Vol. 56.] SILURIAN SEQUENCE OF RHAYADER. 127 



developed in the Lower Birkhill Shales; and we may safely assume 

 that the deposits of the two formations belong to the same 

 geological period. One interesting fact, however, should be re- 

 membered: a thickness of over 1800 feet in the Rhayader district 

 is equivalent to only 52 feet in the Moffat area ; but every zone 

 in this 1800 feet has its proper place in that narrow band of the 

 Southern Uplands of Scotland. 



Turning to the Cab an Group, we find that 12 out of the 

 16 different forms contained in the group have been identified 

 from the Upper Birkhill Shales, while only 4 occur in the under- 

 lying beds. Of these 4, Monograptus lobiferus and Climacograptvs 

 normalis are common to both the Gwastaden and the Caban 

 rocks. M. spinigerus and M. Clingani are of restricted range, and 

 have never been detected out of the Upper Birkhill beds. There 

 can be little doubt that the M.SedgwicJcii Grits are equivalent 

 to part, at least, of the ill.- spinigerus zone of Dobb's Linn. The 

 D.-cometa bands of Moffat would then probably find their repre- 

 sentatives in the lower division of the Caban Conglomerates. 

 There is not, as yet, unfortunately, sufficient fossil evidence from 

 the Gafallt Shales to justify us in placing them in either one of the 

 M.-spinigerus or Rastrites-maooimus zones. We must, therefore, 

 leave them as occupying some position between the two. 



A comparison of the Rhayader Pale Shales with the Gala 

 Group yields remarkable results. Of 19 graptolite species contained 

 in the former beds, 16 occur in the latter, and 11 in the Upper 

 Birkhill. If we throw in the R.-maximus zone with the Gala 

 beds, no less than 18 out of the 19 species are identical in the two 

 groups, while the 11 in the Upper Birkhill are reduced to 5. The 

 forms yielded by these pale shales at the Tannery and the exposures 

 west of Rhayader doubtless belong to the R.-maximus beds of Moffat. 

 Such species as M. crassus and M. turriculatus are more characteristic 

 of the highest zones of the Upper Birkhill than of the lowest beds 

 of the true Gala Group. On the other hand, M.exiguus, M.priodon, 

 M. galaensis, Retiolites obesus — common to the eastern area — are true 

 Gala forms, and are seldom, if ever, detected at lower horizons. It 

 seems only right, therefore, to consider the Rhayader Pale Shales 

 as the representatives of the Gala and part of the Upper Birkhill 

 Groups. AVe are thus confronted with the old difficulty of separating 

 the Upper Llandovery from the Tarannon Shales, and one which 

 may prove troublesome in future research among these Valentian 

 rocks, unless some definite line is to be drawn dividing the two. If 

 the zonal method of mapping is to be followed, no complications 

 should arise. If, on the other hand, a hard-and-fast line must be 

 drawn, then it should be a pakeontological, and not a penological 

 line. The division that naturally suggests itself is one that has 

 already been followed out in the mapping of these rocks at Conway. 1 

 In that locality the Upper Llandovery and the Tarannon Shales were 



1 Quart. Journ. Geol. Soc. vol. lii (1896) p. 273. 



