42 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. [Dec. 4, 



on the one hand, and DiJcelocephalus, Homalonotus, and Calymene on 

 the other. In its coniform glabella, with bent farrows and faceted 

 pleurae, it resembles Conocoryphe ; and in the position of its posterior 

 facial sutures, and in its broad and many-ribbed tail, it resembles 

 Dikelocephalus. The only other Trilobites are two species of Niobe, 

 a well-known Tremadoc genus. In these two genera we certainly 

 recognize a resemblance to far earlier types, and the possible pro- 

 genitors of many succeeding forms. The fauna is exceedingly rich 

 in the Lamellibranchiata, no less than twelve species, belonging to 

 five genera, having been found. This is the more interesting as no 

 Lamellibranchs had previously been discovered in so early a formation 

 in Britain — nor, indeed, so far as I am able to find out, in any other 

 country. The Echinoderms also are represented here by a beautiful 

 starfish of the genus Palasterina, and by an Encrinite of the genus 

 Dendrocrinus, and hence are shown to have existed at a very early 

 period. The Cephalopoda, which have not been found in earlier for- 

 mations, are also represented here by a species of Orthoceras. The 

 remainder of the fauna is made up of Heteropods, Pteropods, and 

 Brachiopods, groups which had representatives in still earlier rocks 

 (see Table). 



The conditions under which the Tremadoc rocks in the neighbour- 

 hood of St. David's were deposited seem to have been intermediate 

 between those of the shoal and shallow water of the Lingula-flag 

 period and those of the deep sea which must have prevailed when 

 the fine muddy deposits of the overlying Arenig slates were being 

 thrown down. This intermediate condition must have been par- 

 ticularly favourable to the existence of life, and was doubtless one 

 of the causes of the appearance at this time of such a varied and 

 important group of organisms. 



In comparing these rocks with the Tremadoc slates of North 

 Wales on stratigraphical grounds, a difficulty arises in the supposed 

 unconformity mentioned by Prof. Ramsay in his memoir on North 

 Wales as occurring there between them and the Lingula-flags — as 

 these two groups, wherever they are exposed in the neighbourhood 

 of St. David's, appear so closely connected, both on stratigraphical 

 and lithological grounds that it becomes quite impossible to know 

 exactly where the boundary-line between them, which of course 

 in such a case is only arbitrary, should be placed. But the pake- 

 ontological evidence goes to j)rove that they (the Tremadoc rocks) 

 are closely allied to, if not identical with the lower portion of the 

 Tremadoc rocks of North Wales*. 



Mr. Homfray, of Portmadoc, who has carefully studied the 



* The faunas, however, are in some respects considerably unlike ; but the 

 difference is doubtless to be attributed to the state of the sea at the time the 

 rocks were deposited in each locality. In North Wales the series is made up of 

 fine-grained slates, and hence of deep-sea deposits, and with a deep-water fauna ; 

 in South Wales, of rough-grained flags and sandstones, indicating much shallower 

 water, and, as shown by the presence of worm-tracks, numerous lamellibranchs 

 and brachiopods of large size, a comparatively shallow- water or mixed fauna. 

 The absence at St. David's, in the Upper Lingula-flags, of the fine black 

 slates which are so characteristic of the series in North Wales and at Malvern, 

 may also be accounted for by the prevalence of shallow water in the one locality. 



