478 Correspondence — A. JR. Sunt. 



some of the Calciferous Sandstone Series," as McCoy supposed, the 

 Mansfield fishes are typically and essentially Carboniferous. "Of 

 the six genera represented in the collection, one {Eupleurogvius) 

 is too imperfectly known for discussion ; four of the others 

 (Acanthodes, Ctenodiis, Strepsodus, and Elonichthys) have hitherto 

 been discovered only in the Permian and Carboniferous of Europe 

 and in the Carboniferous of North America; while the sixth 

 (Gyracanihides) is related to an essentiallj' Carboniferous fish in the 

 northern hemisphere, and bears every mark of belonging to the 

 same late Paleeozoic period," Gyvncanthides, which McCoy was 

 correct in recognizing as a close ally of Gyracanthus, is regarded 

 by Dr. Smith Woodward as a remarkable discovery. The new 

 specimens prove Gyracnvthides to be a typical Acanthodian belonging 

 either to the Diplacanthidas or to a distinct family marking the 

 culmination of this series. The fins exhibit peculiarities indicating 

 a high degree of specialization, which was analogous to that 

 observable in later geological periods among Selachians and 

 Teleosteans. 



The Mansfield fishes are such as often occur in estuarine and 

 fresh-water strata in the northern hemisphere, though all the genera 

 are occasionally found in sediments of marine origin. Their 

 association with the remains of land plants suggest estuarine 

 conditions. 



The systematic descriptions are illustrated by 11 lithographic 

 (i"ock-colour tinted) plates, with three figures in the text, one of 

 which is a restored drawing of the remarkable Acanthodian 

 Gyracanthides Murrayi. 



The following are the species described and figured : — 



Gyracanthides Murrayi, sp. nov. Strepsodun decipiens, sp. nor. 



Acanthodes AustraJis, sp. nov. Elonichthys Sweeti, sp. nov. 



Eupkurognuis CresiceUi, McCoy. E. gibbus, sp. nov. 



Ctenodiis breviceps, sp. nov. 



A. H. F. 



coTa:REis:poisr3D-£]3^0E!. 



THE MODE OF ACCrMULATION OF THE SOUTH DEVON RED 

 SANDSTONES AND CONGLOMERATES. 

 Sir, — I am glad to see that Mr. Hobson ^ has attacked the 

 perplexing problem of the mode of accumulation of the Devonshire 

 red conglomerates. The remark of mine which he quotes was not 

 intended to throw any doubt on the fact that water was the agent : 

 the difficulty is to conceive any probable mode of action. The 

 Torbay evidence is much as follows. In the English Channel, a few 

 hundred yards south of Berry Head, is a rocky islet, and in its 

 limestone there is apparently a pipe, or small swallow-hole, filled 

 with red sandstone. In the eastern face of Berry Head there is, 

 if I remember aright, a small fissure filled with sandstone. In the 



1 See Geol. Mag., July, 1906, pp. 310-320 (PI. XXIa). . . 



