250 DEEP-SEA FISHES OF THE ATLANTIC BASIN. 



Outside of the Mediterraueau it has been identified by Gapello from Lisbon, where he 

 says it is very rare and found only in summer. 



Moreau says he has found it common at Marseilles, and identities with the species in- 

 dividuals observed by him at St. Jean de Luz, where it is very common, and at Biarritz, 

 where it is known by the Basque name Crahra; at Valence and at Arcachon, where it is 

 exceedingly rare, M. Lafout having obtained a single specimen, trawled and brought to 

 market in February, 1871. 



The species also occurs in the western Atlantic in numerous localities, having been 

 first discovered by the Fish Hawk, in 1880, off Narragansett Bay. The western form pre- 

 sents no characters by which it can be distinguished from that of the Mediterranean, except 

 that the lower rays of its pectoral fin are somewliat more prolonged in the examjile studied 

 by us. This also has a perfectly smooth suborbital keel, while most of the western speci- 

 mens have a small spine on the anterior portion of this keel. Tliis character is of little mo- 

 ment. In some of our specimens the spine is present on one side and absent on the other. 



In comparing H. davtylopterus with H. maderensis, special attention should be paid to the 

 preopercular spines. The limb of the preoperculum is much more nearly vertical in the 

 Madeiran type, and the spines are more prominent, and their axes (as shown in Lowe's figure) 

 are parallel with each other and with the central axis of the fish's body. This tendency to 

 parallelism is noticeable in some of our specimens of H. dacti/hptern.i. The Bladeiran fish 

 has a much more heavily aruaed head, and its coloration is different. The back part of the 

 mouth of H. dactyloptefm is plumbeous in both Mediterranean and American specimens, but 

 it seems hardly probable that it cau have been as conspicuously black as in that of Madeira. 



In addition to the American specimens, we have had before us a specimen obtained by 

 President Jordan at Genoa (U. S. Nat. Mus. Cat. No. 29783). 



HELICOLENUS MADERENSIS, Goode & Bean, n. s. 

 Sehastes imperialis, Lowe, Synopsis, Pishes of Madeira, 175 ; Fishes of Madeira, 171. pi. xxiv. 



A Rclicolenm, with body moilerately compressed, as high at the origin of the anal as at 

 that of the dorsal, with its profile flatly arched to a height equal to about one-third the 

 diameter of the orbit. Its greatest height, at the origin of the ventrals, is equal to the 

 length of the head (measured from the tip of the snout), and slightly exceeds one-third the 

 length of the body; the thickness, which is greatest behind the eyes, about one-half of its 

 height. 



Bye large, the upper limb of the orbit encroaching upon the upper pi'ofile of the head ; 

 its diameter equal to the length of the snout, but less than the postorbital portion of the 

 head, and consequently less than one-third the length of tlie head. The interorbital space 

 deep and strongly ribbed, its width scarcely one-half the diameter of -the orbit; a promi- 

 nent depression in the occipital region. Suborbital equally feeble, aculeate, with generally 

 only one slight spine. Preoperculum with five strong spines, equidistant and regularly 

 arranged, the uppermost, which is 0])posite the end of the suborbital keel, or sometimes one 

 below it, a little the most conspicuous, all having the same horizontal or parallel direction, 

 being nearly straight or but slightly hooking upwards. Thescapuhu'yand two suprascapu- 

 lary spines small and crowded, foiiuiiig tlie usual triangle. Two (ionspicuous spines upon 

 the upper part of the opercular flap, below its angle; spines upon the anterior suborbital 

 inconspicmnis. 



The teeth are not described by Lowe. 



Tlie dorsal begins slightly behind the upper anterior angle of the preoperculum, and 

 (as shown in the figure) the roots of the first and second S]nnes seem very close together. 

 The second and fourth spines are nearly equa", the third sliglitly exceeding them in length ; the 

 fifth, sixth, and seventh are nearly equal; the eiglitli, ninth, and tentli sliglitly decreasing 

 from one to the other, the ninth as long as the first, wliicli exceeds the twelfth by as much 

 as this exceeds the tenth, and the tenth exceeds the eleventh. Tliis fin is less regular in 

 the diminution of the lengtli of its si)ines than tliat in IT. durfi/liiptcrtts. 



The anal begins farther forward than in H. dctutyloptcrun, apparently in the vertical from 



