1889-90.] Dr Traquair on Fossil Dipnoi and G-anoidei. 397 
As in the Lower Permian genus Rliabdolepis of Troschel, this plate 
extends back the whole way to the opercular margin, and com- 
pletely separates the operculum from the quadrate plate beneath. 
I considered it as the true “suboperculum,” and reckoned the 
quadrate plate as “interoperculum,” an opinion which I subse- 
quently recanted in my paper on Chondrosteus * in which I restored 
the term “ suboperculum ” to the quadrate plate, and considered the 
small intermediate one as accessory. I had not, at the time I 
instituted the genus Cosmoptychius, observed this little triangular 
accessory plate in Elonichthys , but since then I have seen it 
frequently, and in consequence I feel that there is no sufficient 
ground for separating striatus from the other species of the last 
named genus. 
Rhadinichthys ornatissimus , Ag. — I have long had no doubt as 
to my former R. lepturus from Burntisland being only a young 
specimen of R. ornatissimus , and therefore cancel the species. Ex- 
ceptionally fine specimens of R. ornatissimus have recently occurred 
in the roof of the Dunnet shale at Straiton and Pentland. 
Rhadinichthys carinatus , Ag. — I much regret that a similar fate 
must befall R. Geikiei from Redhall; but the accession of a fine 
series of carinatus from Pumpherston has so much added to our 
knowledge of its characters, that it has become clear that the little 
specimens from the former locality, which I named in honour of 
the Director-general of the Geological Survey, must be absorbed in 
the Agassizian species. In Agassiz’s type of this species from 
Wardie, as well as in other specimens from that locality, the outer 
surface of the scales is imperfectly shown, hence I formerly 
described it as entirely smooth or nearly so ; but the Pumpherston 
specimens, which I cannot avoid referring to carinatus , from the 
general appearance and contour of the body and fins, clearly display 
the same character of scale ornament as I had previously attributed 
to Geikiei ( Proc . Roy. Soc. Edin ., ix. 1877, p. 439). I also, in my 
paper on the Eskdale Ganoids , doubtfully identified the common 
Rhadinichthys of the Glencartholm beds with the supposed species, 
R. Geikiei , and it has been long evident to me that this doubt was 
amply justified. Though closely allied, the Glencartholm fish has 
a considerably coarser serration of the posterior margins of the 
* Geol. Mag ., June 1887, p. 253. 
