466 



Mr. E. L. Gill on an undescrihed Fish 



along the median vertical plane of the fish, but the line of 

 splitting rans slightly across this plane, so that the end 

 of the snout is seen on one side only and the fra»'raent of 

 the tail on the other. As far as the body is concerned the 

 two sides of the nodule are almost identical, so that for 

 purposes of figuring I have thought it sufficient to represent 

 the left side, which includes the snout, and to incorporate 

 in the drawing the few additional details, chiefly of the 

 tail-shaft and caudal fin, which appear only on the right. 

 Unlike that of the body, the state of preservatioii of the 

 head is very different on the two sides. On the left side — ■ 

 shown in the figure — the head has been considerably crushed 

 in laterally. On the right its form is retained as a cav^ity in 

 the nodule, and within this cavity are seen natural but 



PhanerorJujiichus armatus, X 3. Middle Coal Measures, Sparth, nr. Rochdale, Lanes. 

 Manchester Museum, L 8585. Coll. W. A. Parker. 



rough and granular casts of the bones of the palate. The 

 total length of the specimen is ^3*7 cm. ; the complete fish 

 would be about 4 cm. long. 



The specimen had not been critically studied, but 

 palaeontologists who had previously examined it were dis- 

 posed to think it would turn out to be a "super-Palseoniscid," 

 possibly allied to the Permian genus Acentrophorus^ or, at 

 - any rate, a fish of a higher type than any hitherto known 

 from the Carboniferous*. The prominent snout had not 

 then been exposed and the general aspect of the fish was 

 therefore quite different ; and it was presumably the size of 

 the scales, the strong dorsal and ventral ridge-scales and 

 fulcra, and an appearance as of fin-supports equal in number 

 to the dermal rays that suggested a relationship to the 



* A. Smith Woodward, Proc. Geol. Soc. 1915, p. Ixxiii. 



