Mr. Salter on Palaeozoic Fossils from South Africa. 219 



The axis of the thorax is nearly parallel-sided, very convex, and rather wider 

 than the sides, which are curved steeply downwards at a short distance from it. 

 The fulcrum is at less than one-third in front, and much closer behind; and the 

 tips of the nearly direct pleurae have sharply recurved points (see fig. 9). The 

 axis in some specimens (fig. 3) has blunt spines along the middle, — it is difficult 

 to say if this be general or not, as most of our specimens are internal casts. 



The tail is rounded and convex ; its outline is arched in front ; the axis pro- 

 minent, especially at the blunt tip, somewhat conical, and reaching nearly to the 

 margin ; it has six distinct rings ; the sides have four or five distinct furrows, 

 which do not reach the margin ; they are faintly interlined on the outer portion ; 

 tiie edge is serrated, — five teeth on each side, and a short obscure terminal lobe. 

 (The margin is not sufficiently rounded in our fig. 4.) 



The form nearest akin to this species is perhaps one from the summit of the 

 Andes, described by M. D'Orbigny, and of which, through his kindness and that 

 of Mr. Pentland, we have been able to compare the specimen. This rare fossil, the 

 Phacops (Calymene) Verneuilii, D'Orb., differs, however, in some essential points. 

 It has a wider glabella, and the two lower pairs of furrows do not form an oval as 

 in P. Africanus, but are short and direct. The pleurae are spinose, as in ours ; 

 but the tail has much fewer (two or three) ribs on the axis, and the same number 

 on the sides. The specimen does not show the margin of the tail. 



I have little doubt that the Calymene Tristani, quoted by D'Archiac and De 

 Verneuil from the Cedarberg Mountains (Trans. Geol. Soc. 2 Ser. vol. vi. p. 381), 

 was a crushed individual of this or of the following species. The Calymene Blu- 

 menbachii of the same list was identified from our figured specimen, fig. 9. It is 

 rather remarkable that De Koninck should lay stress upon it as the "true 

 species." 



Localities. — In light-coloured soft micaceous rock, black schists, and ferru- 

 ginous nodules, at Gydow Pass ; in light-coloured soft rock and hard dark schist, 

 at Hottentots Kloof ; and in dark-coloured schist and nodules, at Cedarberg. It 

 is associated with Orthis palmata in one specimen from Gydow Pass. 



The specimens illustrated by figs. 6, 7, and 9 were collected by Dr. A. Smith 

 in. the Cedarberg, and were presented by Sir R. I. Murchison to the Museum of 

 Practical Geology. 



Phacops Cafer, sp. nov. PI. XXV. figs. 10-13. 



P. fere 4-uncialis, subdepressus, axi quam lateribus latiore ; capite transverse granulate ; 

 fronte apiculato ; oculis majoribus ; glabella subparallela, sulcis radiatis rectis sequalibus, 

 lobis convexis ; caudae axi conico, baud elevato, multi-annulato ; margine spinoso. 



This species when perfect must have been 4 inches long, judging from the 



