856 REPORT— 1898. 



Eastern Deyonshire. These rest in complete discordance on the flanks of the 

 Palaeozoic highlands, and must be regarded as forming the base of the Secondarv 

 rocks of that district. 



By the Geological Survey this series has hitherto been mapped as Trias, but in 

 the new ' Index-map ' they are coloured as Permian. There is no palseontological 

 evidence which would connect them with the fossiliferous Permians, usually re- 

 garded as of Palaeozoic age ; but it has been evident for some time past that opinion 

 was inclining to revert to the views of Murchisou and the older geologists, moie 

 especially as to the position of the breccias so largely charged with volcanic rocks. 

 The subject was dealt with by Sir A. Geikie in his address to the Geological 

 Society, where he speaks of some of these rocks as presenting the closest resem- 

 blance to those of the Permian basins of Ayrshire and Nithsdale.^ 



One difficulty which presented itself ti'i the Devonshire geologists in accepting 

 the Permian age of the ' red bods' was, that the whole of the lower Second ary 

 rocks appeared as an indivisible sequence, proved by its fossils to be of Keupe'r 

 age at one end, and therefore inferentially of Keuper age at the other. Dr. Irving, 

 however, considered that at the base of the Budleigh-Salterton pebble-bed there is^ 

 a physical break of as much significance as that between the Permian and Trias of 

 the Midlands. In the marls which underlie this pebble-bed he recognised a strong 

 resemblance to the Permian marls of Warwickshire and Nottinghamshire ; and 

 Prgfessor Hull, who had been studying the sections east of Exmouth about the 

 same time, ultimately acceded to this view.'-* Its acceptance by the Survey thus- 

 throws all the Exmouth beds into the Permian ; and that formation, according to 

 the new reading, has an outcrop of some 35 miles from the shores of the English 

 Channel to within 3 miles of Bridgewater Bay. The fertility of these red clays, 

 loams, and marls has long been recognised by agriculturists, and it is not im- 

 probable that the abundance of contemporaneous volcanic material may in some 

 measure have contributed to this result. 



In conformity with the new mapping, the Budleigh-Salterton pebble-bed and 

 its equivalents to the northwards are accepted as of Bunter age, and thus consti- 

 tute the base of the Trias in the south-west. Like most pebble-beds, they are 

 irregularly developed between the Permians and a strip of reddish sandstone 

 (coloured as Keuper), which runs up irom the mouth of the Otter to within a 

 short distance of Bridgewater Bay. The materials of the pebble-beds are not of 

 local origin, like so much of the breccia at the base of the Permian, ^ The general 

 resemblance, both as regards scenery and composition, to the Bunter conglomerate 

 of Cannock Chase has been pointed out by Professor Bonnev, who seems prepared 

 to endorse the recognition of the Budleigh-Salterton pebble-bed as a Bunter con- 

 glomerate. He was not impressed by any marked unconformity with the under- 

 lying series. To some extent we may accept this view, since, whatever maybe the 

 age of the Devonshire breccias and ' red beds,' they, in common with the Trias, 

 must have been deposited under fairly similar physical conditions in a sort of 

 Permo-Triassic lake basin. 



The bulk of the Trias, including the Dolomitic Conglomerate of the ]3ristol 

 district, is still regarded as of Keuper age, though it is now admitted, as insisted 

 on by Mr. Sanders years ago, that the Dolomitic Conglomerate does not neces- 

 sarily occupy the base of the Keuper, but is mainly a deposit of hill-talus, which 

 has been incorporated with the finer deposits of the old Ti-iassic lake as the several 

 Paleozoic islands gradually became submerged. The great blocks which fell from 

 the old cliSs were formerly regarded as proofs of glacial agency, and there are 

 persons who still believe, more especially with respect to the Permian breccias, 

 that such rocks are indicative of a glacial origin. 



' Quart. Journ. Geol. Soc. vol. xlviii. (1892), p. 161. 



^ Cy. Irving Quart. Journ. Geol. Sue. vols. xliv. (1888). p. 149, xlviii. (1892), p. 68, 

 and xlix. (1893), p. 79 ; and Hull, op. cit. vol xlviii. (1892), p. 60. 



» Northwards, e.ff. at Burlescombe, Wiveliscombe, &c., the equivalent conglome- 

 rates are largely, if not entirely, of local origin (cf. Ussher, Quart. Journ. Geol. Soc. 

 vols, xxxii. (1876), pp. 378, 382 ; and xxxiv. (1878), p. 461). Mr. H. B. Woodward 

 confirms this. Note added October 1898. 



1 



