Revieics — The Survey Memoir on the Scottish Uplands. 517 



sheet of Radiolarian chert, which varies hut little in thickness, 

 60 to 70 feet from end to end of the country. From beneath this 

 clieit sheet emerges in its turn a basement volcanic series of 

 pillow lavas, volcanic agglomerates and tuffs, with some Arenig 

 graptolites and other fossils in their highest zone, and answering 

 to the volcanic series of Ballantrae. 



This Northern maze thus satisfactorily unravelled, the Survey 

 officers enter next upon the description of the Girvan district, which 

 is supposed to commence at the great Glen App fault. They give 

 a most interesting account of the coast sections between Glen App 

 and Ballantrae, where at the localities of Portandea, Currarie, and 

 Dc^wnan, they have discovered the complete sequence from the 

 Arenig (Ballantrae) volcanic rocks, through Radiolarian chert bands 

 into mudstones and greywackes (Tappin's group), with a typical 

 Glenkiln fauna. 



The descriptions of these important localities are illustrated by 

 several beautiful photographs by Mr. R. Lunn, showing the ' pillow '- 

 form masses of lava, with their remarkable vesicular structures, and 

 with beds of limestone and Radiolarian chert filling spaces between 

 the pillow-like forms. 



The authors next plunge into the Ballantrae volcanic and plutonic 

 complex north of Stinchar Valley. Tiie key section of Bennane Head 

 is first described ; the relationship of its Arenig graptolitic (or 

 'retrngraptas) bands to the underlying volcanic series and the over- 

 lying Radiolarian chert bands is clearly demonstrated, and is illustrated 

 by capital sketch-maps, sections, and photographs. From this key 

 localitj'^ the beds of the complex are traced stage by stage along the 

 whole Ballantrae coastline to Kennedy's Pass. The authors show 

 how accurate were Professor Bonney's original conclusions as to 

 the volcanic nature of many of these strata, and the similarity of 

 the associated intrusives with those of the Lizard district. They 

 establish in a general way the sequence of the entire succession of 

 the Ballantrae rocks both along the shore and on land, as at 

 Cragneil and Craigbead, and they cite several new localities for 

 Arenig graptolites in the series older than the zone discovered by 

 myself at Bennane Head. Finally, they prove conclusively that 

 the whole of this Ballantrae complex (some of whose plutonic 

 members were erroneously regarded by myself and others to be of 

 post-Girvan age), volcanic and plutonic rocks alike, is of more 

 ancient date than the basal beds of the Barr series, which actually 

 rest upon the volcanic platform, and contain abundant rounded 

 pebbles or fragments of nearly all its known rock-varieties. 



In Chapters XXI and XXII we have a summary of the strati- 

 graphy and palaeontology of the overlying Girvan formations. The 

 classification and sequence of the various members of the succession, 

 as worked out in ray Girvan paper, are again shown to be the natural 

 ones, and the various graptolite zones which enable us to parallel the 

 bwds with the Moffat and Upland succession generally are carefully 

 checked and discussed. The Balclatiihie beds are. however, now 

 transferred to the base of the Ardmillau Series, and the Blair and 



