536 Ff, R. Cowper Reed—Silurian, Haverfordwest. 
A large collection of fossils has been made by Mr. V. M. Turnbull 
from this particular locality during the last. few years and presented 
to the Sedgwick Museum. From a prolonged study of this material 
and comparison with the typical fauna of the Slade Beds as described 
by Messrs. Marr & Roberts, the conclusion has been forced upon me 
that we have here to deal with a higher paleontological horizon than 
the true Upper Bala of this district, and that there is much more 
reason for putting it in the Silurian than for keeping it in the Ordovician. 
The stratigraphical evidence which has subsequently become available 
is in support of this view. A list of the fossils from this bed 
may be useful at the present moment, particularly as the Sedgwick 
Museum possesses by far the most extensive series which has been 
collected from it (see opposite page, 537). 
The rock in which they occur is a greenish-grey, fine-grained, 
argillaceous mudstone with occasionally minute specks of mica and 
a little arenaceous material, but usually of a very uniform smooth 
texture. It breaks in an irregular, often subconchoidal manner, and 
is devoid of any regular or well-marked planes of division. Exposed 
surfaces and the fossils themselves are generally more or less iron- 
stained. In appearance and general characters it is difficult or 
impossible to distinguish the rock from the higher beds of the Slade 
Series, but the fauna is completely different. 
The mudstone lies above a series of unfossiliferous(?) black shales, 
and intercalated in the latter is found a persistent bed of sandstone 
or grit. This whole series of beds may be conveniently termed the 
“‘St. Martin’s Beds,” as a local designation seems necessary; and the 
fossiliferous mudstone in it may bear the name of the “St. Martin’s 
Mudstone.”’ 
The true Ordovician occurs near Portfield House as a more 
micaceous mudstone lying fifty yards below the band of sandstone, 
according to Mr. Turnbull, and yielding Zrinucleus seticornis, Orthis 
biforata, Leptena sericea, etc. But until the whole succession has been 
worked out we may defer attempts at correlation, and a detailed 
comparison of the St. Martin’s and associated beds with those of other 
areas is at present undesirable. 
The occurrence of Phacops mucronatus, Strophomena siluriana, 
Meristella cf. crassa, and Diplo. cf. modestus are sufficient in themselves 
to show that we have here to deal, not with the highest Ashgillian, 
but with the base of the Silurian. The strongly marked paleonto- 
logical break between the St. Martin’s Mudstone and the Slade Beds 
proper seems abundantly to justify the drawing of the Silurian 
line below and not above them. All the typical Slade species are 
| absent; no Zrinucleus occurs, and the higher affinities of the fauna are 
indicated by the occurrence of certain species ( Cornulites scalariformis, 
Goniophora ct. eymbeformis, Meristella cf. crassa, Atrypa ci. flexuosa, 
Orthis cf. Edgelliana, Scenidium Lewist, Crania Grayi var., Lingula 
Symondst, Diplo. cf. modestus) which are not known from undoubted 
Ordovician beds. The Llandovery Beds of the Gas Works have, on 
the other hand, a completely distinct fauna. 
With the exception of the Brachiopods (especially the genus Orthis) 
the fauna of the St. Martin’s Mudstone is poor; the Gasteropods, 
