A review of ecological and behavioural interactions between cultured and wild Atlantic salmon

ICES Journal of Marine Science, 54: 1031 1039. 1997 A review of ecological and behavioural interactions between cultured and wild Atlantic salmon B. Jonsson Jonsson, B. 1997. A review of ecological and behavioural interactions between cultured and wild Atlantic salmon. ICES Journal of Marine Science, 54: 1031 1039. Cultured Atlantic salmon (Salmo salar L.) may be introduced into natural systems intentionally or accidentally. As smolts or post-smolts, they move to the feeding areas of wild salmon in the North Atlantic Ocean. As maturing fish, they return to the area of release and enter rivers to spawn. Lack of juvenile river experience is the prime reason why cultured salmon often enter fresh water later in the season than wild fish. During spawning, cultured female salmon from fish farms make fewer nests, tend to breed for a shorter period of time, are poorer at nest covering, and retain greater amounts of unspawned eggs than wild females. Cultured male salmon from fish farms exhibit less combat and display behaviour, have greater difficulty in acquiring access to mates, show less quivering and courting behaviour, and have lower reproductive success than wild males. However, cultured male salmon are more involved in prolonged, reciprocal fights than wild males and are, therefore, more often wounded. The reproductive success of cultured salmon increases with the time the fish have lived in nature before maturing sexually; for cultured females released in nature at the smolt stage, reproductive success is similar to that of wild females. The relative reproductive success of cultured males is smaller than that of corresponding females. Within both sexes of cultured and wild salmon, competitive spawning ability increases with body size. As a phenotypic response to increased growth rate during the first year of life, cultured salmon tend to have smaller sized but more numerous eggs than wild fish of the same size. Offspring of cultured salmon are more generally aggressive, more risk prone, and have a higher growth rate than wild offspring. Consequently, their survival rate in nature may be lower. 1997 International Council for the Exploration of the Sea Key words: cultured, dispersal, hatchery, migration, reproductive success, salmonids, Salmo salar, spawning survival. B. Jonsson: Norwegian Institute for Nature Research, Dronningens gt 13, PO Box 736 Sentrum, N-0105 Oslo, Norway. Introduction Cultured salmon in nature have a varying degree of impact on wild conspecifics (Allendorf, 1983). These impacts may be mediated through ecological competition (Fleming and Gross, 1993), genetic introgression (loss of local adaptation and genetic homogenization) (Hindar et al., 1991), or the spread of parasites and infectious diseases (Egidius et al., 1991). In Atlantic salmon (Salmo salar L.), the interactions may occur at juvenile and adult stages in freshwater and marine environments. Cultured salmon are introduced into natural systems with or without intent. They are released as alevins, parr, or smolts to enhance natural stocks or for the development of a culture-based fishery (e.g. river cultivation and sea ranching) (Jonsson and Fleming, 1993). Moreover, cultured salmon are artificially reared from broodstock in freshwater hatcheries for 6 months to 1 year before being transferred to marine net pens where they are reared until harvest. The fish can escape at any time during culture, either in fresh water or at sea. The abundance of cultured salmon in nature is large. For example, between 25 and 40% of the salmon in the North Atlantic Ocean (Hansen et al., 1993a), and more than 90% of the salmon in the Baltic Sea (Jonsson and Fleming, 1993) are cultured fish. Similar high figures have been reported from some Norwegian salmon rivers (Lund et al., 1996). Furthermore, because of the transportation of cultured salmon over large areas, intentional and unintentional releases of cultured fish have resulted in a mixture of individuals from different populations to an extent which has never occurred before (Hindar and Jonsson, 1995). Owing to broodstock selection (e.g. of large, fastdeveloping fish from specific stocks) and artificial 1054 3139/97/061031+09 $25.00/0/jm970302 1997 International Council for the Exploration of the Sea

1032 B. Jonsson conditions during rearing (substratum, temperature, and feeding regimes, current velocity, lack of predators, fish density), cultured salmon diverge rapidly from their wild origin through environmental and evolutionary processes (Cross and Challanain, 1991; Youngson et al., 1991; Fleming et al., 1994). Moreover, cultured salmon often have a non-local origin. Therefore, when released in nature, they pose a risk of impacts on their wild conspecifics through ecological interactions (Heggberget et al., 1993a). This review summarizes knowledge on ecological and behavioural interactions between cultured and wild Atlantic salmon in nature: dispersal, migration, and spawning of cultured salmon; behavioural interference between cultured and wild spawners in rivers, and the consequences of these interactions for their reproductive success; observations on fecundity and survival; and evidence for interactions between the young in rivers. Dispersal and migration Migration to sea The distribution pattern of cultured Atlantic salmon depends on their developmental stage and the time of the year when they are released or escape. Smolts or post-smolts released in Norway move in the same general direction as the prevailing water currents along the coast for some time before turning out into the ocean, where they enter the feeding areas of wild salmon in the North Atlantic (Jonsson et al., 1993). The mean migratory speed of post-smolts along the Norwegian coast has been estimated at 7.5 km d 1 (Jonsson et al., 1993). Sexually maturing cultured salmon which have escaped are more inclined to stay in coastal areas than juvenile escapees, although some also move to sea to feed in the North Atlantic (Hansen et al., 1987). In the ocean, cultured fish supplement the stock of wild salmon and may be exploited there and in coastal waters (Hansen et al., 1993a). Few cultured salmon, however, contribute to the fishery at West Greenland (Hansen et al., 1997). Cultured salmon may return to fresh water within 1 year of release, or they may remain at sea for 1 or more years before they enter rivers to spawn (Jonsson et al., 1990a). Homing and river ascent Immature cultured salmon which escape to the sea in spring, summer, or autumn appear to return to the area of release when sexually mature (Hansen and Jonsson, 1991). Those which escape in winter, on the other hand, do not return to any specific area when sexually mature. Salmon which escape as maturing fish in their second summer appear to have lost the ability to return to their home river or place of release; they enter rivers to spawn over a large area (Hansen et al., 1987). Thus, there is a finite period when the fish are able to pick up navigational cues that they use during the return migration sub-adult and adult salmon have lost this ability (Hansen et al., 1993b; Hansen and Jonsson, 1994). Escaped, cultured salmon enter rivers to spawn (Lund et al., 1991; Webb et al., 1991). The time of river ascent of cultured salmon is influenced by whether or not they have juvenile experience of river life. Cultured salmon released in fresh water as parr or smolts will usually enter the river of release earlier than those released at the river mouth, although the time of arrival in fjords is similar between these two groups (Jonsson et al., 1990a, 1994). When smolts and post-smolts are released at the coast, they return as adults to the area of release and enter nearby rivers to spawn, rather than returning to their natal river (Hansen et al., 1989, 1993b). Salmon released as post-smolts in the open ocean north of the Faroe Islands have been recaptured as maturing fish on the Norwegian coast. Hansen et al. (1993b) therefore suggested that cultured salmon may possess an innate, general sense of direction, helping them to move towards their home continent. They have, however, no inborn preference for their river or area of origin. Lack of juvenile river experience is probably the prime reason why cultured salmon on their spawning migration tend to enter fresh water later in the season than the wild salmon (Jonsson et al., 1990a). Although hatcheryreared fish may migrate thousands of kilometres in the ocean before returning to fresh water for spawning, a difference of only a few hundred metres in the location of release as smolts influences the timing of upstream migration, as shown by experiments in the River Imsa (Jonsson et al., 1994). There, sea-ranched, cultured salmon which lack river experience released at the river estuary, ascend on average 17 d later than corresponding fish released 800 m upstream in the river and wild fish utilizing the same river. This pattern is independent of sex. The time of river ascent of adult salmon in the River Imsa is negatively correlated with body size (Jonsson et al., 1990a, b; Fleming et al., 1997). The reason is probably that large salmon are delayed in entering the river by low water flows early in the season (Jonsson et al., 1990b). It is not known if the difference in pattern of river ascent of salmon with and without prior river experience is linked to the geography of the site of release, or their lack of juvenile experience from running waters in general. River experience does not influence the ability of the adults to locate the river on their return migration in the sea. Studies in the River Imsa have shown that straying rates to other rivers are similar for cultured smolts released at the river mouth or 800 m upstream (Jonsson et al., 1994). Similarly, the duration of the juvenile river residency has little influence on the ability of the fish to

Interactions between cultured and wild salmon 1033 locate their river as adults. Although cultured smolts migrate to sea within hours of release, they return with similar precision to the river of release as wild smolts which have lived in the river for 1 to 3 years (Jonsson et al., 1994). When smolts or post-smolts are released in fjords a few metres from a river mouth, however, the rate of straying to other rivers greatly increases (Hansen and Jonsson, 1991; Hansen et al., 1993b; Jonsson et al., 1991a). In river movement and descent In rivers, cultured spawners ascend to the spawning grounds as rapidly as wild salmon (Heggberget et al., 1993b; Økland et al., 1995), and they appear to exhibit the same preference for spawning areas as wild fish. Trapping studies in the River Imsa have shown that cultured salmon are less stationary in the spawning area than wild salmon (Jonsson et al., 1990a, 1991a). Similarly, radio telemetry data from the River Alta indicate that cultured females which have escaped from fish farms are less stationary in a particular part of the river than wild conspecifics (Økland et al., 1995). This difference may be a consequence of their later arrival or their lack of previous river experience (Jonsson et al., 1990a), or their competitive inferiority to wild conspecifics (Fleming et al., 1996). Cultured salmon leave the river earlier after spawning than wild conspecifics, and the tendency to spend the winter after spawning in fresh water is less pronounced in cultured fish than wild fish. Cultured males in particular return rapidly to the sea, and a substantial proportion may leave the river without having spawned (Jonsson et al., 1990a). Cultured fish, and males in particular, appear to be more seriously wounded during spawning than wild fish, even when the two groups originate from the same population (Jonsson et al., 1990a). This may be because cultured males are more involved in prolonged fights during spawning than wild salmon (Fleming et al., 1997). Spawning in nature Spawning of cultured salmon of both sexes has been verified through visual inspection and by detection of synthetic astaxanthin in eggs following excavation of nests (Lura and Sægrov, 1991a, b; Webb et al., 1991, 1993a, b; Lura and Økland, 1994). Investigations by Lura et al. (1993) indicated that cultured salmon deposited fewer eggs per nest than wild fish, perhaps because the cultured fish in this study were smaller than the local wild salmon (cf. Fleming et al., 1996). Lura et al. (1993) found no other difference between the nests of cultured and wild salmon in rivers and concluded that the basic components of spawning behaviour are maintained in cultured female Atlantic salmon. Time of spawning of cultured and wild Atlantic salmon was studied in the Norwegian River Vosso (Lura and Sægrov, 1993). Although cultured salmon usually enter streams after the wild fish, they found that the cultured salmon spawned more than 20 d earlier than the wild salmon present, resulting in the offspring of cultured salmon hatching approximately 10 d before those of wild fish. The consequences of this early breeding of cultured fish are unpredictable but may reduce the reproductive success. However, these studies of salmon in natural rivers do not provide detailed information of the spawning behaviour of cultured salmon. Reproductive behaviour and success The experiments Fleming et al. (1996) studied the spawning behaviour of cultured and wild Atlantic salmon under semi-natural conditions in 47 m 2 circular outdoor tanks which provided stream environments that simulated natural breeding conditions. Fifth-generation cultured fish from the breeding programme at Sundalsøra, Norway, were compared with wild fish from the River Imsa. Sundalsøra salmon are widely cultured both in Norway, where they represent over 80% of salmon used in the fish farming industry, and in Great Britain. Wild Atlantic salmon were trapped 100 m above the estuary of the River Imsa during their spawning migration. The Sundalsøra salmon had been in culture all their life. Preparation for spawning When held in mixed groups, wild and cultured females showed similar levels of aggressive and submissive behaviour. Aggressive behaviour included chasing (attack of another fish without reciprocation), fighting (reciprocal attacks between two fish) and agonistic display (aggressive posturing towards another fish). Submissive behaviour is the fleeing from an aggressive opponent. Female aggression towards males was directed more often at wild than at cultured individuals. Cultured females were less active than wild females in cruising (swimming over an area without chasing or being pursued) and digging (a series of body flexures while turned on one side on the streambed, usually associated with nest construction), and they were courted (attended by a male at the nest) less often than wild females, presumably reflecting their lower levels of digging and associated nesting activity. Cultured males were less aggressive and exhibited less combat and display behaviour than wild males. A similar difference has been observed between cultured and wild coho salmon (Fleming and Gross, 1992, 1993), and

1034 B. Jonsson may have both a genetic and an environmental origin. Male aggression was directed more frequently at wild rather than cultured males. Cultured males exhibited no difference in their aggression towards cultured and wild females, whereas wild males were more aggressive to wild rather than cultured females. Cultured males had difficulty acquiring access to mates, showing less quivering (vibration of the body next to a female) and courting behaviour than wild males. There were only minor differences in behaviours displayed by wild males and females competing with cultured fish and wild fish competing only among themselves. A part of the lower competitive ability of cultured males may be due to morphological changes occurring during hatchery rearing, such as a less pronounced kype (hook on the lower jaw), which male salmon use during fights (Fleming et al., 1994; Fleming, 1996). In the absence of wild salmon, cultured females: exhibited reduced levels of aggression, particularly combat behaviour; displayed less digging behaviour; were courted less often; and cruised more frequently than when wild salmon were present. The low female activity may at least partly be a consequence of the low male activity. The behaviour of cultured males was similar in the presence or absence of wild males. Spawning In mixed groups of cultured and wild salmon, cultured females made fewer nests and there was a tendency for cultured females to breed for a slightly shorter period of time than wild females. The majority of females constructed all their nests within a single redd. Cultured females constructed fewer nests than wild females, but the total number of eggs recovered per nest was similar. Cultured females dug less frequently during the first 5 min following oviposition and took longer to cover their eggs. There were no significant differences between cultured and wild females in the depth of nests constructed or the gravel quality used for spawning. Cultured females retained a significantly greater weight of unspawned eggs than wild females. This may be because of their poorer physical condition with weaker muscles and a poorer ability to extrude all eggs during spawning. Nest destruction, as a result of overcutting by other females, also occurred more frequently to nests constructed by cultured than wild females. Survival of eggs, which included the effects of poor fertilization and mortality, of cultured females was significantly lower than for those of wild females. As a consequence of greater egg retention and poorer survival of deposited eggs, the survival of eggs of cultured females was less than one-third that of wild females. Cultured females made fewer nests in the absence of wild salmon than in the presence of wild salmon, but more eggs were deposited in each nest. The reproductive success of cultured females in the experiments was dramatically reduced in the absence of wild fish. Egg retention increased and fewer of the eggs spawned survived, as the fertilization rate was low. In the absence of wild males, only 10% of the nests constructed by cultured females contained eggs with live embryos, while 98% of the nests contained live embryos when wild males were present. In the absence of wild fish, the reproductive success of cultured females was only onetenth of what it was in the presence of wild fish. Thus, the presence of active, wild males appeared to stimulate the spawning activity of cultured females. In the competition experiments with cultured and wild males, cultured males took longer to the onset of breeding, but spawned for a shorter period of time. For wild males, neither the onset nor the duration of spawning activity was influenced by the presence of cultured salmon. Wild males did, however, take part in more spawnings and had higher reproductive success when cultured fish were present than when held in groups comprised solely of wild fish. The breeding and reproductive success of cultured males were not influenced by the presence of wild males. Even in the absence of wild males, cultured males were hardly able to spawn. Courting cultured males approached the nest and released sperm when the female oviposited during only two out of six spawnings. This behaviour may explain Lura and Sægrov s (1991a) observation that eggs in several nests spawned by escaped cultured female Atlantic salmon were unfertilized. Competitive ability In salmon, body size is a principal determinant of the expression of competitive and breeding behaviours in both males and females. This was clearly shown in competition experiments between wild and sea-ranched salmon (cultured fish released as smolts and sampled as maturing adults) (Fleming et al., 1997). Larger individuals were more aggressive and more active at courting (males) and being courted (females), while smaller individuals were more likely to be submissive. Large and small individuals, however, used the two forms of aggressive behaviour, combat and display, in similar proportions. In terms of combat, chase was most often directed towards small individuals, while fighting usually involved pairs of males of similar size. Males are more aggressive than females, both in terms of combat and display behaviour. Furthermore, compared to females proportionately more of the aggressive behaviour shown by males involved displays. Males directed little of their aggression towards females. Females, on the other hand, directed a large part of their aggression towards males. Both males and females were

Interactions between cultured and wild salmon 1035 more aggressive towards smaller individuals of the opposite sex. In contrast to migratory behaviour, differences in early experience do not influence competitive behaviour, except that fish with hatchery experience tend to be more involved in prolonged, reciprocal fights than wild males. In terms of the breeding behaviour, there is no difference in frequency of digging or courting behaviours, and no indication of assortative mating by fish type (wild or cultured). The behaviours of the sea-ranched fish were not significantly altered by competition with wild fish, with the exception of male courting. Sea-ranched males were less frequently able to be primary courters in the presence of wild males. This reflected a tendency for wild males to dominate the primary courting position from which they could obtain better access to spawning females and thus improve their fertilization success during inter-group competition. More of the spawnings by wild males were solo, involving only a single male, whereas those of searanched males often involved two or more males. Wild males therefore achieved higher reproductive success than sea-ranched males during inter-group competition. Female reproductive success was similar for sea-ranched and wild females. To the extent that cross-breeding between cultured and wild salmon occurs, hybridization between wild males and cultured females takes place more often than between cultured males and wild females. The reason for this is the relatively higher reproductive success of cultured females than males compared with corresponding wild fish, and the consequence is a greater gene flow from the cultured females than from the cultured males to the wild population. The reproductive success of cultured salmon increases with the time the fish live in the natural environment before maturing sexually (Fleming et al., 1997). The prime reasons for this appear to be improved physical condition and a morphological appearance which closely resembles wild fish. Most of the phenotypic differences between cultured and wild fish disappear after 1 year in nature (Fleming et al., 1994). The reduced reproductive success of cultured fish may also be due to domestication selection and selective breeding in the hatcheries. Their reduced aggression may be an evolutionary response to generations of artificial breeding, where the costs of aggression are not offset by any reproductive advantage since breeding opportunities are determined artificially (Fleming and Gross, 1992, 1993). Egg quality The reproductive success of salmon in nature is influenced by the quality of the eggs (and sperm). Cultured salmon have smaller but higher numbers of eggs per unit weight than wild fish (Thorpe et al., 1984). This is probably a phenotypically plastic response to different development rates during the first year of life (Jonsson et al., 1996). The more rapidly the young fish grow during their first summer in fresh water, the smaller the eggs they will produce as adults, producing a higher relative fecundity. This phenotypic variation in egg size probably influences the future success of the offspring. Smaller eggs may contain less yolk, and the effect may be smaller juveniles (Glebe et al., 1979; Kazakov, 1981) with lower juvenile growth rate (Bagenal, 1969) and reduced susceptibility to starvation (Hutchings, 1991), which may be a major problem under keen juvenile competition for food. On the other hand, small eggs may survive better than large eggs under low O 2 conditions (van den Berghe and Gross, 1989). The variation in egg size and numbers both with juvenile growth rate and size of the mother may indicate that the gamete production is adapted to spawning conditions (e.g. large adults may spawn in coarser gravel and more rapidly flowing water with better oxygen conditions than smaller adults) (Fleming and Gross, 1990; Quinn et al., 1995) and the richness of juvenile feeding conditions (Jonsson et al., 1996). Competitive ability of the offspring Cultured Atlantic salmon parr released in rivers have similar feeding and territorial behaviour, at least within 1 2 months after release (Shustov et al., 1980), and parr releases are used to enhance natural populations (review in Mills, 1989). However, the success of naturally spawned offspring of cultured salmon may be lower than that of wild conspecifics. Reisenbichler and McIntyre (1977) observed lower survival in the offspring of cultured steelhead trout (Oncorhynchus mykiss) and cultured/wild trout hybrids than in wild offspring after two generations of cultivation. Moreover, Chilicote et al. (1986) found that cultured steelhead had only 28% of the success of wild steelhead trout in producing smolt offspring. Experiments with artificially spawned offspring of Pacific salmon (Oncorhynchus spp.), however, indicate that cultured fish may have competitive advantages over wild conspecifics by being more aggressive in competitive contests (Chandler and Bjornn, 1988; Swain and Riddell, 1990, but see Berejikian et al., 1996), so they can contribute to the displacement of wild fish in nature (Nickelson et al., 1986). Holm and Fernö (1986) provided experimental evidence that the aggression among cultured Atlantic salmon parr was higher than that of wild parr. Moreover, Einum and Fleming (1997) undertook a series of experiments, both in the hatchery and in the natural

1036 B. Jonsson environment, to examine interactions between cultured, native and hybrid 0-group parr derived from controlled crosses and reared under common conditions. The cultured salmon were seventh-generation fish from the Sundalsøra strain, and the native salmon were from the Rivers Imsa and Lone, Norway. In the hatchery, cultured salmon were more aggressive than both native populations and tended to dominate them in pairwise contests. Cultured salmon were more risk prone, leaving cover sooner after a simulated predator attack, and had higher growth rates than native fish. Cultured wild hybrids showed intermediate expression of the above traits. However, there was evidence of hybrid vigour in the Lone/cultured crosses which were able to dominate both pure Lone and cultured parr in pairwise contests. In the wild, observations of habitat use and diet suggested that the populations competed for territory and food, and both cultured fish and hybrids had higher growth rates than native fish. Similarly, Johnsson et al. (1996) found that hatchery selection increased the competitive ability of brown trout (Salmo trutta). These results suggest that innate differences in behaviour and growth, which are closely linked to fitness, pose a threat to native populations through superiority in competition for food. Swain and Riddell (1990) demonstrated that newly emerged coho salmon (Oncorhynchus kisutch) fry from populations that had been cultured for several generations were more aggressive than fry from geographically close wild populations. Experiments with fry of steelhead trout (Oncorhynchus mykiss) which had been reared in captivity for four to seven generations, on the other hand, indicated that the aggression may be reduced with domestication in newly emerged fry, but not in older fry (Berejikian et al., 1996). Newly emerged cultured steelhead trout required a size advantage of about 4% to dominate wild competitors. Thus, the problem of the offspring of cultured fish dominating the wild fish may be reduced with number of generations in culture. The reason for different aggressiveness between cultured and wild parr may be unintentional artificial selection (selection of particularly aggressive fish as broodstock) or natural selection to the domestic environment in the hatchery (Ruzzante, 1994). However, increased aggressiveness may not give a growth advantage when food is in excess and antagonistic interactions are not rewarded with better access to the food (Ruzzante and Doyle, 1990). If the agonistic behaviour pays off through better access to food, this may result in a growth advantage among the competitors (Metcalfe et al., 1989; Huntingford et al., 1990; Metcalfe and Thorpe, 1992). On the other hand, aggression has an energetic cost, and Holm and Fernö (1986) found that the growth rate was lower in groups of the more aggressive salmon parr. Offspring of cultured salmon may thus have a negative competitive impact on the offspring of wild salmon, although the success of the cultured offspring probably is lower than that of wild offspring in natural rivers. Survival In hatcheries, salmon have increased survival relative to conspecifics in nature. For example, the freshwater survival of cultured fish to the smolt stage is approximately 20 times higher than that of the wild fish (Jonsson and Fleming, 1993). This situation changes when cultured fish are released in nature. Cultured smolts released in the spring have about half the survival from smolt to adult as wild salmon (Jonsson et al., 1991a). For smolts and post-smolts, the survival is greatest for those escaping in spring (between April and June) and gradually decreases for those released in summer, autumn, and winter (Hansen and Jonsson, 1989). Higher mortality of the offspring of cultured salmon may be a consequence of their willingness to take greater risks. Experiments with juveniles of steelhead trout and steelhead domesticated rainbow trout hybrids showed that the hybrids were more risk-prone and exposed themselves more to predators (Johnsson and Abrahams, 1991). In the hatchery environment, there is no selection by predators against risky foraging behaviour and excessive aggressiveness. In brown trout, Johnsson et al. (1996) found that cultured fish had a less pronounced anti-predator response than wild conspecifics. A similar change in anti-predator behaviour is induced by increasing the level of growth hormone. Under natural conditions cultured fish, with or without manipulated growth hormone levels, may therefore have reduced survival relative to wild fish. The survival of the offspring of cultured fish in competition with wild progeny was tested in a large-scale release of genetically marked first-generation cultured brown trout. Systematically, the brown trout is the nearest relative of the Atlantic salmon, and the two exhibit similar mating behaviour (Jones and Ball, 1954; Fleming, 1996). Cultured spawners were liberated in the spawning grounds of two natural populations in the River Øyreselv, Norway (Skaala et al., 1996). The cultured trout spawned with other cultured trout and with indigenous anadromous and freshwater resident trout. The F 1 generation was sampled over the next three years. At age-0 the offspring of the cultured trout constituted 16% and 19% of the two populations, respectively. The estimated survival-rate of the wild fish was three times higher than cultured fish, and consequently the frequency of the genetically marked alleles in the F 1 generation gradually declined. There is

Interactions between cultured and wild salmon 1037 no information on the performance of the F 1 generation beyond age 2. Based on this finding, higher survival of offspring of wild fish compared to cultured salmon should be expected in nature, as has been found for steelhead trout (Leider et al., 1990). On the other hand, the offspring of cultured salmon may contribute to the displacement of wild salmon. Conclusions (1) Escaped cultured salmon enter rivers to spawn, but this usually occurs later in the season than for wild fish owing to the lack of juvenile river experience. (2) Cultured salmon produce offspring in nature, but their reproductive success is generally lower than that of similar sized wild salmon. This is most pronounced in males. The reproductive success is higher in the presence of wild fish. (3) The reproductive success is lower for cultured salmon escaping at late life history stages than for those escaping at an early stage. (4) Reproductive success increases with body size in both sexes of cultured and wild salmon. (5) The success of the offspring of cultured salmon is uncertain. Competitive ability and growth rate may be higher, but the survival rate is probably lower than in wild fish due to higher risk taking. The outcome of competitive interactions in nature will depend on local conditions. Acknowledgements I am grateful to N. B. Metcalfe, J. E. Thorpe, and an anonymous referee for helpful comments on the paper. References Allendorf, F. W. 1983. Conservation biology of fishes. Conservation Biology, 2: 145 148. Bagenal, T. B. 1969. Relationship between egg size and fry survival in brown trout, Salmo salar L. Journal of Fish Biology, 1: 349 353. Berejikian, B. A., Mathews, S. B., and Quinn, T. P. 1996. Effects of hatchery and wild ancestry and rearing environments on the development of agonistic behavior in steelhead trout (Oncorhynchus mykiss). Canadian Journal of Fisheries and Aquatic Sciences, 53: 2004 2014. Chandler, G. L. and Bjornn, T. C. 1988. Abundance, growth, and interaction of juvenile steelhead relative to time of emergence. Transactions of the American Fisheries Society, 117: 432 443. Chilcote, M. W., Leider, S. A., and Loch, J. J. 1986. Differential reproductive success of hatchery and wild summer-run steelhead under natural conditions. Transactions of the American Fisheries Society, 115: 726 735. Cross, T. F. and Challanain, D. N. 1991. Genetic characterisation of Atlantic salmon (Salmo salar) lines farmed in Ireland. Aquaculture, 98: 209 216. Egidius, E., Hansen, L. P., Jonsson, B., and Nævdal, G. 1991. Mutual impact of wild and cultured Atlantic salmon in Norway. Journal du Conseil International pour l Exploration de la Mer, 47: 404 410. Einum, S. and Fleming, I. A. 1997. Genetic divergence and interactions in the wild among native, farmed and hybrid Atlantic salmon. Journal of Fish Biology, 50: 634 651. Fleming, I. A. 1996. Reproductive strategies of Atlantic salmon: ecology and evolution. Reviews in Fish Biology and Fisheries, 6: 379 416. Fleming, I. A. and Gross, M. R. 1990. Latitudinal clines: a trade-off between egg number and size in Pacific salmon. Ecology, 71: 1 11. Fleming, I. A. and Gross, M. R. 1992. Reproductive behavior of hatchery and wild coho salmon (Oncorhynchus kisutch): does it differ? Aquaculture, 103: 101 121. Fleming, I. A. and Gross, M. R. 1993. Breeding success of hatchery and wild coho salmon (Oncorhynchus kisutch) in competition. Ecological Applications, 3: 230 245. Fleming, I. A., Jonsson, B., and Gross, M. R. 1994. Phenotypic divergence of sea-ranched, farmed and wild salmon. Canadian Journal of Fisheries and Aquatic Sciences, 51: 2808 2824. Fleming, I. A., Jonsson, B., Gross, M. R., and Lamberg, A. 1996. An experimental study of the reproductive behaviour and success of farmed and wild Atlantic salmon (Salmo salar). Journal of Applied Ecology, 33: 893 905. Fleming, I. A., Lamberg, A., and Jonsson, B. 1997. Effects of early experience on the reproductive performance of Atlantic salmon. Behavioral Ecology, (in press). Glebe, B. D., Appy, T. D., and Saunders, R. L. 1979. Variation in Atlantic salmon (Salmo salar) parr and adults reared in sea cages. Canadian Special Publication of Fisheries and Aquatic Sciences, 89: 24 29. Hansen, L. P., Døving, K. B., and Jonsson, B. 1987. Migration of farmed adult Atlantic salmon with and without olfactory sense, released on the Norwegian coast. Journal of Fish Biology, 30: 713 721. Hansen, L. P., Jacobsen, J. A., and Lund, R. A. 1993a. High numbers of farmed Atlantic salmon, Salmo salar L., observed in oceanic waters north of the Faroe Islands. Aquaculture and Fisheries Management, 24: 777 781. Hansen, L. P. and Jonsson, B. 1989. Salmon ranching experiments in the River Imsa: effects of timing of Atlantic salmon (Salmo salar) smolt migration on survival to adults. Aquaculture, 82: 367 373. Hansen, L. P. and Jonsson, B. 1991. The effect of timing of Atlantic salmon smolt and post-smolt release on the distribution of adult return. Aquaculture, 98: 61 71. Hansen, L. P. and Jonsson, B. 1994. Homing of Atlantic salmon: effects of juvenile learning on transplanted post-spawners. Animal Behaviour, 47: 220 222. Hansen, L. P., Jonsson, B., and Andersen, R. 1989. Salmon ranching experiments in the River Imsa: is homing dependent on sequential imprinting of the smolts? In Proceedings of the salmon migration and distribution symposium, June 23 25 1987. Second International Symposium, pp. 19 29. Ed. by E. Brannon and B. Jonsson. School of Fisheries, University of Washington, Seattle. Hansen, L. P., Jonsson, N., and Jonsson, B. 1993b. Oceanic migration in homing Atlantic salmon. Animal Behaviour, 45: 927 941. Hansen, L. P., Reddin, D. G., and Lund, R. A. 1997. The incidence of reared Atlantic salmon (Salmo salar L.) of fish farm origin at West Greenland. ICES Journal of Marine Science, 54: 152 155. Heggberget, T. G., Johnsen, B. O., Hindar, K., Jonsson, B., Hansen, L. P., Hvidsten, N. A., and Jensen, A. J. 1993a.

1038 B. Jonsson Interactions between wild and cultured Atlantic salmon: a review of the Norwegian experience. Fisheries Research, 18: 123 146. Heggberget, T. G., Økland, F. R., and Ugedal, O. 1993b. Distribution and migratory behaviour of adult wild and farmed Atlantic salmon (Salmo salar) during return migration. Aquaculture, 118: 73 83. Hindar, K. and Jonsson, B. 1995. Impacts of aquaculture and hatcheries on wild fish. In Protection of aquatic biodiversity. Proceedings of the World Fisheries Congress, Theme 3. pp. 70 87. Ed. by D. P. Philipp, J. M. Epifanio, J. E. Marsden, J. E. Claussen, and R. J. Wolotira, Jr. Oxford and IBH Publishing, New Delhi. Hindar, K., Ryman, N., and Utter, F. 1991. Genetic effects of cultured fish on natural fish populations. Canadian Journal of Fisheries and Aquatic Sciences, 48: 945 957. Holm, M. and Fernö, A. 1986. Aggression and growth of Atlantic salmon parr II. Different populations in pure and mixed groups. Fiskeridirektoratets Skrifter Serie Havundersøkelser, 18: 123 129. Huntingford, F. A., Metcalfe, N. B., Thorpe, J. E., Graham, W. D., and Adams, C. E. 1990. Social dominance and body size in Atlantic salmon parr, Salmo salar L. Journal of Fish Biology, 36: 877 881. Hutchings, J. A. 1991. Fitness consequences of variation in egg size and food abundance in brook trout, Salvelinus fontinalis. Evolution, 45: 1162 1168. Johnsson, J. I. and Abrahams, M. V. 1991. Interbreeding with domestic strain increases foraging under threat of predation in juvenile steelhead trout (Oncorhynchus mykiss). Canadian Journal of Fisheries and Aquatic Sciences, 48: 243 247. Johnsson, J. I., Petersson, E., Jönsson, E., Björnsson, B. T., and Järvi, T. 1996. Domestication and growth hormone alter antipredation behaviour and growth pattern in juvenile brown trout, Salmo trutta. Canadian Journal of Fisheries and Aquatic Sciences, 53: 1546 1554. Jones, J. W. and Ball, J. N. 1954. The spawning behaviour of brown trout and salmon. British Journal of Animal Behaviour, 2: 103 114. Jonsson, B. and Fleming, I. A. 1993. Enhancement of wild populations. In Human impact of self-recruiting populations, pp. 209 238. Ed. by G. Sundnes. Royal Norwegian Society of Sciences and Letters Foundation, Tapir Publishers, Trondheim. Jonsson, B., Jonsson, N., and Hansen, L. P. 1990a. Does juvenile experience affect migration and spawning of adult Atlantic salmon? Behavioral Ecology and Sociobiology, 26: 225 230. Jonsson, B., Jonsson, N., and Hansen, L. P. 1991a. Differences in life history and migratory behaviour between wild and hatchery reared Atlantic salmon in nature. Aquaculture, 98: 69 78. Jonsson, N., Jonsson, B., and Hansen, L. P., 1990b. Partial segregation in the timing of migration of Atlantic salmon of different ages. Animal Behaviour, 40: 313 321. Jonsson, N., Hansen, L. P., and Jonsson, B. 1991b. Variation in age, size and repeat spawning of adult Atlantic salmon in relation to river discharge. Journal of Animal Ecology, 60: 937 947. Jonsson, N., Jonsson, B., and Hansen, L. P. 1993. Migratory behaviour and growth of hatchery-reared post-smolt Atlantic salmon Salmo salar L. Journal of Fish Biology, 42: 435 443. Jonsson, N., Jonsson, B., and Hansen, L. P. 1994. Juvenile experience influences timing of adult river ascent in Atlantic salmon. Animal Behaviour, 48: 740 742. Jonsson, N., Jonsson, B., and Fleming, I. A. 1996. Does early growth rate cause a phenotypically plastic response in egg production of Atlantic salmon? Functional Ecology, 10: 89 96. Kazakov, R. V. 1981. The effect of the size of Atlantic salmon, Salmo salar L., eggs on embryos and alevins. Journal of Fish Biology, 19: 353 360. Leider, S. A., Hulett, P. L., Loch, J. J., and Chilcote, M. W. 1990. Electrophoretic comparison of the reproductive success of naturally spawning transplanted and wild steelhead trout through the returning adult stage. Aquaculture, 88: 239 252. Lund, R. A., Økland, F. R., and Hansen, L. P. 1991. Farmed Atlantic salmon in fisheries and rivers in Norway. Aquaculture, 98: 143 150. Lund, R. A., Østborg, G. M., and Hansen, L. P. 1996. Escapes of farmed salmon in marine homewater and in the riverine fisheries in the period 1989 1995. NINA Oppdragsmelding, 411: 1 16. (In Norwegian, English summary.) Lura, H., Barlaup, B. T., and Sægrov, H. 1993. Spawning behaviour of a farmed escaped female Atlantic salmon (Salmo salar). Journal of Fish Biology, 42: 311 313. Lura, H. and Økland, F. R. 1994. Content of synthetic astaxanthin in escaped farmed female Atlantic salmon, Salmo salar, in Norwegian rivers. Fisheries Management and Ecology, 1: 205 216. Lura, H. and Sægrov, H. 1991a. Documentation of successful spawning of escaped farmed Atlantic salmon, Salmo salar, in Norwegian rivers. Aquaculture, 98: 151 159. Lura, H. and Sægrov, H. 1991b. A method of separating offspring from farmed and wild Atlantic salmon (Salmo salar) based on different ratios of optical isomers of astaxanthin. Canadian Journal of Fisheries and Aquatic Sciences, 48: 143 150. Lura, H. and Sægrov, H. 1993. Timing of spawning in cultured and wild Atlantic salmon (Salmo salar) in the River Vosso, Norway. Ecology of Freshwater Fish, 2: 167 172. Metcalfe, N. B., Huntingford, F. A., Graham, W. D., and Thorpe, J. E. 1989. Early social status and the development of life-history strategies in Atlantic salmon. Proceedings of the Royal Society of London Series B, 236: 7 19. Metcalfe, N. B. and Thorpe, J. E. 1992. Early predictors of life-history events: the link between first feeding date, dominance and seaward migration in Atlantic salmon, Salmo salar L. Journal of Fish Biology, 41 (Suppl. B): 93 99. Mills, D. 1989. Ecology and management of Atlantic salmon. Chapman and Hall, London. 351 pp. Nickelson, T. E., Solazzi, M. F., and Johnson, S. L. 1986. Use of hatchery coho salmon (Oncorhynchus kisutch) presmolts to rebuild wild populations in Oregon coastal streams. Canadian Journal of Fisheries and Aquatic Sciences, 43: 2443 2449. Økland, F. R., Heggberget, T. G., and Jonsson, B. 1995. Migration behaviour of wild and farmed Atlantic salmon (Salmo salar) during spawning. Journal of Fish Biology, 46: 1 7. Quinn, T. P., Hendry, A. P., and Wetzel, L. A. 1995. The influence of life history trade-offs and the size of incubation gravels on egg size variation in sockeye salmon (Oncorhynchus nerka). Oikos, 74: 425 438. Reisenbichler, R. R. and McIntyre, J. D. 1977. Genetic differences in growth and survival of juvenile hatchery and wild steelhead trout, Salmo gairdneri. Journal of the Fisheries Research Board of Canada, 34: 123 128. Ruzzante, D. E. 1994. Domestication effects on aggressive and schooling behavior in fish. Aquaculture, 120: 1 24. Ruzzante, D. E. and Doyle, R. W. 1990. Behavioral and growth responses to the intensity of intraspecific social interaction among medaka Oryzias latipes (Temminck and Schlegal) (Pisces, Cyprinodontidae). Journal of Fish Biology, 37: 663 673.

Interactions between cultured and wild salmon 1039 Shustov, Yu. A., Schurov, I. L., and Smirnov, Yu. A. 1980. On the rates of adaption of hatchery young of the Atlantic salmon, Salmo salar L., to riverine conditions. Journal of Ichthyology, 20: 758 761. Skaala, Ø., Jørstad, K. E., and Borgstrøm, R. 1996. Genetic impact on two wild brown trout (Salmo trutta) populations after release of non-indigenous hatchery spawners. Canadian Journal of Fisheries and Aquatic Sciences, 53: 2027 2035. Swain, D. P. and Riddell, B. E. 1990. Genetic variation in agonistic behavior of juveniles between hatchery and wild stocks of coho salmon (Oncorhynchus kisutch) rearing in a lake and its tributary stream. Canadian Journal of Fisheries and Aquatic Sciences, 47: 566 573. Thorpe, J. E., Miles, M. S., and Keay, D. S. 1984. Developmental rate, fecundity and egg size in Atlantic salmon, Salmo salar L. Aquaculture, 43: 289 305. van den Berghe, E. P. and Gross, M. R. 1989. Natural selection resulting from female breeding competition in a Pacific salmon (coho: Oncorhynchus kisutch). Evolution, 43: 125 140. Webb, J. H., Hay, D. W., Cunningham, P. D., and Youngson, A. F. 1991. The spawning behaviour of escaped farmed and wild adult Atlantic salmon (Salmo salar L.) in a northern Scottish river. Aquaculture, 98: 97 110. Webb, J. H., McLaren, I. S., Donaghy, M. J., and Youngson, A. F. 1993a. Spawning of farmed Atlantic salmon, Salmo salar L., in the second year after their escape. Aquaculture and Fisheries Management, 24: 557 561. Webb, J. H., Youngson, A. F., Thompson, C. E., Hay, D. W., Donaghy, M. J., and McLaren, I. S. 1993b. Spawning of escaped farmed Atlantic salmon, Salmo salar L., in western and northern Scottish rivers: egg deposition by females. Aquaculture and Fisheries Management, 24: 663 670. Youngson, A. F., Martin, S. A. M., Jordan, W. C., and Verspoor, E. 1991. Genetic protein variation in Atlantic salmon in Scotland: comparison of wild and farmed fish. Aquaculture, 98: 231 242.