458 Scientific Proceedings, Royal Dublin Society. 



the matter of barbel, seems to feed extensively on large and active fish. In the 

 Whiting we meet with a fish very usually provided with a barbel in its younger 

 existence ; but, in the adult condition, usually, though not invariably, destitute of such 

 an appendage, and furnished with powerful teeth. It appears to prey chiefly on fish. 

 The Pollack is a fish which throughout life has a considerably projecting under jaw^ 

 and seems always to seek a more or less pelagic diet. The Coal-fish presents us Avith 

 both conditions. Having a barbel and protruding upper jaw in its young condition 

 (up to about 12 inches in length), it certainly derives a considerable proportion of its- 

 food on the bottom, whilst in approaching the adult stage the under jaw outgrows the 

 upper, and the barbel becomes the merest rudiment ; and the full-grown fish, pre-eminently 

 suited for an active life, becomes one of the greatest persecutors of the Herring. In 

 the Herring and Mackerel we find the under jaw protruding ; both are well known to- 

 derive their food from pelagic sources — copepods and the like. The adult Herring 

 being destitute of teeth, whilst in the transparent postlarval condition, living near 

 bottom and preying upon relatively large organisms, it is furnished with rather powerful 

 teeth. In fine, the projecting under jaw is almost always associated with a pelagic 

 diet. It is found, I believe, in the pelagic postlarval stages of all bony fishes, whether 

 or no it persists in the adult. Unfortunately our observations during the Survey were 

 almost entirely confined to the food of mature or fair-sized fish. 



EED GUENAED— r>7V7/« cumlus. 



During the Survey 7 stomachs were examined. 2 were empty. Of the remaining- 

 5, Crustaceans occurred in 4 ; unidentified crabs in 3 ; Common Shrimps in 2. 

 Galathcea in 1. Fish, Sand-eels in 1. 



So far as they go, these observations appear to show that the Eed Gurnard feeds, 

 more on crustaceans and less on fish than the Grey and Sapphirine Gurnards. 



SAPPHIEINE GUENAED— Tri^/a Mrundo. 



During the Survey 10 stomachs were examined. 4 were empty. Of the remaining^ 

 6, Annelids {Nereis) occurred in 1 ; Crustaceans in 2 ; Swimming Crabs {P. holsatus), 

 Mask Crabs {Corystes) and Common Shrimps each in 1 ; Lammellibranclis, Eazor 

 Shells (Solen) in 1. Fish in 4 ; Sand-eels in 2 ; Lesser Weever in 1 ; unidentified fish 

 inl. 



The figures are too small to be much of a guide. The food seems to be much the 

 same as that of the Grey Gurnard. 



GEET GJJB.'N AnB—Triffla gurmrdus. . 



In 1890 64 stomachs were examined. 10 were empty, and in 3 the nature of the 

 food was iindistinguishable. The records do not show the exact distribution of the 

 different sorts of food in the remaining 51. Annelids occurred in 3 ; Crustaceans in 

 about 28 ; Crabs in 6 ; Shrimps [Crangon) in about 5 ; Mysis in 2 ; Amphipods,. 

 chiefly Gammarids, in about 4 ; Zoese in 2 ; unidentified Crustaneans in 9 ; Fish in 

 about 30 ; Sand-eels in about 18 ; Sprats in about 5 ; Grystallogobius in 2; Dabs and 

 Dragonets each in 1 ; unidentified fish in 3. 



In 1891 110 stomachs were examined. 33 were empty. In 1 the nature of the 

 food was not distinguishable. Of the remaining 76, Annelids occurred in 1 ; Crustaceans 

 in 39, 51 per cent.; Shrimps [Crangon) in 23 ; Schizopods in 10; Prawns, Falcemon 



