Vol. 55.] GEOLOGY OF KORTHERN ANGLESEY. 641 



Llandeilo forms. The latter species has hitherto been recorded, I 

 believe, only in the Scottish Llandeilo. 



Eor stratigraphical reasons it is almost certain that the pebbly 

 sandstones of Ogof Gynfor and the neighbourhood lie below the 

 Dicranograptus-heds, of Penterfyn. All the species, too, occur as 

 low as the Llandeilo, so that the Ogof Gynfor rocks may be looked 

 upon as Upper Llandeilo. The commonest form in these beds was 

 a small Ortliis {? Zygosjpira) with simple ribs, which I could not 

 identify. The Lingula is large, 1 inch long by nearly | inch wide. 

 Its frontal margin is missing, but in contour and size the specimen 

 agrees with L. tenuigranula.ta. The line granulation is not pre- 

 served, owing to the coarseness of the matrix. 



It was most interesting to find, associated with these fossils, 

 specimens of the so-called species Ortliis Carausii^ because the 

 occurrence of this form in the Ortlik-grit^ of Central Anglesey seems 

 to be the principal reason why Prof. Hughes assigns the basement 

 Palaeozoic rocks there to the period of the Tremadoc Slates, though 

 Dr. Callaway has questioned the correctness of this conclusion.^ It 

 only occurs occasionally at Ogof Gynfor, but one band in the Craig 

 Wen tramwav-beds was crowded with casts of the shell, and 

 single individuals are to be found in other parts of the tramway. 

 I have examined a large number of specimens and compared them 

 with Davidson's figures and descriptions, and (by the kindness of 

 Prof. Hughes) with specimens from Central Anglesey in the Wood- 

 wardian Museum. The concentric lines of growth are not so well 

 preserved, though this may be due to the nature of the matrix, 

 and a few examples very slightly exceed the dimensions given by 

 Davidson ; but the majority of the specimens are within those limits, 

 and all of them, in number of ribs, relative convexity of valves, 

 and internal characters, agree with Orthis Carausii. 0. Carausii 

 was separated as a distinct species from 0. calligramma as much on 

 account of its occurrence at a lower horizon (namely, Tremadoc and 

 Lower Arenig) as from its smaller dimensions.^ But we now see 

 that in J^orthern Anglesey the Qarausii-iorm. occurs with 0. testu- 

 dinaria, 0. vespertilio, etc. : that is to say, on the horizon of the 

 ordinary calligramma-form ; and the last-named is a very variable 

 shell, frequently blending into the Cara?/m-type. There does not 

 therefore seem sufficient justification for their separation into distinct 

 species, and I follow Salter's first identification of this brachiopod 

 as a dwarf variety of Orthis calligramma. 



It is evident that Orthis Bailyana is not pre-Cambrian, and we 

 may with confidence accept Davidson's statement that in Wexford 

 (the only other locality where this species occurs) it is found with 

 0. calligramma, 0. biforata, Leptcena sericea, and Favosites Jihrosus. 

 These are taken to indicate a Bala age in Wexford, but the Anglesey- 

 beds are more likely to be Upper Llandeilo. 



1 Quart. Journ. Geol. Soc. vol. xxxvi (1880) p. 238 & vol. xl (1884) p. 580. 



2 Davidson, Monogr. Pal. Soc. (1882-1884) 'Brit. Sil. Brach.' Suppl. p.. 183. 

 Q. J. G. S. No. 219. 2t 



