeded in interest the establishment of the facts, that the 

 typical members of the disputed rock series of Ballantrae are true 

 contemporaneous lavas and tuffs, and that all the unquestionably 

 intrusive masses associated with them ai-e, with the exception of 

 a few Tertiary dykes, of pre-Bala age. The Ballantrae lavas belong 

 to a great basement volcanic series of Arenig age, which, underlying 

 the sheet of Radiolarian chert, has now been proved by the Survey to 

 actually show through at the surface, in numberless eroded anticlines, 

 over a collective area of some 1,600 square miles. The lavas are mainly 

 diabases or diabase porphyrites, and are characterized by a remarkable 

 ' pillow structure.' Mr. Teall points out with striking effect how 

 both the structure and the composition of these pillow lavas resemble 

 those of the volcanic rocks similarly associated with Eadiolarian 

 cherts in such widely separated localities as California, Cornwall, 

 and Saxony, while their accompanying agglomerates and tuffs call to 

 inind those of the disputed disti'icts of Rhobell Fawr and Predazzo ; 

 and further, in almost all these widely separated localities, British and 

 foreign, these peculiar volcanic rocks, both effusive and pyroclastic, 

 are pierced by plutonic or intrusive rocks which agree in certain 

 peculiarities of composition, texture, and structure, and in the 

 difficulties which they present to the geologist who endeavours to 

 determine their relative order of intrusion. ]t is possible that, as 

 in the parallel case of the igneous rocks with pillow structures, and 

 associated pyroclastic rocks, etc., of the Ordovician of Shropshire and 

 Radnor, we are dealing with a distinctly submarine igneous series, 

 probably extruded at great depths ia the sea. 



