Fishes of the Western North Atlantic 565 



It has been found that American Smelts average somewhat larger at maturity in the 

 Bay of Chaleur, Quebec, than in the estuary of the St. Lawrence River (64: 71-85), 

 and that those of the Bay of Chaleur have a larger number of vertebrae on the 

 average (62.18-62.9 according to locality) than those from various places between 

 Matane on the southern side of the estuary and Quebec City (60.32—61.92). The 

 first of these differences may be associated with the food supply; the second illus- 

 trates the long-known rule that among fish species that occur over a wide range 

 of temperatures, the number of vertebrae average the most numerous among popu- 

 lations that pass their growth stages in colder than in warmer water. Nothing else 

 is known in this regard for any other parts of the American coast. For the American 

 Smelt in fresh water we know only that they run much larger in some lakes than in 

 others. 



Racial studies of the subspecies mordax of the White Sea region, Arctic Asia, 

 and eastern Siberia by Petrow (as dentex) did not reveal any regional differences call- 

 ing for recognition in nomenclature {jy. 177—188). For the European subspecies 

 eperlanus, Hubbs {43: 52) has pointed out that the gill rakers average more numerous 

 in specimens from Scandinavia (32-37) than in those from the English Channel 

 (25—32). And Jensen has reported slight differences, based on extremely detailed com- 

 parisons of smelts from different lakes in Denmark, especially in the number of verte- 

 brae (49: 73-1 09). 12 



Abundance. In colonial days, according to Captain John Smith, American Smelt in 

 1622 were in "such abundance, that the Salvages doe take them up the rivers with 

 baskets, like sives" {56: 244). While it seems certain that they were far more plentiful 

 along the coasts of New York and Maine up to the early part of the nineteenth 

 century than they are today, by 1849 a decrease was reported for northern New 

 Jersey, and by 1885 for the streams of Long Island, New York. By the end of 

 the century, at Cold Spring Harbor, the New York State Hatchery, which planted 

 48 million fry in 1898, produced only a few million in 1899 for lack of ripe females 

 and eggs from nearby waters, and after 1900 its operations were discontinued. Many 

 reports of their depletion in the waters of Massachusetts have been received also; 

 there the number of eggs from the Weir River, tributary to Boston Harbor, fertilized 

 at the Palmer Hatchery, fell from around 100 and iio million in 1918 and 1919 to 

 32 million in 1920, after which Smelt hatching was given up.^' A general decrease 

 seems also to have taken place along the coast of Maine since the end of the last century, 

 and from the same causes that produced the depletion southward, i. e. from contraction 

 of favorable spawning grounds by dams close to the mouth of streams, and perhaps 

 from overfishing. In 1889, for example (J5: 79), United States waters yielded 1,209,385 

 pounds of American Smelts in contrast to a yearly average of only 473,550 pounds 

 for the period 1 951— 1954. But the catch records do not suggest any extensive or 



12. For a recent survey of variations in the taxonomic features of fishes in general, as associated with regional dif- 

 ferences in their environment, see Vladykov {113: 122). 



13. For details and for general information as to the smelt, see 84. 



