90 OLD RED SANDSTONE FISHES. 



Pterichthys, Agassiz, 1840. 



Asterolepis, Pander (pars). 



Generic Characters. — Pre-median plate of the head not notched in front ; 

 anterior median dorsal plate of body overlapping the anterior dorso-lateral, but 

 overlapped by the posterior dorso-lateral. Tail covered with rounded scales, and 

 showing a triangular dorsal and a heterocercal caudal fin. Other characters 

 as in Asterolepis, but the carapace is more elevated in shape. 



History. — Pterichthys, first discovered at Cromarty about the year 1831 by Hugh 

 Miller, and again independently at Lethen Bar by Stables and Malcolmson in 1839, 

 received its name from Agassiz, to whom specimens collected by Miller were trans- 

 mitted for examination. The name was first published by Sir Roderick Murchison 

 in the report of the meeting of the British Association at Glasgow in 1840, but no 

 description appeared until 1841, when Hugh Miller gave an account of it in his 

 " Old Red Sandstone." This description was accompanied by restored figures of 

 both the dorsal and ventral surface of the animal, which, if neither perfect nor fault- 

 less, certainly showed that Miller had managed even at that time to attain a 

 remarkable insight into the structure and configuration of this singular creature. 



In 1844 Agassiz, in his ' Poissons fossiles du vieux gres rouge,' described the 

 genus not only from Cromarty specimens supplied to him by Hugh Miller, but 

 also from material from other localities, such as Lethen Bar, Clune, and Orkney, 

 contained in the collections of Lady Gordon Cumming, Lord Enniskillen, Sir Philip 

 Grey-Egerton, Professor Traill, and others. But it is surprising that Agassiz, 

 in this work, seems to have paid no attention to Miller's comparatively correct 

 description and figures, but, on the other hand, gave a most inaccurate description, 

 in which he even mistook the ventral for the dorsal surface, while his " restoration," 

 unfortunately copied and recopied into too many geological text-books, bore no 

 more than a very remote resemblance to the creature it was meant to represent. 

 Strange indeed, for one of Hugh Miller's specimens figured by Agassiz (PI. IV, 

 figs. 1 — 3) shows both dorsal and ventral surfaces as well as the head, and is, to 

 this day, one of the most instructive examples known. To Pterichthys Agassiz 

 allotted eight species, the rectification of which will be presently considered. 



In 1848 Sir Philip Grey-Egerton published a paper on the structure of 

 Pterichthys 1 containing copious extracts from letters written to him by Hugh 

 Miller, and in which Agassiz' more glaring mistakes were rectified. Nevertheless, 

 contrary to the more correct representation given by Miller in the " Old Red 

 Sandstone," the front part of the anterior ventro -lateral, to which the arm is 

 articulated, is marked off as a distinct " thoracic " plate, while the hinder 

 1 ' Quart. Journ. Geol. Soc.,' vol. iv, 1848, pp. 302—314. 



