FISHES OF THE GULF OF MAINE 327 



narrow limits. Thus in winter it deserts the shoalest bays where the flats are 

 laid bare at low tide (Duxbury Bay, for example), no doubt to avoid the extreme 

 cold, reappearing there in March, moving out again in midsummer when the water 

 on the Hats is warmest, to work back to the shallows once more in September or 

 October. In the estuaries of the Bay of Fundy, however, where the summer 

 temperature of the shoal water is lower than in similar locations in other parts of 

 the Gulf, longhorn sculpins are most plentiful during the warm months and almost 

 all of them move out into the open bay to pass the winter. Along the southern 

 shores of New England, where the coastal waters are much warmer in summer than 

 in most parts of the Gulf of Maine, this sculpin is abundant in autumn and winter, 

 but rarel}' taken in very shoal water in summer. 



These migrations are probably induced by temperature and they are sufficient 

 evidence that this species avoids both the warmest and coldest water — that is, 

 temperatures higher than 55° to 60° and lower than 35°. However, even freezing 

 temperatures are not fatal to it, for we find no evidence that sculpins are ever 

 killed by cold when caught on the flats in severe freezes, a fate that occasionally 

 overtakes cunners and tautog. 



Breeding habits. — All that is known of the breeding habits of this sculpin is 

 that it spawns in November and December about Woods Hole, its eggs sinking 

 together in clumps like those of the shorthorn sculpin, and often being thrown 

 up on the beach in masses. Presumably the spawning season is the same in the 

 Gulf of Maine, but whether it gathers in particular localities or at any precise depth 

 to spawn or does so indifferently on all kinds of bottom is yet to be learned. The 

 presence of sculpins of all sizes, from very young fry to adult, proves that it breeds 

 generally along the coasts of Massachusetts Bay and Maine, probably along 

 western Nova Scotia as well; but in the Bay of Fundy it seems restricted as a breeder 

 to the Scotian side, the absence of young on the New Brunswick shore proving 

 that the half-grown and adult fish that are so plentiful there are immigrants either 

 from across the bay or from the Gulf outside. 



The larval stages have not been described previously; hence it was interesting 

 to tow four young sculpins in April, 1920 (three on the eastern part of Georges 

 Bank and one in the Eastern Channel), which probably belonged to this species 

 because of their long cheek spines. The smallest of these larvae was 13 mm. long, 

 showing the first traces of the dorsal and anal fin rays, and the largest was 21 mm. 

 long. The larvse are more slender than corresponding stages of the shorthorn 

 sculpin (p. 321), and differ from them in the outline of the dorsal fin, for in the 

 longhorns (if our identification be correct) it is continuous from end to end, 

 only the largest of them showing a shallow notch to separate its spiny from its 

 soft portion, whereas in the shorthorn the two sections are separate from the time 

 the fin first takes finite form. Nothing definite is known of the rate of growth 

 of this fish nor of the age at which it matures. 



Commercial importance. — The only commercial value of this sculpin is as bait 

 for lobster pots, for which it is used in great numbers, being speared in some locali- 

 ties and caught on hook and line in others. It is a fairly good fish on the table 

 and a few are eaten by the foreign-born population, but there is no market for them. 

 102274— 25 f 22 



