2i8 ANNALS OF SCOTTISH NATURAL HISTORY 



Traquair's figure for this dimension is one-half, while Day's 

 is two-thirds. There are 20 rays in the left fin and 21 in 

 the right. The caudal fin is 2.5 inches long, and is thus 

 practically one-sixth of the total. This is Traquair's pro- 

 portion ; it is less than that of Cuvier and Valenciennes or 

 of Day. There are 38 dorsal and 22 ventral fin rays, but 

 no other specially distinctive external characters. 



Mr. Sim in his note comments upon the cesophageal 

 pouches and speculates as to their function. These structures 

 appear to be characteristic of the Families Stromateidse and 

 Tetragonuridas, to the former of which Centrolophus belongs. 

 They are carefully described by Sim, who, however, omits to 

 note that the ridges which divide the pouches into compart- 

 ments are simply continuations of the cesophageal ridges. 

 They are here deeper, bridging over the cavity of the pouch, 

 and are fibrous in character. In their anterior parts these 

 ridges are very large, and project prominently into the lumen 

 of the gullet as expanded plates. The edges of the ridges as 

 well as of these plates are beset with numerous recurved 

 setiform spines, which, as Sim points out, are simple and not 

 barbed as stated by Gunther. As to the function of these 

 pouches, it may be of interest to note that I found their 

 cavities filled with a soft, creamy, pulpy substance similar to 

 the contents of the stomach and pyloric caeca. The only 

 difference observable was that the material in the latter 

 organs was in a more fluid condition ; it was, in fact, in a 

 further advanced stage of digestion. Remarkable as it 

 undoubtedly is, the facts seem to suggest that these fishes 

 regurgitate their food ; and as these pouches are so very 

 thoroughly supplied with spines it seems possible that some 

 sort of rumination is indulged in. Certainly the substance in 

 the pouches had no resemblance to recently swallowed 

 material unacted upon. Boulenger states that the Stroma- 

 teidae feed on crustaceans, medusas, and the fry of other 

 fish ; the circumstances under which this example was 

 caught suggest the last named at the time of capture, yet 

 there were no recognisable traces of herring fry, either in the 

 pouches or in the stomach. 



NATURAL HISTORY DEPARTMENT, 

 THE UNIVERSITY, ABERDEEN. 



