522 THE CARB0NIFEK0T7S SUCCESSION IN GOWER. [NOV. 19II. 



In one respect, however, the radiolarian cherts differ from all the 

 Modiola phases of Gower, — they are devoid of included fragments 

 of contemporaneous sediments. 



On the other hand, further confirmatory evidence is yielded by 

 the black Pendleside or Posidonomy •a Limestones of the Midlands 

 and elsewhere. These rocks include the horizon of the Gower 

 cherts and appear, from their peculiar characters, to have been 

 deposited in lagoon-areas 1 ; their occurrence over a wide extent 

 of country testifies to a wide prevalence of lagoon-conditions on 

 that horizon. Some of them contain an abundance of bodies which, 

 Mr. Howe suggests with much hesitation, may be radiolaria (op. cit. 

 p. 400) ; and in North Devon similar limestones, there known as the 

 ' Venn Limestones,'' are associated with radiolarian cherts which, 

 as we shall presently see, probably include the horizon of the 

 radiolarian cherts of Gower. 



The Culm radiolarian cherts also a lagoon -phase. — 

 "We may now tabulate (Table IV, p. 523) the features of the radio- 

 larian phase, including some which have been observed in the Devon 

 radiolarian cherts, for comparison with those of the Gower Modiola 

 phases. The radiolarian nature of the Lower Culm cherts (the 

 Codden-Hill Beds) of Devon was long ago pointed out by Dr. Hinde 

 & Mr. Fox,- who, however, for the reasons mentioned above, 

 regarded them as deep-sea deposits, fhe associated Venn Lime- 

 stones being considered as but somewhat shallower in origin. 3 

 Dr. AVheelton Hind has more recently shown 4 that these lime- 

 stones, together with part, at least, of the cherfs, are correlative 

 with the Pendleside Series. The correlation of the radiolarian cherts 

 and overlying beds of Gower with the same Series tends to confirm 

 the view, suggested by De la Beche and strengthened by recent 

 work of the Geological Survey, 5 that the radiolarian cherts of the 

 two areas occur at one horizon. 6 (It is probable, of course, that 

 the Devon cherts, which are much thicker than those of Gower, 

 include other horizons also.) The close similarity 7 of the cherts of 

 Devon and Gower is such strong presumptive evidence of community 

 of origin, that it is difficult to believe that those of the one area 

 have been formed in comparatively deep water and those of the other 

 in shallows. And my own observations (see Table IV, p. 523) in 



1 With this view Mr. J. A. Howe, who first adequately described the rocks, 

 Q. J. G-. S. vol. lvii (1901) p. 399, tells me that he agrees. 



2 Q. J. G. S. toI. li (1895) p. 609. 



3 Of the shallow-water origin of the Yenn Limestones there can, however, 

 be no doubt. 



4 Geol. Mag. dec. 5, vol. i (1904) p. 392. 



5 See A. Strahan, Swansea Memoir, pp. 22-25. 



e It may be useful here to mention that the horizon in Gower with which 

 the Codden-Hill Beds are homotaxial is that of the radiolarian cherts, 

 not of the underlying beds with triiobites and brachiopods referred to by 

 Dr. Wheelton Hind (Geol. Mag. 1904, p. 402) ; the latter belong to D 2 . 3 . 



7 The similarity has been remarked by others, including Dr. G. J. Hinde 

 (Swansea Memoir, p. 25), and extends to the fine lamination and minute wedge- 

 bedding that I have adduced as evidence against a deep-sea origin for the 

 Gower cherts. 



