316 NOTES. 



fish ; and the foreign examples from rocks older than the up- 

 permost Ludlow are all easily demonstrable as invertebrate 

 remains. But although, when the above paper was written, 

 no true fish had been detected below the horizon of the 

 ' bone bed ' of the Ludlow rock, the supposed coprolite no- 

 dules in the Lower Silurian still left a doubt on the matter 

 (Quart. Geolog. Jour., vol. vii. 1851). This doubt also has 

 been dissipated, since Mr Hunt of the Canadian Geological 

 Survey has shown how phosphotic riodules occur deep down 

 in Silurian rocks, and even at their base, which are clearly 

 derived from the animals and horny shells of Lingula and 

 Discina, the commonest of ancient fossils. 



" Later researches have made it manifest that true placoids 

 are not the oldest known forms of vertebrate life. The spines 

 and the shagreen of shark-like fish were certainly the chief 

 constituents of the 'bone bed/ but with them (and the 

 doubtful remains called Plectrodus) another genus occurs be- 

 longing to the puzzling group of the Cephalaspidse. This is 

 Pteraspis, now for some time known to be a distinct genus. 

 It has been detected in the bone bed, and in the Upper Lud- 

 low rock itself, beneath the bone bed, by the persevering geo- 

 logists of Ludlow. (See * Siluria,' second edition, 1859, p. 

 267.) And while these pages are preparing for press, a spe- 

 cimen of this genus is announced by them from the Lower 

 Ludlow rock ! This remarkable group, then, of fishes, whe- 

 ther they be abnormal ganoids, as Agassiz supposed, or have 

 closer relations with the Teleostean or bony fishes, to which 

 the Siluroids are now referred, are undoubtedly, so far as our 

 present knowledge goes, the oldest of known fijsh." 



