10 



BRITISH FOSSILS. 



ptycliius. In tlie opercuiar apparatus, the operculum and sub- 

 operculum are large, subquadrate, and nearly equal in size. There 

 are large dendrodont teeth (very well shown in a large specimen in 

 Sir P. Egerton's collection) upon the inner side of the mandible. 

 The principal jugular plates are large, but no specimen I have seen 

 gives clear evidence of others. There is a well marked lateral line. 



Apart from what has been done by Agassiz, Miller, and Pander, 

 I think I can venture to assert from my own investigations that 

 the woodcut fig. 7 gives an essentially faithful restoration of Glyp- 

 tolepis.^ But a comparison of this figure with that of Holoptycliius, 

 given above, is sufiicient to prove the close affinity of the two 

 genera, — in fact, their family relationship. 



Pausing now, to look back over the ground which has been 

 traversed, we find that the six genera which have been discussed, 

 viz., Glyptolcemus, Glyptopomiis, Gyro2^tychius, Holoptychius, 

 Platygnathus, and Glyptolepis, possess the following characters in 

 common : — Two dorsals, acutely lobate paired fins (ventrals of 

 Glyptolepis ?) ; principal and lateral jugular plates, and no bran- 

 chiostegal rays ; more or fewer large teeth with grooved bases, and 

 consequently folded dentine ; sculptured scales and cranial bones, — 

 among Vi^hich last are to be noted three occipital plates, — large, 

 distinct, parietals, and equally distinct frontals. In short, they con- 

 stitute a family of Ganoids, which I propose to call Glyptodipterini, 

 and which may again be subdivided into two groups, or sub- 

 :ftimilies, the one, which might be called the rhombiferous Glypto- 

 dipterini, containing the genera Glyptolcemus, Glyptopomus, and 

 Gyropty chilis, with diphycercal tails, and for the most part rhom- 

 boidal scales ; and the other, which might be termed the cycliferous 

 Glypto dipterini, containing Holoptycliius, Flatygnathus, and Gly2^- 

 tolepis, with heterocercal tails and cycloid scales. 



Professor Pander has endeavoured to prove that the teeth known 

 as Bendrodus belong to fishes of the genus Gyroptycliius. The 

 evidence brought forward in support of this view, however, appears 

 to me to be hardly sufficient to demonstrate its accuracy ; though 

 I think it extremely probable that the teeth and jaws, which have 

 been referred to the genera Bendrodus, Gricodus, Lamnodus, 

 Pkdygnathus, and Rliizodus, will turn out to belong to allies of 

 Gyroptycliius, or, in. other words, to fishes belonging to the family 

 of Glypto dipterini. And again I ca,nnot adopt the family of 

 Dendrodonts" which Professor Pander has established for Gyro- 

 ptycliius, Gricodus, &c., partly because, as he defines it, it seems to 



* It may be that the ventral fins are lobate, but I have seen no specimen justifying 

 that conclusion. 



