CENTROLOPHUS NIGER ON THE SCOTTISH COAST 217 



ing on young herring. None of them has ever seen such a 

 fish before." This interesting find on examination proves to 

 be a specimen of a somewhat rare visitor to northern waters, 

 viz. CentrolopJius niger, Gmelin. regarding which notes have 

 already been made in these " Annals." It is generally 

 regarded as a Mediterranean fish ; Regan (" Ann. and Mag. 

 Nat. Hist.," 1902, p. 195) gives its distribution as " Mediter- 

 ranean and N. Atlantic." According to Traquair, it has also 

 occurred in Ireland, and Jordan and Evermann record it 

 from Dennis, Mass., although Traquair throws doubt upon 

 the identity of the species in this latter case. The present 

 specimen is the fourth recorded from the east coast of Scot- 

 land ; three of these have been found between Lossiemouth 

 and Aberdeen, the fourth occurred in Largo Bay (see notes 

 in the "Annals" for 1902 by Dr. Traquair and Mr. George 

 Sim). The earliest record is by Dr. Gordon in 1841. 



In Dr. Traquair's communication some particulars are 

 given as to dimensions, in which he notes in the proportions 

 of his specimen some differences from the current descriptions 

 and figures. Since the Cullen example proves on the whole 

 intermediate it may be of some value to record particularly 

 its more important characteristics. 



Along the dorsal region the fish is a deep blue-black 

 colour ; ventrally it is lighter, being of a greyish slaty blue. 

 This lighter colour is in part, however, due to the fish having 

 been immersed in spirit. The total length is 14.7 inches ; 

 the greatest depth, just in front of the anus, is about 3 inches. 

 It is thus contained! slightly less than five times in the total 

 length. Traquair's specimen is stated to be in body depth 

 "slightly over four times in the total." Regan gives the 

 depth of the body as about four times. From snout to 

 posterior margin of operculum this latest example measures 

 2.9 inches, almost exactly one-fifth of the total length. 

 This accords with the proportion given by both Gunther and 

 Day, viz. one in five. Traquair's measurement is " rather 

 less than one in five and a half." Regan gives this dimension 

 as four and a third to five times, hence Traquair's figure ex- 

 tends the range of variation of this dimension. The pectoral 

 fin measures 1.6 inches to the tip of the longest ray, giving a 

 proportion of rather more than half the length of the head. 



