Fishes of the Western North Atlantic 537 



virens), silver hake (Merluccius bilinearis), and perhaps spiny dogfish {Squalus acanthias)^ 

 and by cormorants and gannets. The larger Salters are so active that they are not likely 

 to suffer seriously from these predators, but seals doubtless take a toll of them. During 

 their sojourn in fresh water they suffer from attacks by herons and eels, as shown by 

 the scars sometimes seen on their sides {y^: 182); mink and otter are also poten- 

 tial enemies. 



N. C. White has contributed the information that when the Salters re-enter the 

 streams they are heavily infested with sea lice (parasitic copepods, genus Lepeophtheirus) 

 similar to those that infest fresh-run Atlantic salmon, but these drop off soon after the 

 fish have entered fresh water. 



Variations. Anglers have long been aware that the populations oi fontinalis in dif- 

 ferent lakes, ponds, and streams differ widely in color, relative size of head, depth of 

 body, size at maturity, and maximum size of growth. Hatchery experience, however, 

 has proven that the rate of growth depends primarily on the supply of food available 

 to each individual fish combined with the yearly duration of the season when the water 

 is neither too cold nor too warm for them to feed actively. 



Fat fontinalis are clearly deeper-bodied than lean ones, and Ricker has pointed 

 out that the maximum size to which fontinalis grow seems correlated with the size of 

 the body of water in which they live, and with the presence of suitable large food, 

 i.e. fish or crayfish {^y. 73). These generalizations find their expression in the large 

 sizes to which they tend to grow in large streams and lakes in the northern part of 

 their range, contrasted with the stunted but overplentiful populations that inhabit 

 many of the smaller mountain streams of southern North Carolina and northern Georgia, 

 w\ieve fontinalis "rarely pass the dimensions of fingerlings" (Jordan, ^l: 112) though 

 they are self-sustaining. Also it seems from casual firsthand observation that the heads 

 average larger, relatively, among the slow-growing populations than among fast- 

 growing fish, quite apart from sexual dimorphism. 



In ^enerdX, fontinalis average paler when living over a pale sandy bottom in brightly 

 illuminated waters than over a dark bottom in shady situations, but their nuptial 

 coloration usually is more brilliant in the second case than in the first. But this rule is 

 not invariable (for an interesting exception among Labrador fish, see Weed, J 2: 131). 

 Brook Trout that are dark when fresh-caught soon fade after death. In short, it seems 

 unlikely that any of the local populations oi fontinalis represent recognizable subspecies, 

 except perhaps those in Dublin and Center ponds and Christmas Lake, New Hamp- ■ 

 shire, which were made the basis of a separate species (agassizii) by Garman {2^: 78). ^^ 



The fontinalis that are taken in salt water in estuaries or at sea differ so widely 

 in their pale coloration and silvery sides from the freshwater Brook Trout that they 

 have been mistaken by various authors (60: 141, 142; 5J: 131, 132; ^: 206; 5J: 

 2555) for the sea-run form of Salmo trutta., the brown trout (p. 498); and those from 

 Hudson Bay, Labrador, and from Newfoundland were described by Suckley (6^: 310) 

 as a separate species, Salmo hudsonicus, which has been accepted as a subspecies by 



31. For historical account, diagnostic characters, and excellent colored figures of agassizii, see Kendall (45: 56-68, pi. 6). 



