882 DR RAMSAY H. TRAQUAIR ON FOSSIL FISHES COLLECTED BY THE 



scales of the tail and body behind the head is admirably shown in a specimen lent me 

 by Mr Macnair, and from which the drawings of magnified scales given in Plate I. 

 figs. 7 and 8 were taken. 



Lanarkia horrida, Traquair. 



A specimen of Lanarkia horrida from Birkenhead Burn (Geological Survey Collec- 

 tion), represented, natural size, in Plate II. fig 1, distinctly displays, even when dry, two 

 small round spots in the same relative position as those in the two examples of 

 Thelodus Scoticus described above, and in like manner, when the fossil is wetted, stand 

 out with all the greater distinctness. On examination with a powerful lens or with 

 the compound microscope (one inch objective), nothing more can be made out than the 

 appearance of a dark bituminous-looking stain at the spots concerned. The white dot 

 seen in the centre of the left dark spot in fig. 1 is due to the carbonate of lime filling 

 up the pulp cavity of a dermal spinelet which has become broken across at this place. 



That the above-described dark-stained spots represent the position of the eyes in 

 Thelodus and Lanarkia will, I think, scarcely be questioned, and indeed the pheno- 

 menon does not stand alone in the field of fossil ichthyology. In the case of the little 

 Cyprinodont fish Prolebias cephalotes (Ag.), which occurs in shoals in the Lower 

 Oligocene of Aix in France, the pigment of the eye, as well as of the abdominal cavity, 

 is preserved ; and on a slab of fissile grey shale from China, and probably of Jurassic 

 age, now before me, numerous specimens of a small Lycoptera show a conspicuous 

 round black spot occupying the well-defined interior of the orbit. That no definite 

 orbits are observable in the case of the Coelolepidse is inseparable from the non-coherent 

 nature of the elements of the exoskeleton. 



It is, of course, interesting to find that these eye -spots in the Coelolepidse occupy 

 exactly the same relative position on the head as the orbits, or supposed orbits, of 

 Pteraspis and Drepanaspis, seeing that, if my view of the interaffinities of these 

 creatures is correct, such a position is the one we would naturally expect. But 

 upholders of the arthropod idea of vertebrate derivation may also point out that the 

 lateral eyes of certain Eurypterids (Pterygotus, Slimonia) also occupy the same 

 situation in the head. Other members of the same group (Eurypterus, Stylonurus) 

 have their lateral eyes on the upper surface of the cephalic shield towards the middle, 

 as the Cephalaspidse, — in either case I fear we have to do with what the Germans call 

 a " Convergenz-Erscheinung." 



? Lanarkia, sp. 



In Plate I. fig. 2 is represented the anterior part of a fish apparently belonging to 

 the Coelolepidse, and here provisionally referred to Lanarkia, seeing that, so far as the 

 lens can discover, the exoskeleton seems to consist of minute, pointed, hollow spinelets. 

 What exists of the counterpart is shown in fig. 3. 



