part 2] THE TALENTIAIf SEEIES. 151 



then available for the correlation of the graptolitic facies of Birkhill 

 and Grala with the divisions of the older classification. After 

 summarizing the evidence, he remarks (^op. cit. vol. v, p. 47) that 

 three distinct groups of strata separated by fairly marked uncon- 

 formities are found in man}^ parts of Wales between the summit 

 of the Bala and the base of the Wenlock Shale : — 



' It has g-enerally been the habit to call the first of these subformations by 

 the title of the Lower Llandovery, the second May Hill or Upper Llandovery, 

 and the third the Taraimon shale. The first is usually believed to be most 

 intimately allied in its palseontological characters to the Bala formation,, 

 and the last to be hardly separable from the Wenlock shale. My own re- 

 searches impel me to the conclusion that these three subformations are far 

 more closely allied to each other than they are to the beds above or below^ 

 and that they should be considered as the three consecutive members of a 

 single formation. In the South of Scotland (Valentia) these three sub- 

 formations are recognizable, superposed in conformable sequence with clear 

 relations to the Bala below and to the Wenlock above, and unitedly covering 

 an area of several thousands of square miles. Until geologists are willing to. 

 include the Tarannon in the Llandovery, it will therefore be best to speak of 

 this great Scottish formation and its equivalents as the Valentian forma- 

 tion, its three divisions (Lower, Middle, and Upper) representing respectively 

 the Lower Llandovery, Upper Llandovery, and Tarannon of Wales and Siluria.'' 



It will be observed that Lapworth apparently implies the corre- 

 lation of the Lower Birkhill, Upper Birkhill, and Gala (including 

 Hawick) subdivisions which he had previousl}^ established in the 

 South of Scotland with Lower Llandovery, Upper Llandoverj^, and 

 Tarannon respectivel}^. If this suggestion is examined it w411 be 

 found, however, that the meaning of both Upper Llandovery 

 and Tarannon varies accordms^ to the facies of the rocks which 

 is in question. It is doubtful, moreover, whether such a cor-^ 

 relation was intended by Lapworth, for, on another page {oj). cit. 

 vol. V, p. 364), he writes : 



' In the entire series we seem at present to recognize five subgroups of 

 tolerably equal systematic importance — the (1) Lower, (2) Middle, and (3) 

 Upper Birkhill Shales, and (4), (5) the Gala and Grieston groups. Of these, 

 Nos. 1, 2, and 3 are possibly included in the so-called Lower Llandovery of 

 South Wales, while the fourth and fifth correspond to the Upper Llandovery 

 and Tarannon ' ; 



and on p. 365 he uses Gala and Tarannon as if they were 

 synonymous terms. The correlation of the Birkhill wath the 

 Lower Llandoverj' is in accordance with his conclusions in the paper 

 on the Moffat Series (see above); while he had previously demon- 

 strated 1 that the Gala rocks pass up conformably into the Kiccartoui 

 Beds — the representatives of the Wenlock in the South of Scotland. 



Most of the confusion which attaches to the classification of the 

 Yalentian rocks of different areas is due to the ambiguous relation 

 one to the other of the Tarannon and Upper Llandoverj^ groups.. 

 From their lithological characters and their position immediately 

 underneath the Wenlock Shale, the pale shales ot the neighbourhood 

 of Llandovery and of Shropshire w^ere believed by Aveline to be 

 the same rocks as those at Tarannon. Aveline traced these 



1 Geol. Mag. 1876, p. 550. 



