98 ACANTHOPTERYGII. 



near Balliiitrae, on the coast of Ayrshire, on the 29lh of November, 1837, after a 

 severe storm. 



" It was taken to Dr. Wylie, of the village, who, on learning from the fishermen 

 that the species was imknown to them, most liberally transmitted it to me.* 

 In consequence of its size, and its being received in a perfectly recent state, I 

 here transcribe some of my notes, made on comparing the specimen with the 

 descriptions of various authors, before it was transferred to spirits. 



" The largest English Cepola on record, is described by Mr. Couch in the Lin- 

 nsean Transactions (vol. xiv. p. 76) to have been 15 inches in length. Cuvier 

 and Valenciennes observe (Hist, des Poiss. t. x. p. 398) that their specimens were 

 a foot long ; but add, that the species has been found a foot and a half in length. 

 The present specimen, although broken off near the tail, is 19§ inches long; and 

 as the body, when perfect, tapers to a point, and that of the individual under 

 consideration is 2 lines deep at the fracture, I should consider, judging from the 

 gradual diminution of its depth before this part, that it must have been from 

 about two to three inches longer. The depth of the head is 1 inch and \ a line ; 

 the greatest depth of the body (just behind the gill covers) is 11 lines, or I3 

 line less than the depth of the head, and thence it tapers gradually towards the 

 tail. Its thickness close to the head is 4| lines, at the centre 1^ line, and at the 

 extremity ^ a line. Its weight is scarcely 1 oz. 



" The species has been generally described as destitute of scales. Mr. Yarrell, 

 however, states, that a specimen sent to him by Mr. Couch, ' exhibits, here and 

 there, an occasional thin, oval, semi-transparent scale.' — {Br. F. vol. i. p. 197.) 

 It is remarked by Cuv. and Val. — ' Les ecailles de la Cepole sont extremement 

 petites, ovales, lisses, entieres, insensibles au tact, ne s'imbriquent point, et 

 se presentent a la loupe comme autant de petits pores enfonces et disposes en 

 quinconce serre ; ce n'est qu'en raclent la peau, qu'on en detache et qu'on pent 

 les voir separement : la tete et les nageoires n'en ont aucunes.' (t. x. p. 397.) 

 My specimen entirely coincides with this description, but it may be further ob- 

 served, that its scales increase gradually in size from the head towards the tail, 

 and that in approximatmg the latter they are apparent to the naked eye ; 

 from being more sunk in the skin, in addition to their smaller size, they are not 

 thus visible on the anterior part of the fish ; — with a low magnifying power the 

 longitudinal stria of the scales on the posterior portion are conspicuous. 



" The Cepola rubescetis and C. ternia are described by authors who hold them 

 to be distinct, the former as possessing one, and the latter two, rows of teeth 

 in the lower jaw. Donovan (British Fishes, No. 105) and Yarrell f have consi- 

 dered that this difference may be owing to the age or size of the individual. 

 Risso in his Histoire, (ed. 1826, tome iii. p. 294,) in which the C ruhescens and 

 C. tmiia are brought together, though in his Ichthyologie they were regarded 

 as distinct, attributes 14 teeth to the upper and 16 to the lower jaw. Cuvier 

 and Valenciennes enumerate 17 or 18 teeth on each side of the upper jaw, and 

 10 on each side of the under, behind which 2 appear, and add that they vary a 

 little in individuals. My specimen, considerably exceeding in magnitude the 

 CepolcB examined by these authors, exhibits 41 teeth in the upper jaw (cavities 

 denote that many are wanting) and 25 in the lower, of which latter 17 are in a 

 tolerably regular row, inside of which is 1 tooth, and outside it 7, which are 

 equal in length to the largest in the row, but not so much hooked. The tongue 

 is smooth. 



" The lateral line is apparent only on close examination, being a mere faint-co- 



* This specimen afforded an illustration of the correct application of Tani- 

 oidei, or " Poissons en ruban," to the family in which it ranks, in a point of view 

 that, in all probability, was overlooked by Cuvier. Although 19| inches long, it 

 was folded up like a riband, and forwarded through the post office, under cover 

 of a franked letter of ordinary size and legal weight. 



t In a specimen 7^ inches long, this author found one tooth in the line of the 

 second row; and in an individual 13 inches in length, six teeth constituted this 

 row. — British Fishes, vol. i. p. 197. 



