10 FOSSIL FISHES OF THE ENGLISH CHALK. 



clavicle. The <>Teater part of its outer face is smooth, but behind aud below a deep 

 vertical groove Avliicli traverses two thirds of its depth, it exhilnts a rugose orna- 

 ment. From its antero-inferior angle there extends forwards a smooth rod-like 

 process (j>.), wliicli articulates with the pterotic. The supraclavicle (scL) is more 

 than three times as deep as broad, truncated above, tapering beloAv, and Avith a 

 thickened anterior margin. Its outer face is very feebly rugose. The clavicle 

 (PI. I, fig. 2, ^7.) is much sigmoidally bent, with a large external lamina, Avhicli is 

 as feel)ly ornamented as the supraclavicle. Its hinder margin is notched just 

 above the attachment of the coraco-scapular mass (c.) ; and at tlie angle there is 

 sometimes a trace of a post-clavicle (pcJ.). Tliere is a persistent suture betAveen the 

 coracoid and scapula, as shoAvn on the left side of the original of PI. I, fig. 2 ; and 

 the scapula is pierced l)y a large oval foramen (/.). The l)asall)ones of the pectoral 

 fin are unknoAvn, and the fin itself has ncA^er been Avell ol)served. As indicated by 

 B. M. no. P, 6465, it is very small, comprising from eight to ten rays, of Avhicli the 

 uppermost and longest cannot have exceeded the premaxilla in length. Only 

 scattered fragments of its delicate fin-rays are shoAvn in PI. I, fig. 1, y. The pelvic 

 fins are unknoAvn, but a trace apparently of one of the supports in B. M. no. 37751, 

 seems to show that their insertion Avas slightly l)eliind that of the pectoral fins. 



The median fins are ahvays imperfect in the fossils, but some of their characters 

 are observable. The dorsal fin seems to arise opposite the insertion of the pehnc 

 pair, and is much extended. There are traces of free fin-supports in advance of it 

 (PI. I, fig. 1, n.) ; and its Ioav anterior portion consists of A^ery stout, nearly smooth, 

 closely arranged spines (PI. I, fig. 4). These spines, six in numlier, gradually in- 

 crease in length, and are foUoAved by much longer I'ays, Avhich are closely articu- 

 lated distally but only represented by their bases in the specimens figured (PI. I, 

 figs. 1, 4). The articulated dorsal rays must haA^e been at least tAv©nty-fiAT in 

 number, as shown by B. M. no. P. 6535. The anal fin must also have been con- 

 siderably extended, arising in adA'ance of the middle point 1)etAA'een the pectorals 

 and the caudal, and reaching at least as far back as the seventh A'ertebra froin the 

 caudal extremity. Its foremost support (PI. I, fig. 1, a) exhilnts at its loAver end a 

 large triangular expansion, eAadently for the attachment of poAverful spines. The 

 caudal fin must liaA^e been delicate, and Avas probably forked. Only fragmentary 

 remains of it are shown in PI. I, fig. 1. 



The aspect of the scales Agarics much in the different specimens, but this circ\un- 

 stance seems to be due to accidents in preserA'ation. When they are undisturbed 

 and complete, as in the front part of PI. I, fig. 1, their exposed portion is observed 

 to be almost rhombic in shape, the hinder border being not gently curved but 

 rather sharply bent at its middle. When the surface is Avell preserA'ed it is 

 distinctly rugose, and some of the delicate markings at the hinder margin radiate 

 in such a manner as to give it a ])ectinated appearance. The smoothness of the 

 scales in the originals of 1^1. T, figs. 1,4, and in similar specimens, is almost certainly 



