58 Br. R. H. Traquair — A New Genus of Coccosteidce. 



middle of that bone acutely backwards and inwards, ending on or 

 near the centre of ossification of the central plate. Just at the 

 point where the backward turn commences, a short branch is given 

 off which ends in the post-orbital angle or prominence. 



Mr. Whiteaves represents the main groove as again continued for- 

 wards at an acute angle so as to end at the front of the shield near 

 the ante-orbital prominence. Judging from analog}' with Coccosteus, 

 one might expect it to do so, but this continuation is not exhibited 

 in any of the specimens which I have examined. 



Associated in the same deposit with the cranial shields are found 

 various other isolated plates, which from their sculpture probably 

 belong to the same fish. Of these the only one which seems to be 

 clearly identifiable is the median dorsal plate (Whiteaves, op. cit. 

 pi. ix. fig. 2). The plate which he has figured as "left pre-ventro- 

 lateral " (ib. fig. 3), if it is so, must belong to the right side of. the 

 body, but his " ventro-median (?) " cannot be referable to a median 

 position as it is unsymmetrical. The Edinburgh Museum possesses 

 a number of such detached plates of different forms, but I am cer- 

 tainly not prepared to speculate at present as to their position on 

 the body cuirass. One thing is at least evident, namely, that if 

 those plates really belong to Phlyctcenius, their difference of form 

 from those of Coccosteus certainly gives additional emphasis to 

 the distinctness of the genus. No trace of the maxilla? or mandibles 

 of Ph. Acadicus has, so far as I am aware, been yet discovered. 



Phlyctcenius Anglicus, sp. nov. 



A good many years ago a small lot of fossils from Herefordshire 

 was purchased from a London dealer for the Edinburgh Museum, 

 and among them I found a small cranial shield, which, being 

 obviously referable neither to Cephalaspis, nor to Pteraspis, nor to 

 Scaphaspis, was rather puzzling in its appearance. Being, however, 

 at the time specially engaged with other subjects, this shield, from 

 Cradley, lay rather neglected, till one day I bethought me of it when 

 examining our collection of fish remains from the Devonian rocks of 

 Canada, and I was then greatly interested to find that the English 

 fossil was in reality a Coccostean, and a Coccostean not of the type 

 of Coccosteus decipiens, but of .Phlyctcenius Acadicus. This is of 

 special geological interest, seeing that both in England and Canada 

 this type is associated in the same beds with Cephalaspis, whereas 

 not a trace of any Cephalaspidean has ever occurred in those northern 

 Old Ked Sandstone deposits (Orkney, Caithness, and Moray Frith) 

 in which the typical Coccosteus is abundant. Nor does Coccosteus 

 occur in Forfarshire, where Cephalaspis is characteristic. 



At the time I made this discovery no one seemed to know of the 

 existence of a Coccostean in the Cradley beds, though indeed a piece 

 of the. shield of this very species is figured by Lankester in his 

 Monograph of the CephalaspidaB (pi. viii. fig. 4) as a " fragment of 

 doubtful character" in connection with Zenaspis Salweyi. 1 How- 



1 Mr. 'Wm. Davies seems to have believed in the occurrence of " Placodermi " in 

 the Herefordshire beds, as he labelled some fragments in the British Museum 

 " Pterichthys.' n They do not, however, belong to that genus. 



