374 BULLETIN OF THE BUREAU OF FISHERIES 



domus, and Fusus), sea clams (Mactra), and other shells, which the wolfEsh crushes 

 easily in its vicelike molars. Sometimes, however, these mollusks are swallowed 

 whole. In north European waters wolffish are said to subsist largely on mussels, 

 and one taken and examined by Vinal Edwards at Woods Hole was full of these; 

 but although mussels are plentiful on Georges Bank none were found in stomachs 

 of the fish caught there, which points to a definite preference for the other shellfish 

 just mentioned. The wolf is also known to feed on large hermit crabs, ordinary 

 crabs and other crustaceans, starfish, and sea urchins, a quart of the latter having 

 been taken from one caught at Eastport;^' and Mr. Clapp's observations that every 

 one he has opened contained food of some sort is good evidence of its constant 

 search for anything edible. With such a diet it is not surprising that wolffish are 

 more often caught on hand lines baited with cockles or clams than on line trawls, 

 which are usually baited with herring. 



Breeding habits.-*-The breeding habits of the wolffish have not been followed on 

 this side of the Atlantic. In north European waters it spawns chiefly from November 

 until January, ^^ and apparently the breeding season is about the same in the Gulf 

 of Maine, for we have taken larvae of 20 to 22 mm. (fig. 188) — that is, 2 to 3 months 

 old from the time the eggs were deposited — as early as January 30 in 1913, and as 

 late as March 4 in 1920. 



The e^s, 5.5 to 6 mm. in diameter, among the largest fish eggs known, yellowish, 

 opaque, and with an oil globule of 1.75 mm., are laid on the bottom in shoal water 

 where they stick together in lai^e loose clumps among weeds, stones, etc. The 

 fish have been described as making an annual shoreward journey for spawning pur- 

 poses, but there is Uttle evidence of this. The precise duration of incubation is yet 

 to be learned. Probably it is long, as with most fishes laying eggs' that lie on bottom. 



The slender transparent larvae are about half an inch (12 mm.) long at hatching, 

 with an enormous baglike yolk sac inclosed in a net of highly developed blood 

 vessels (see fig. 187, p. 371), thus remotely suggesting salmon or trout larvae in 

 appearance. This yolk gradually shrinks as the larva grows, but it persists much 

 longer than in species producing buoyant eggs. Wolffish hatched by Mcintosh and 

 Prince in the aquarium at St. Andrews, Scotland, did not absorb the yolk sac wholly 

 until about 3 >^ months old and upward of 20 mm. long, but in natural surround- 

 ings larvae as small as 17 mm. have been found free of yolk, nor is any trace of it 

 visible in the larvae of 21 mm. and upward which we have towed in the Gulf of 

 Maine. Larvae of 20 to 22 mm. show the dorsal and ventral spines and fin rays 

 in their final number, but the large head, enormous eyes, and tiny teeth, and the fact 

 that there is no definite separation between the anal and dorsal fins and the caudal, 

 give the young fishes an aspect very different from that of the adult until they are 

 llito\% inches long. In life the wolf is silvery on the sides at this stage, but this 

 metallic hue fades after preservation, leaving only the dark brown pigment granules 

 with which the sides are thickly dotted. The largest fry we have seen (44 mm. 

 long) show similar but somewhat paler pigmentation. 



" VerrUl, 1871, p. 400. 



»» It was formerly thought to spawn in spring, but Mcintosh and Prince (Transactions, Royal Society of Edinburgh, Vol. 

 XXXV, Part III (No. 19), 1890), to whom we owe all that is known of its early larval development, proved it an autumn and 

 winter spawner both by examination of its ovaries and by the discovery of its eggs. 



