populations be protected from pollution including acid precipitation, habitat deterioration, 

 overfishing, and introduction of competitors. It deserves protection because of its rarity 

 in its own right as a species and because of its value as a forage fish for sport and 

 commercial fishes for introduction Into certain other lakes. A survey of specimens is under 

 way by Sylvie Laframboise and the second author to determine if the pygmy smelt occurs in 

 other lakes . 



ORIGIN OF PYGMY SMELT 



We find it interesting to speculate on the origin of the pygmy smelt. Given the 

 widespread occurrence of anadromous populations of the rainbow smelt in the Atlantic, 

 Pacific and Arctic waters, it seems most likely to us that Osmerus spectrum with its 

 restricted lacustrine range, arose from Osmerus mordax . The close morphological and 

 biochemical homogeneity (Copeman, 1977 and unpublished studies by Larry Speirs) of the two 

 species, does not call for an origin earlier than the Pleistocene or perhaps even the 

 Wisconsin. 



We hypothesize that Osmerus spectrum arose in lakes close to the edge of a Wisconsin or 

 an earlier ice sheet, perhaps during a glacial maximum, on the Atlantic seaboard. Or it 

 might have evolved in one of the proglacial lakes during the Wisconsin glacial retreat, e.g. 

 Glacial Lake Hitchcock in the valley of the Connecticut River which may have endured 4,100 

 years starting not long before 13,000 BP (Schafer and Hartshorn, 1965). However we 

 personally feel that the level of differentiation, greater than that of the Newfoundland 

 dwarf, suggests an age greater than 15,000 years and an orgin going back to about the raid or 

 early Wisconsin. 



Allopatric isolation in a plankton-rich but fish-depauperate lake in the presence of a 

 vacant niche would have favoured retention of the planktivorous over-development of the 

 piscivorous habit as an adult. Or conceivably the pygmy smelt might have evolved 

 sympatrically through interruption of the spawning run by a natural disaster such as a 

 glacial clay land-slide, mutation for a later spawning period, or through selection acting 

 on stocks homing to different spawning grounds in the same lake. 



Neoteny with sexual maturity at an earlier age and smaller size while still 

 planktivorous, may have occurred. Selection for more gill rakers in association with 

 prolonged planktivory, and fewer scales and vertebrae because of smaller maximum size in 

 accordance with Lindsey's pleomeristic rule (Lindsey, 1975) may have occurred in initial or 

 secondary sympatry. 



Dispersal may have occurred northwards in proglacial lakes or in drainages fed by melting 

 glaciers through New England to southern New Brunswick, with southern populations being 

 extirpated as lakes warmed. Deglaciation of lake basins known to presently contain pygmy 

 smelt in Maine and New Brunswick took place between 13,000 and 14,000 BP (Geological Survey 

 of Canada Map 1257A. Retreat of Wisconsin and Recent ice in North America). 



Populations in southern Quebec can be no older than deglaciation estimated (Map 1257A. 

 op. cit.) at 10,800 to 11,500 BP. The glacial history here is more complicated 



26 



