54 



OLD RED SANDSTONE FISHES. 



largest in the large shields, as may be seen by comparing PI. XII, figs. 2 and 6, and 

 PI. VIII, fig. 4, and their less frequency, distinguish the ornament of this species from 

 that of Eucephalaspis Agassizii, specimens of nearly equal size being compared. The 

 thickness of the substance of the shield is relatively great, and it has a bright polished 

 surface. The cornua are long, subcylindrical, and solid, differing in this, as does the 

 shield-substance in thickness, from Eucephalaspis Agassizii ; the tubercles on the surface 

 are oblong and closely packed. The ends of these cornua are not unfrequently found 

 detached, having survived the destruction of the rest of the shield, from which they appear 

 to have been readily broken. The postorbital valley is shorter and broader relatively in 

 this species than in other Cephalaspids. Sir Philip Egerton attached importance to the 

 great breadth between the eyes, but the size of the individual and variations in pressure 

 are liable to affect this character. A greater interspace between the orbits is, no doubt, 

 present than between those of Ea. Agassizii, if orbits of equal size or nearly equal size 

 be compared. But the real point is that the orbits are mucli smaller in this species, 

 relatively to the size of the whole shield, than in Eucephalaspis. 



General Remarks. — After some hesitation I have decided to associate C. Salwegi and 

 C. asterolepis as one species, not being able, on careful examination, to find any character 

 which should separate the large specimen described by Dr. Harley from Sir Philip Egerton's 

 original C. Salwegi. A magnificent specimen has been kindly lent to me by Mr. Lee, of 

 Caerleon, from which the outline woodcut has been in great measure drawn ; two cornua, 

 belonging to Dr. MacCullough, and found in the same quarry at Abergavenny, furnishing 

 the evidence of these parts which are wanting in Mr. Lee's specimen. This specimen 

 came to hand after my plates had been completed ; it has been photographed and pub- 

 lished in a recent volume of the * Proceedings of the Woolhope Nat. Eield Club.' Another 

 specimen, which is better than either of those drawn in my plate, has been recently obtained 

 for the British Museum. It is very similar to Mr. Salwey's specimen drawn in PI. XII, 

 fig. G, but it has the long cylindrical cornua attached to the angles of the shield, which 

 are Avanting in this specimen. The ornament is best shown on these two shields, the 

 irregular ' splashed ' character of the tubercular markings being well seen. In parts 

 they assume a much more regular character, each tubercle being hemispherical, and all 

 nearly of equal size (PI. VIII, fig. 4). Bits showing this character of marking from shields 

 of twice the size of that drawn in PI. XII, fig. 6, and therefore with much larger tubercles, 

 are not uncommon, and have been regarded distinctively as C. asterolepis, but I cannot 

 believe that the difference is anything but one of size. None of the large specimens 

 usually referred to C. asterolepis show the tubercles of the surface of the shield in such a 

 way that tliey can be compared with those of the smaller specimen drawn in fig. 6, 

 PI. XII. On such specimens a few of the larger tubercles are left here and there 

 adhering. I do not doubt that if we have had the ornament properly shown, we should 

 find oval, circular, and irregular tubercles of various sizes on parts of the shield, as in 

 typical C. Saliuegi. I say typical C. Salwegi, for though PI. XII, fig. 5, is the figured type 



