these two lakes clustered together in canonical analyses (Fig. 1, Copeman, 1977) In closer 

 proximity to other rainbow smelts from Lake Huron and Lake Heney. Divergence of Introduced 

 populations after more than 60 years was equivalent to that between other natural 

 populations. But the changes resulting from direct environmental differences during 

 development and from selection was not such as to cluster any population with the pygmy 

 smelt . 



Similarly Copeman and McAllister (1978) interpreted the results of smelt transfer 

 experiments in Maine to show that, although growth may be initially accelerated or prolonged 

 in a population transplanted into a previously rotenoned lake, that in a second phase these 

 effects subside and growth characteristics return to those of the source populations. 

 Copeman and McAllister (1978) found that gill raker and vertebral counts were not 

 significantly changed between source and host lakes in the initial or secondary phases of 

 transfer experiments. 



Pygmy smelt from Utopia Lake, New Brunswick, were transplanted to Meach Lake, Gatineau 

 County, Quebec about 1924 (Dymond, 1939). A collection of 14 specimens made in Meach Lake 

 on 8 May 1963 (NMC71-0651) was located just before the present manuscript was completed. 

 Chi-square tests showed that neither gill raker nor vertebral counts were significantly 

 different (p more than 0.30) in samples from the two lakes, despite climatic differences and 

 the passage of nearly 40 years. It may be noted that gill rakers were shown in Copeman's 

 (1977) and the present study to be the best merlstic diagnostic character. 



DISCUSSION 



Taxonomy 



In this discussion we shall attempt to answer whether the pygmy smelt is a species 

 distinct from the rainbow smelt and what scientific name should be applied to the pygmy 

 smelt. Species are understood here to consist of a population or populations of actually or 

 potentially interbreeding organisms which are reproductively isolated from other such groups 

 and are usually morphologically distinct. 



Morphological evidence from Delisle (1969a), Copeman (1973, 1977), and the present study 

 indicate that pygmy and rainbow smelts are morphologically distinguishable. Transplantation 

 experiments cited demonstrate that the number of gill rakers and vertebrae have a genetic 

 basis and are little affected by transplantation to environments very different in 

 temperature, productivity and other factors. Growth under equilibrium conditions of lakes 

 (second phase in reclamations), appears to exhibit an underlying genetic basis. 



In Heney, Green and Utopia lakes evidence has been cited that pygmy and rainbow smelts 

 spawn at different times and that the spawning periods do not overlap in sympatric 

 populations. Rupp and Redmond's (1966) data on transplants in Maine indicate that spawning 

 time is under genetic control and is not significantly influenced by transfer to another 

 lake. A transplanted population in Meach Lake, Quebec, spawned at essentially the same time 

 (8-19 May) as the source population in Utopia Lake, New Brunswick (12-22 May). 



21 



