422 APPENDIX. 



Finally, we may state the essential differences between our b)-acanthiis and 

 Cuvier's lucertia, in the following manner : — 



T. lucerna Cuv. iv. p. 72. T. hracanthus. 



L'angle anterieur superieur de I'or- No dentations whatever above or 



bite sont trois dentelures. before the orbits, and only one 



behind. 



Lepointes dubas deson pri^opercule Xo prickle whatever on the pre- 



celles de son opsrcule, et de son opercule. 



epaule sont moins aigues et moins 



sallantes guana. — Rouget. 



S scales at the lateral line; " ont Tlie scales rot sinuated, and only 



au milieu sur leger sinus ventrant one sixth height of the body. 



— leur hauteur fait pr^s de quart 



de celle du corps. 



Dorsal 10, 18.; anal not mentioned. Dorsal 8, 19 ; anal 17. 



Considering how very constant are the number and proportion of the 

 spines in this genus, paticularly those of the head, and looking to the above 

 differences, we entertain the belief that they are distinct species, more 

 particularly as the number of the dorsal rays likewise differs. It still re- 

 mains to be ascertained what are the colours of the pectoral fin in Cuvier's _ 

 lucerna f what the number of its anal rays? and whether these rays, 

 and those of the second dorsal, are simple or branched ? 



Obs. Regarding this, and one or two other Gurnards here characterised as 

 new, we have to choose one of three opinions : — 1. Either that they are 

 subject to a much greater variation than naturalists are accustomed to 

 attach to the idea of species ; or 2, that the existing descriptions we have 

 quoted are inaccurate in an unusual degree ; or thirdly, that they are 

 really distinct Every one, of course, will form his own conclusions on the 

 superior degree of probability which he may attach to one or other of these 

 suppositions. With respect, however, to the Trigia bracanf/u/s, I may 

 observe that I attach, with Cuvier, a primary importance to the armature, 

 or spines and prickles, upon the head, &c ; and in proof that these do not 

 in the least vary in individuals from different localities, I must remark 

 that the existing descriptions of northern, or British specimens of T. lyra, 

 apply most accurately to those of Sicily ; and that the same may be said of 

 Cuvier's description of the T. aspira of Viviani. "We cannot, therefore, 

 reasonably adapt the first supposition, because, the very reverse is applicable 

 to others of the same group. 



TRIGLA leucoptera. WTiite-finned Gurnard. 



The first dorsal spine serrated in front ; pectorals ex- 

 ternally white, half as long as the trunk ; ventral rays 

 forked, lateral line smooth ; orbits with 2 prickles 

 before and one behind. 



Dorsal S or 9; the second 16; anal 15; ventral 6; pectoral 

 10; caudal — ? 



Inhabits the western coasts of Sicily, but very rare. 



Length 3 — 5 inches, 



The above specific character will' at once distinguish this most elegant 

 little species from all its allies. Its form is not so thick, and the head is 

 more lengthened than in the last. The muzzle is cleft, each lobe being ser- 

 rated with five or .six small spines; before the eye there are two, and 

 behind is oti£ spine ; behind this latter, and placed on the upperpart of the 



