part 2] THE YALENTIAN SERIES. 165 



Wills & Smith divide the Yalentian Series into Llandovery or 

 BirkhilHan and Tarannon. The Corwen Grit was formerly con- 

 sidered to be the base of the Llandoverj', and all the strata up to 

 the Denbighshire Grits (Wenlock) were referred to that formation. 

 At Grlju Ceiriog the fauna of the Glyn Grit was believed to indi- 

 cate closer affinities with the underh^ing Dolhir Beds (Upper Bala) 

 than with the Fron-Frys Slates. In the later description of the 

 area these oiits are assisfned to the Ordovician. 



Whatever may be their relations, there is no doubt that the suc- 

 ceeding Grey Slates of Corwen correspond generally to the Fron- 

 Frj^s Slates. In the former district they are associated with a band 

 of graptolitic shale, but contain in addition a sparse shelly fauna 

 in their lower layers. At Glyn Ceiriog the graptolitic band was 

 not found, though shell}' fossils are abundantly represented. 



The Pale Slates of the one district are similar in lithological 

 charactei- to the Ty-draw Slates of the other, and the two 

 formations are believed to be equivalent. 



The graptolitic shales have yielded a fauna which appears to 

 indicate a horizon in the upper part of the Monocfrapfus-friangii- 

 l((tus Zone or the succeeding Mesograj^fiis-magnns Zone; that is, 

 nearly the same hoi'izon as the shales above the Newlands and 

 Woodland Beds of Girvan. Wills & Smitli record, also, the zones 

 of Monograpfus cgplnis and M. convohifiis. An approximate 

 superior limit to the Grey Slates is thus given by reference to the 

 graptolitic scale. 



B}^ the kindness of Dr. Groom & Mr. Lake I was allowed to 

 examine their collection of fossils from the Fron-Frys Slates soon 

 after the publication of their paper, and was able to confirm in 

 essential particulars their determinations. The brachiopods from 

 the area collected by the officers of the Geological Survey were also 

 examined. The fauna is rather poor in species : the charactei'istic 

 forms are JBarrandella unclata, like the Haverfordwest form, 

 Atrgpa marginaJis, Meristina suhundata . early variants of Plect- 

 amhonites duplicafus, small species of Dalmanella, and a few 

 Strophomenids. This fauna can be matched very closely in the 

 JBarrandella-iDidafa or Cartlett Beds of Haverfordwest and the 

 lower part of the Gasworks Mudstones ; but the characteristic 

 fauna of the uj^pei* part of those mudstones does not appear to be 

 represented. It may be suggested that they, as well as the 

 succeeding Gasworks Sandstone, are equivalent to the graptolite- 

 shales of the Corwen area. 



In the Ty-draw Slates Groom & Lake only found JSLonograptiis 

 marri, but further work by the Geolo^-ical Survey has resulted in 

 the discovery of the four main zones of the Tarannon district. 

 The upper part of the Birkhill is not represented, its absence being* 

 attributed to a non-secpience. 



In the Welshpool district graptolitic horizons are also inter- 

 calated among sandstones and conglomerates with shelly fossils. 

 The relations between some of the groups attributed to the Yalen- 

 tian indicate an unconformity within the series. In some localities 



