356 Mr. J. W. Davis on the Teleostean 



slender tail, and extending along the ventral surface of the 

 body ; the spinous and interspinous bones of the body, and the 

 osseous rays of the fin ; the presence of a clavicle in the 

 shoulder-girdle ; and the fact of there being branchiostegal 

 rays with teeth attached supporting the gills. In the speci- 

 men described by Dr. Goldtuss the two anteriorly-pointing 

 conical cavities are considered, apparently with much reason, 

 to be the nasal cavities ; and if this be so, they are of decidedly 

 teleostean character, the nasal orifices of the Plagiostoinous 

 fishes being in all cases situated beneath the snout and in the 

 Chimera and Rays close to the angles of the mouth, the mouth 

 and nasal orifice in some instances being connected by a 

 groove. 



From a consideration of the peculiarities existing in the 

 fishes composing the genus Pleuracanthus we are led to the 

 conclusion that it claims a nearer relationship with the bony 

 fishes than with the Elasmobranchs, though there is equal 

 reason to believe that it possesses many characters in common 

 with the Sharks and Rays. We are therefore further led to 

 place the genus in an intermediate position between the two. 

 The distinguishing characters used in modern classification to 

 distinguish the Sharks and Ganoids, viz. the many-valved 

 muscularly-contracting bulbus arteriosus, the spiral valve of 

 the intestine, and the chiasma of the optic nerves, are of such 

 a perishable nature that they are not likely to be found in 

 fossil fishes. In the absence of these we are driven to select 

 such characters for classification as may be preserved, and to 

 collate them as best we may with those of existing forms. 

 Prof. Huxley, in the tenth decade of the Geological Survey, 

 has pointed out the seeming relationship between some of 

 the Ganoid fishes (so called) of the Old Red Sandstone and 

 the modern Siluroids. In the structure of the head of Coc- 

 costeus the general arrangement of the bony exoskeleton much 

 resembles that of the tropical fish Clarias ; whilst the peculiar 

 form of the mandibles, and the expansion of the bony elements 

 usually considered to be homologous with the coracoid and 

 radius of other fishes so as to form a large ventral shield, 

 offer many points of resemblance to the Siluroid Loricaria. 

 The Devonian Ptericlithys is also in several ways closely 

 related to the modern Siluroids ; its osseous envelope can 

 only be compared to the box-like cincture of the modern 

 Ostracion ; and the fossil fish Cephalaspis has also certain re- 

 semblances to Callichthys and Loricaria. Prof. Huxley 

 remarks, " At any rate, I think the prima facie case in 

 favour of the Teleostean nature of Coccosteus is so strong that 



