DUTIL AND COUTU: EARLY LIFE OF ATLANTIC SALMON 



Table 9. — Relative abundance in terms of catch per unit of eHort 

 (CPUE) at 2 stations run in 1983 and 1984 using similar nets at tfie 

 same sites. 



'CPUE: catch of 1 net in 24 hours. 



similar nets to 1983 and 1984. The catch was 

 nearly nonexistent: 5 salmon for 120 unit effort 

 ( 1 unit effort, is 1 net x 24 hours). Thus, based on 

 limited observations, numbers of salmon near 

 shore in summer and autumn seem to be highly 

 variable from year to year. 



DISCUSSION 



Postsmolts of Atlantic salmon stay much longer 

 near our coasts than is usually believed. Though 

 early months in the marine environment have 

 been shown to have a marked influence on 

 salmon runs 1 and 2 years later (e.g., Christensen 

 and Larsson 1979; Scarnecchia 1983, 1984), 

 postsmolt biology has been a neglected area of 

 investigation. North American smolts are as- 

 sumed to migrate rapidly out of the estuaries of 

 their home rivers, to feeding areas located far out 

 in the North Atlantic east of the Grand Bank 

 (Templeman 1968; Reddin 1985) and north to 

 Labrador and Greenland (Saunders 1966; Tem- 

 pleman 1967). They return 1 or 2 years later to 

 spawn in home rivers. This study, and early 

 records, indicate that some postsmolts remain in 

 coastal areas as late as autumn before moving 

 offshore. This was clearly suggested by Comeau 

 (1909) who stated that 0.5-1.5 lb postsmolts were 

 regularly taken in autumn along the north shore 

 of the Gulf of St. Lawrence. Fishermen inter- 

 viewed in 1982 on the north shore of the Gulf of 

 St. Lawrence, from Pointe-des-Monts to Blanc 

 Sablon, declared incidental catches of postsmolts, 

 mainly in the months of August and September. 

 The majority declared taking postsmolts each 

 year. Smolts were also shown to linger in estuar- 

 ies of the north shore of the Gulf of St. Lawrence 

 (this study; Power and Shooner 1966; Randall and 

 Power 1979). Huntsman (1939) mentioned their 



occurrence in autumn in herring nets near the 

 mouth of Gaspe Bay. Fall catches also occurred in 

 New England (Kendall 1935). Recently, smolts 

 released in New England were caught as post- 

 smolts in coastal areas of Canada (Meister 1984). 

 There are stocks in the Bay of Fundy (as the 

 stocks in the Baltic) that do not go on extensive 

 migrations in the North Atlantic (Huntsman 

 1939; Jessop 1976): postsmolts of these stocks are 

 regularly taken in herring nets in Passa- 

 maquoddy Bay and off Grand Manan Island (El- 

 son 1953, 1964; Allen et al. 1972). Hence the pres- 

 ence of postsmolts near shore in autumn (or in 

 summer and autumn) is a characteristic of the 

 marine life of Atlantic salmon in North America. 

 Timing of migration has been described for 

 hatchery-reared smolts released in Sweden 

 (Larsson 1974) and Finland (Jutila and Alapassi 

 1985). Tags were returned mostly from a distance 

 of less than 10 km between days and 10, 20-50 

 km between days 10 and 20, and 50-100 km 2 

 months past their release, in the brackish waters 

 of the Gulf of Bothnia (Jutila and Alapassi 1985). 

 Behavior of postsmolts is similar to that of 1- 

 and 2-sea-year salmon. Postsmolts in this study 

 occurred mainly near the surface as indicated by 

 their distribution in the nets. LaBar et al. (1978) 

 concluded that smolts migrated near the surface 

 in the Penobscot estuary. Templeman (1967, 

 1968) also found salmon to occur near the surface 

 in the Northwest Atlantic: 62% occurred in the 

 top 0.6 m and 90% in the top 1.5 m in July and 

 August 1965. Similarly in 1966, most salmon 

 were taken in the top 1.5 m, the number caught 

 decreasing sharply below 0.6 m. Similar observa- 

 tions were made on Baltic salmon (Carlin and 

 Lundin 1967; Christensen 1968). There is less in- 

 formation available on schooling. Postsmolts did 

 not regularly have a clustered distribution in 

 nets, but considering that a net does not retain all 

 the salmon striking it, there were still many in- 

 stances of salmon moving in schools. Thurow 

 (1968) came to the same conclusion for older 

 salmon in the Baltic. Templeman (1967) pre- 

 sented limited evidence for salmon in the North- 

 west Atlantic, but reached negative conclusions 

 later (Templeman 1968). Finally there are lim- 

 ited data in the literature concerning the rhythm 

 of activity of salmon in the marine environment. 

 Christensen and Lear (1980) showed that in West 

 Greenland best catches occurred early in the 

 morning (0600-0800), decreased sharply between 

 0800 and 1000, and were low between 1000 and 

 1400. The nets were not set between 1400 and 



207 



