46 



OLD RED SANDSTONE FISHES. 



pectoral fin has quite a different structure, which is well seen in fig. 2, PI. XI. 

 There is no evidence of a separation of the calcareous matter into regular scales, but 

 crescent-like areae appear to be marked out by the greater thickness of this matter in 

 some parts than in others; neither can any fin-rays be traced in these very curious 

 organs. 



2. Eucephalaspis Agassizii. PI. IX, figs. 2, 3, 6. 



Cephalaspis Lyellii, Agassiz (in part). Pois. foss., vol. ii, pi. 1 b, fig. 3, 1835. 



Name. — Named after Prof. Louis Agassiz, the author of the ' Poissons fossiles.' 



StratigrapMcal Position. — The Cornstones of Herefordshire and Worcestershire. 



Characters. — The woodcut, fig. 11, gives what is, I believe, an accurate indication 

 of the form of the shield in this species. So many specimens of this are found that 

 it is possible to get a more trustworthy idea of the form of the shield when uncrushed, 

 and in all its parts, than it is in the case of any other Cephalaspis. Numerous as these 

 specimens are, only one has come to hand showing clearly the form of the orbital region 

 and the surface-markings. This has been used in drawing these parts in the woodcut. 



Fig. 18 a. 



Fig. 18 b. 



Ornament of Eucephalaspis Agassizii, magnified 7 diameters. 

 a, from the orbital region. b, from the posterior angle. 



fig. 11, and was lent among many others by Dr. Grindrod, of Malvern. It is unnecessary 

 to attempt any concise definition of the species in words when the woodcut figures express 

 so much more clearly what is meant than a compressed sentence can be made to do. 

 The relative position of the orbits in this species and in C. Lyellii may be seen on com- 

 paring the figures in PI. VIII and PI. IX. 



General Remarks. — ^The greatest variation in form is exhibited by the abundant head- 

 shields of this species which have been exhumed in the West of England — a variation 

 due simply to pressure. Some shields are quite flat, and yet tolerably perfect ; the 

 outline in this case becomes very much widened, and the cornua are directed quite away 

 from one another, instead of running backwards in nearly parallel directions ; others are 

 compressed laterally, but many are to be obtained which are regularly convex, and by 

 their symmetry lead one to believe that they have not suffered any distortion. No speci- 

 mens show the form and proportion of head-shield which is attributed to the following 



