1866.] YOUNG — PLATYSOMrS. 305 



are separated for great part of their length by the frontals, the remain- 

 ing space being occupied by the ethmoid, a small bone (Mh.) which 

 separates the upper ends of the nasals {Na.), two narrow elongate 

 bones, in contact for great part of their length, and received into 

 notches of the ascending premaxillary processes. The upper end of 

 each nasal is dilated into a process separated by a notch from the A. 

 Fr., and forming the inner wall of the nasal aperture, whose outer 

 boundary was either a suborbital bone, or was formed by the vertical 

 portion of the maxilla. The narrow premaxillaries {Pma\) give oif 

 ascending plates of considerable relative breadth and height, receiving 

 superiorly the lower end of the nasal bones. The maxillary over- 

 laps the upper mandibular margin behind, and rises above to a 

 right-angled triangle, whose apex is at the premaxilla. The spatu- 

 late mandible (Mn.) is shorter and stouter than that of Palceoniscus, 

 which it much resembles. The teeth are perfectly smooth. The 

 suspensorium forms a solid bar, never shares in the dislocations 

 which have made clear the other head-bones. It consists seemingly 

 of a hyomandibular {Hy.), preopercular {Pr.O.), and quadrate {Qu.). 

 If a symplectic were distinct, its structure would coincide with that 

 of living Teleostei. At a point corresponding to the upper edge of 

 the quadrate, traces of the hyoid articulation are frequently seen. 

 Along the anterior margin of this bar two triangular plates are 

 applied, the lower is evidently the slightly expanded quadi^ate ; the 

 upper, part of the palato-pterygoid arch. It is overlapped and 

 apparently in close union with the maxillary vertical plate, by which 

 the other members of the arch are concealed, but above whose mar- 

 gin a long irregular plate appears, which, coming as it does into 

 relation with the anterior frontal, is doubtless the edentulous palate- 

 bone {PL). The operculum {Ojj.), shorter but as broad as the 

 suboperculum (S.Op.), is attached to the hyomandibular, and the 

 latter to the posterior frontal, thus proving the squamosal nature of 

 that bone. No remains are preserved of the inferior cranial bones, 

 which were probably cartilaginous. A bar of bone answers in posi- 

 tion to the basisphenoid, but is very imperfect, and wants any 

 sphenoidal expansions. The vomer is rarely seen, but, in a specimen 

 of PI. gibhosus (Brit. Mus. 28279.), is provided with a fine brush of 

 teeth. As many as four similar tufts occur in the Staffordshire 

 specimens, so as to suggest their pharyngeal position, but the bran- 

 chial arches are invariably wanting. The numerous occipital plates 

 of Lepidostevs are not represented unless by some large seale-Hke 

 plates, which are probably nuchal, not cranial. The small suprasca- 

 pular (S.Sc.) is distinct. 



The rest of the body presents no points in addition to those already 

 noted by Agassiz, save that there is no evidence for the existence of 

 a ventral fin. Indeed, perfect specimens demonstrate its absence, and 

 thus furnish a valuable diagnosis from the Pycnodonts which have it. 



The range of the genus is from the Coal-measures to the top of the 

 Permian. PI. striatus is the only species common to both periods 

 and to England and Germany, a specimen from Derbyshire being in 

 the Jermyn Street Collection. 



