Holt — Survey of Fishing Grounds, West Coast of Ireland. 459 



and Tandalus, each in 1 ; Swimming Crabs and 2oecs each in 1 ; unidentified 

 Crustaceans in 4 ; Cephalopods, Squid (i. media) in 1 ; Fish in 41, 53 per cent. ; 

 Sand-eels in 21; Crystallogohius in 3 ; Norway Pout, Dab, and Dragonet, each in 1 ; 

 unidentified fish in 11 . 



It appears, therefore, that Fish and Crustaceans together contribute almost the whole 

 of the food, the proportion of the former slightly exceeding that of the latter, whilst 

 Annelids and Molluscs — small Squids, are occasionally eaten. Of the fish Sandeels 

 appear to be most sought after, Grystallogobvus, Sprats, Norway Pout, Dragonet and 

 small Dabs being less preyed upon. Of Crustaceans the more active kinds seem to be 

 the chief sufferers, especially Common Shrimps, and, to a less degree, Schizopods. 



Smith's observations in the Firth of Forth are much to the same effect, but the 

 proportion of fish seems to be higher than that of Crustaceans, whilst he found 

 Hermits contributing a rather large share of the latter, whereas they do not appear in 

 our observations. He notices a single occurrence of an Echinoderm out of 407 stomachs- 

 which contained food. 



PIPER— TW^to lyra. 



During the Survey 21 stomachs were examined. In 1 the nature of the food was. 

 indistinguishable. Of the remaining 20, Echinoderms (Brittle Stars) occurred in 16^ 

 SO per cent.; Opliioglypha lacertosa in 15; 0. albida in 2, and Amphiura in 1 ; 

 Crustaceans in 14, 70 per cent.; Hermits in 11 ; Crabs in 4, viz. Gonoplax'va. 2; 

 Swimming Crabs in 1, and unidentified Crabs in 1 ; Common Shrimps in 1 ; uniden- 

 tified Crustaceans in 2 ; Gastropod shells {ISatiea, Turritella and Aporrheus) in 3, 

 probably the "Houses " of Hermits ; Fish in 1. 



It appears, therefore, that the Piper is the most exclusive bottom feeder of its 

 genus (as indeed its structure would seem to indicate) and derives its food almost 

 entirely from Echinoderms (thus markedly differing from T. gurnardus) and Crustaceans.. 

 Of the former, 0. lacertosa, and of the latter. Hermits are evidently preferred. 



ANGLER — Lophius piseatorius. 



During the Survey 7 stomachs were found to contain food. The contents were in all 

 cases fish. Sprats, Common Dab, Pole Dab, and Sole occurred each in 1 ; unidentified 

 fish in 5. 



Smith concludes that without doubt fish forms almost the whole food of the Angler, 

 having only found an invertebrate, a Hermit, in one stomach out of 33 which contained 

 food. Day has collected a number of instances in which this fish has been known to 

 swaUow various sea birds, and other objects of no service as food, and remarks, on the- 

 authority of Pennant and another, that they are returned to sea by fishermen in 

 certain localities on account of their eating Dogfish. This may, perhaps, be looked upon 

 as somewhat doubtful policy. 



Mr. Lane made some experiments with one of these fishes on the deck of the 

 " Harlequin." He found that on touching the top of the erected filament with a stick 

 the Angler immediately made a snap, closing his jaws so as to exactly catch that part 

 of the stick which had touched the filament. This was continued until the fish became 

 exhausted. It may, therefore, be that the action of the fish is to some extent involun- 

 tary, and that the filament serves not only as a lure, but to direct the snap. It would 

 seem unlikely that a dying fish would voluntarily snap at any object, and contact of 

 the stick on any other part had no effect in stimulating it. 



