IN THE OLD RED SANDSTONE OF OBAN. 593 



Fig. 3 in Plate presents us with a specimen in which not only the head but 

 also a portion of the body is seen, though in both cases only in impression. The 

 contour of the shield is well exhibited, and its proportions agree in all essentials with 

 those of the specimens described above. The impressions of the inner surface of the 

 lateral body scutes show a sort of corrugated appearance, which does not, however, in 

 my opinion, indicate that these inner surfaces were not smooth as usual. 



In fig. 4 of the same Plate we have an entire specimen lying on its side, but, unfor- 

 tunately, not very well preserved. Only a portion of the impression of the cephalic 

 shield, not including the orbits, is here seen, but it shows the peculiar short cornu (c. ) on 

 left side. We have also in this specimen, on the left side, the impression of the peculiar 

 obtuse flap-like organ (c.o.) which passes back from the head between the cornu and the 

 posterior angle of the shield in all the species of this genus where the specimens are 

 sufficiently entire to show it. Of the body scutes, only the inner surfaces of some, and 

 the impressions of the outer surfaces of others, are seen, whence it appears that they 

 were smooth inside and minutely granulated outside. Some remains of the dorsal fin 

 are exhibited opposite the point marked d. in the figure, and as regards the caudal it is 

 present but not expanded. 



An imperfect cranial shield, not figured, must, judging from its proportions, have 

 been originally 4 inches in length, and its dimensions must have consequently exceeded 

 those of any other British species except C. Salweyi, Eg., and C. magnified, Traq. 

 This specimen, consisting only of part of the right side of a shield, is interesting as 

 showing the tesselation of the middle layer, and also the impression of that strange 

 marginal strap-like internal space which lies close to and parallel with the outer rim, 

 beginning close to the cornu behind and extending nearly to the middle line in front. 

 This space has long been known in Eukeraspis pustuliferas* in which it is subdivided 

 into six loculi ; but I had not observed it in Cephalaspis until some time ago Mr Smith 

 Woodward pointed it out to me in specimens of C. Murchisoni in the British Museum. 

 Geological Position and Locality. — All the specimens as yet known were collected 

 by the Geological Survey of Scotland in the Lower Old Red Sandstone of the neighbour- 

 hood of Oban. One specimen has also turned up in the island of Kerrera. 



EXPLANATION OF PLATE. 



Fig. 1. Detached cranial shield of Cephalaspis Lornensis, Traq., from a cliff on the roadside near Carrick 

 Villa, Ohan. 



Fig. 2. A similar shield, much compressed, from the railway cutting near Dalintart, ahout one mile from 

 Oban. 



Fig. 3. A specimen, showing the head and a portion of the body, from the same locality as that repre- 

 sented in fig. 1. 



Fig. 4. A specimen, nearly entire, lying on its side, showing the cornu (c.) ; cornual flap (c.o.) ; and re- 

 mains of the dorsal fin (d.). From the shore 100 yards west of Gallanach Lodge, 3 miles south-west of Oban. 



* Lankester, Brit. Cephalaspideans, Pal. Soc, 1870, p. 58 ; Smith Woodward, Gat. Foss. Fishes Brit. Mus., pt. 

 ii. p. 193. 



