Changes in the density of hosts

A major determinant of the transmission rate of a parasite is the availability of susceptible hosts (Anderson and May, Reference Anderson and May1979; May and Anderson, Reference Anderson and May1979; Arneberg et al. Reference Arneberg, Skorping, Grenfell and Read1998). This has to exceed a minimum density – the critical threshold – to allow enough transmissions to maintain the parasite population. Human activities that alter the density of hosts can consequently have a major effect on host-parasite encounter rate and, hence, on parasite transmission (e.g. Farnsworth et al. Reference Farnsworth, Wolfe, Hobbs, Burnham, Williams, Theobald, Conner and Miller2005).

Human activities that drastically reduce the density of hosts are expected to reduce encounter rate and thereby parasite transmission rate and abundance (Lyles and Dobson, Reference Lyles and Dobson1993; Keeling and Grenfell, Reference Keeling and Grenfell1997; Arneberg et al. Reference Arneberg, Skorping, Grenfell and Read1998; Morand and Poulin, Reference Morand and Poulin1998; Dunn et al. Reference Dunn, Harris, Colwell, Koh and Sodhi2009). In support of this, Altizer et al. (Reference Altizer, Nunn and Lindenfors2007) found that primates threatened because of various anthropogenic disturbances carry fewer parasite species than sister taxa that are less threatened. Similarly, fish populations that are declining because of overfishing and other human-induced changes are less parasitized than stable populations (Amundsen and Kristoffersen, Reference Amundsen and Kristoffersen1990; Wood et al. Reference Wood, Lafferty and Micheli2010). For instance, the decline of lake trout populations Salvelinus namaycush in the Great Lakes of North America because of intensive fishing correlates with lower parasite prevalence. Inspection of museum specimens revealed that trout caught before 1925 were more often infected by the nematode Cystidicola stigmatura than trout caught after 1925 (Black, Reference Black1983, Reference Black1985). Amundsen and Kristoffersen (Reference Amundsen and Kristoffersen1990) showed experimentally that a reduction in the density of hosts can decrease parasite prevalence. When they reduced the density of whitefish Coregonus spp. in Lake Stuorajarvi in Norway, the prevalence of the cestode Diphyllobothrium ditremum decreased although the abundance of its intermediate and definitive hosts – copepods and birds – did not change. Further support for an influence of human-induced reductions in host density on parasitism is provided by a literature survey. This found infectious diseases to decrease over time in marine fishes that are declining, but to increase over time in marine taxa that are not declining (Ward and Lafferty, Reference Ward and Lafferty2004).

Correspondingly, human activities that increase host density are expected to raise encounter rate and, hence, parasite abundance, as a higher host density intensifies contacts among hosts. In support of this, the stocking of Lake Kilpisjärvi in Finland with whitefish increased parasite infections. However, a cessation of stocking did not reduce parasite abundance (Tolonen and Kjellman, Reference Tolonen and Kjellman2001). This suggests that other factors than density also influenced the abundance of parasites.

Human activities also can influence parasite transmission rate indirectly, by influencing the density of the predators that prey upon the hosts. For instance, the introduction of the predatory fish salmon Salmo salar into Lake Kvernavann in Norway has led to a dramatic decrease in the prevalence of the cestode Schistocephalus solidus in the population of threespine stickleback Gasterosteus aculeatus in the lake, as the salmon selectively preys upon parasitized stickleback (Jakobsen et al. Reference Jakobsen, Johnsen and Larsson1988).

Human-induced environmental changes that alter the distribution of hosts and, hence, local density, also alter parasite transmission rate. Habitat fragmentation, for instance, that forces individuals to aggregate in a few localities, can increase encounter rate and, thus, parasite transmission rate (Garnett and Holmes, Reference Garnett and Holmes1996; McCallum and Dobson, Reference McCallum and Dobson2002). An example is the recent growth of human populations in eastern Kenya, which has fragmented the habitat of the endemic monkey Tana River red colobus Procolobus rufomitratus (Wieczkowski and Mbora, Reference Wieczkowski and Mbora2000; Mbora and Meikle, Reference Mbora and Meikle2004a , Reference Mbora and Meikle b ). This has increased the density of monkeys in isolated patches, which in turn has made the species more vulnerable to parasites (Mbora and McPeek, Reference Mbora and McPeek2009).

Human-induced alterations in the density of a host also can influence encounter rate with the next host in the life-cycle of the parasite, as many parasites infect several host species sequentially. For instance, human-induced increases in the growth of algae in aquatic ecosystems have increased the density and size of planorbid snails, the primary intermediate hosts of the trematode Ribeiroira ondatrae, and allowed the snails to tolerate longer infections. This has nearly doubled the number of cercariae larvae that are released from the snails and exposed to the next hosts in their life-cycle, fishes and amphibians (Johnson et al. Reference Johnson, Chase, Dosch, Hartson, Gross, Larson, Sutherland and Carpenter2007). This has increased the prevalence of malformations in amphibians (Fig. 2, Johnson and Chase, Reference Johnson and Chase2004; Johnson et al. Reference Johnson, Chase, Dosch, Hartson, Gross, Larson, Sutherland and Carpenter2007).

Fig. 2.

An example of a complex life-cycle of a parasite. The life-cycle of the digenetic trematode Ribeiroia ondatrae involves both free-living stages, which are directly exposed to environmental perturbations, and parasitic stages, which rely on other organisms for development and reproduction and, hence, are indirectly influenced by effects on these organisms. The miracidia larvae infect aquatic snails, in which they develop to sporocysts. The sporocysts produce cercariae asexually, which are released into the water where they search for amphibians to infect. In the frogs, the cercariae develop into metacercariae, which cause malformations in amphibians. Predation by birds transmits the metacercariae to the definitive bird host, where the metacercariae mature and reproduce sexually. Anthropogenic disturbances that affect at least one of the different stages can compromise the completion of the whole life-cycle of the parasite.

Changes in the density of parasites

Human activities that alter the density of parasites can similarly alter the encounter rate between hosts and parasites and, thus influence parasite transmission rate. Reductions in parasite density are likely to reduce parasite transmission rate, while increases in parasite density can increase parasite transmission rate.

Human-induced environmental changes influence the density of parasites both directly – such as global warming that influences the growth rate of parasites – and indirectly by altering the density of hosts (discussed in the previous section), of competitors, or of predators that prey on parasites. An example of an indirect effect of alterations in predator density on parasite density is the removal by humans of the fishes that prey on the crab Mithrax nodosus. The crab is a predator on parasitic eulimid snails – Sabinella shaskyi and Pelseneeria spp. – that infect the sea urchin Eucidaris galapagensis. Thus, an increased density of the crab because of overfishing has reduced the density of the parasitic snails (Sonnenholzner et al. Reference Sonnenholzner, Lafferty and Ladah2011). Correspondingly, a recent meta-analysis (Zhao et al. Reference Zhao, Neher, Fu, Li and Wang2013) found human use of herbicides to benefit plant-parasitic nematodes by suppressing the density of the predatory nematodes that feed on these parasitic nematodes.

Global warming because of anthropogenic activities is expected to have positive effects on the abundance of many parasites, as temperature can accelerate their growth and maturation rates (reviewed in Chubb, Reference Chubb1977, Reference Chubb1979, Reference Chubb1980, Reference Chubb1982). Accordingly, elevated temperature has been found to accelerate the development rate of trematode eggs and allow trematodes to complete their life-cycles more rapidly. This has increased the production and transmission rate of trematodes (Poulin, Reference Poulin2006; Paull and Johnson, Reference Paull and Johnson2011). Elevated temperature was also found to accelerate the growth rate of the cestode S. solidus in its secondary intermediate host, the threespine stickleback (Macnab and Barber, Reference Macnab and Barber2012). As higher parasite weight increases the fecundity of the parasite in its definitive host, birds (Tierney & Crompton, 1992), elevated temperatures could increase egg production and thus parasite transmission rate.

Anthropogenic pollution is another factor with major effects on parasite abundance and thereby on the transmission of parasites to hosts. For instance, Kelly et al. (Reference Kelly, Poulin, Tompkins and Townsend2010) showed that the exposure of the snail Potamopyrgus antipodarum to moderate levels of the common herbicide glyphosate increases the production and release by the snail of the infective cercarial stages of the trematode Telogaster opistorchis. This raises in turn the density of parasites for the next host in the life-cycle of the trematode, juveniles of the freshwater fish Galaxias anomalus. Moreover, synergistic effects between the herbicide and the parasite amplify the negative impact of the parasite on infected fish, causing malformations and death of fish (Kelly et al. Reference Kelly, Poulin, Tompkins and Townsend2010).

Changes in the temporal distribution of hosts and parasites

The often ephemeral distribution of hosts forces parasites to adjust their temporal distribution to that of their hosts (Lewis et al. Reference Lewis, Campbell and Sukhdeo2002). For example, helminth parasites time the hatching and emergence of their larvae to abiotic factors associated with the presence of susceptible hosts, such as temperature, light and humidity (Pietrock and Marcogliese, Reference Pietrock and Marcogliese2003). This tracking of the presence of hosts can cause highly seasonal parasite epidemics (Altizer et al. Reference Altizer, Dobson, Hosseini, Hudson, Pascual and Rohani2006). Human-induced changes in abiotic factors that alter the temporal distribution of hosts will consequently also alter that of parasites. This can in turn induce evolutionary changes in the phenology of both hosts and parasites (Lafferty, Reference Lafferty2009).

A growing human-induced environmental problem that is affecting the phenology of both hosts and parasites is global warming (Rogers and Randolph, Reference Rogers and Randolph2006). For instance, higher temperature has been found to prolong the time window during which mosquitoes can breed and extend the period when mosquito-borne parasites are transmitted to humans (Reiter, Reference Reiter2001). This prolonged transmission period has been suggested to have contributed to the spread of the bluetongue virus in Europe during recent years (reviewed in Purse et al. Reference Purse, Mellor, Rogers, Samuel, Mertens and Baylis2005).

Global warming can also reduce the reliability of environmental cues as indicators of the ideal timing of life-history events (Schweiger et al. Reference Schweiger, Settele, Kudrna, Klotz and Kühn2008). This mis-timing can have particularly deleterious consequences for parasites that spend part of their life-cycle outside the host, where the host's internal temperature does not buffer them against changing environmental conditions. Little evidence exists, however, of mis-timed phenologies between hosts and parasites, probably because of the scarcity of long-term data series for the prevalence and abundance of parasites (Møller, Reference Møller2010).

Another anthropogenic disturbance that can have profound effects on the temporal distribution of both hosts and parasites is thermal pollution of aquatic ecosystems. For instance, a long-term study by Aho et al. (1982) found thermal discharges in cooling reservoirs to extend the transmission period of the metacercariae of the eyefluke Tylodelphys scheuringi. This increased infections in populations of mosquitofish Gambusia affinis.

Changes in the diversity of hosts and parasites

The encounter rate between a parasite and a host depends not only on the density of the target host, but also on that of alternative hosts and non-hosts, i.e. on the composition of the community (Thieltges et al. Reference Thieltges, Jensen and Poulin2008a ; Johnson and Thieltges, Reference Johnson and Thieltges2010). Human-induced introductions or removals of species can consequently have positive or negative effects on native host-parasite interactions, depending on how the change influences encounter rate with compatible hosts.

Human-induced introduction of novel host species promotes parasite transmission if a new compatible host-parasite association is formed from which parasites can be ‘spilled back’ to native hosts (reviewed in Kelly et al. Reference Kelly, Paterson, Townsend, Poulin and Tompkins2009b ). This can occur when the introduced and the native hosts are ecologically similar, i.e. the encounter filter is open, and when the two hosts are closely related so that the probability of compatibility with local parasites is high, i.e. the compatibility filter is open. Both the native and the introduced host will then experience higher parasitic pressure. The host that suffers the greatest negative impact might eventually go extinct (Anderson and May, Reference Anderson and May1979), or diverge ecologically and evolve lower overlap.

When humans introduce species that are not compatible with native parasites, the introduced species will experience a low risk of infection, in lines with the enemy release hypothesis (Torchin et al. Reference Torchin, Lafferty and Kuris2002, Reference Torchin, Lafferty, Dobson, McKenzie and Kuris2003; Torchin and Lafferty, Reference Torchin, Lafferty, Rilov and Crooks2008). This can dilute infection risks also for native hosts by increasing the parasite's encounters with and infection attempts on incompatible hosts and, thus, reduce the abundance of the parasite (Fig. 3, Keesing et al. Reference Keesing, Holt and Ostfeld2006; Johnson and Thieltges, Reference Johnson and Thieltges2010). An example of this is the introduction by humans of the European freshwater snail Lymnaea stagnalis to New Zealand at the end of the 19th century (Hutton, Reference Hutton1881). The introduced snail is less compatible than the native snail P. antipodarum with the native trematode Microphallus sp., i.e. the encounter filter is open but the compatibility filter is closed (Kopp and Jokela, Reference Kopp and Jokela2007). Thus, increased infection attempts on the introduced, incompatible snail has reduced the prevalence of the parasite, which in turn has reduced encounters with the compatible, native snail and released it from infections (Kopp and Jokela, Reference Kopp and Jokela2007). Similar dilution effects because of human-introduced species have been reported for various parasites of molluscs, fishes and mammals (e.g. Telfer et al. Reference Telfer, Bown, Sekules, Begon, Hayden and Birtles2005; Thieltges et al. Reference Thieltges, Reise, Prinz and Jensen2008b ; Kelly et al. Reference Kelly, Paterson, Townsend, Poulin and Tompkins2009a ; Paterson et al. Reference Paterson, Townsend, Poulin and Tompkins2011).

Fig. 3.

The dilution of infection risks through increases in the density of incompatible hosts. Individuals of the white host population carry one parasite species able to differentially infect other host populations in the community. Black hosts are compatible hosts for the parasite (i.e. sources for the parasite population), while grey hosts are incompatible hosts (i.e. sinks for the parasite population). Although the population of white hosts remains constant under the three community structures, alterations in the composition of the host assemblage affect infection risk for the white hosts.

When human-introduced species bring with them non-indigenous parasites, these can be ‘spilled over’ to native hosts and affect them negatively. The topic has received considerable attention in the literature (Prenter et al. Reference Prenter, MacNeil, Dick and Dunn2004). A classic example is the massive mortality of amphibians worldwide because of global trade with African-clawed frogs Xenopus sp. The frogs, which are used in research, as pets, and to conduct pregnancy tests, carry the chytrid fungus Batrachochytrium dendrobatidis on their skin, which is spread to native species when the frog is introduced into novel environments (Weldon et al. Reference Weldon, Du Preez, Hyatt, Muller and Speare2004; Ouellet et al. Reference Ouellet, Mikaelian, Pauli, Rodrigue and Green2005; Rachowicz et al. Reference Rachowicz, Hero, Alford, Taylor, Morgan, Vredenburg, Collins and Briggs2005). Another classic example is the worldwide spread of the swim-bladder nematode Anguillicola crassus with the commercial export of Japanese eel Anguilla japonica (Køie, Reference Køie1991). The low specificity of the nematode for intermediate hosts has allowed it to rapidly spread to other eel species, where it causes pathological damage to the swimbladder (Moravec, Reference Moravec1996; Moravec and Škoriková, Reference Moravec and Škoriková1998). This could have contributed to the dramatic decline of European eel Anguilla anguilla populations in recent years (Køie, Reference Køie1991).

Changes in search and avoidance strategies

The search strategy of the parasite and the avoidance strategy of the host depend crucially on environmental conditions. Human disturbances that influence these strategies can consequently alter host encounter rate. An insidious human-induced problem that is currently altering the behaviour of both parasites and hosts is chemical pollution. Chemical cues are often used by free-living stages of parasites to locate their hosts (Lewis et al. Reference Lewis, Campbell and Sukhdeo2002). This means that human activities that influence the chemical composition of the environment also alter host-parasite interactions. For instance, the increased use of pesticides has impaired the search behaviour of parasitic wasps by affecting their longevity, activity and orientation (Desneux et al. Reference Desneux, Decourtye and Delpuech2007).

Human-induced increases in the concentration of heavy metals pose a serious threat to host-parasite interactions, as the metals often influence the behaviour of both parasites and hosts. For example, Morley et al. (Reference Morley, Crane and Lewis2005) showed that a mixture of zinc and cadmium influences the energy expenditure of the cercarial larvae of the trematode Diplostomum spathaceum. The larvae normally shed their tails when their energy store is almost depleted in order to reduce energy use. When the larvae were exposed to a single metal, the frequency of tail losses increased with age. However, when the larvae were exposed to a mixture of metals, tail losses were delayed (Morley et al. Reference Morley, Crane and Lewis2002, Reference Morley, Crane and Lewis2005). This is expected to increase energy use and reduce energy store, and, hence, impair the activity and transmission rate of the parasite. Together with a reduction in the release of larvae from the intermediate snail host population, because of negative effects of heavy metals on snails (Coeurdassier et al. Reference Coeurdassier, De Vaufleury, Scheifler, Morhain and Badot2004), this could drastically reduce the prevalence of the parasite.

Changes in parasite avoidance behaviour of hosts because of human-induced environmental changes can be categorized into two main groups: (1) changes in pre-contact avoidance behaviour, such as habitat choice and selective foraging, and (2) changes in post-contact avoidance behaviour, such as grooming (Thieltges and Poulin, Reference Thieltges and Poulin2008). These changes are expected to influence host-parasite interactions. However, the impact that human-induced environmental changes have on avoidance behaviour has received surprisingly little attention. One exception is the study by Rohr et al. (Reference Rohr, Swan, Raffel and Hudson2009) on the influence of the herbicide atrazine on parasite avoidance behaviour of tadpoles of the American toad Bufo americanus.

Plastic changes in virulence and resistance

The degree to which human-induced environmental changes cause plastic alterations in parasite virulence and host resistance depends on the genetic constitution of the individuals, as genes determine reaction norms and, thus, how traits vary across environments. Common causes of plastic adjustments of parasite virulence are alterations in host condition or in the intensity of competition for hosts (e.g. Tseng, Reference Tseng2006; Tschirren et al. Reference Tschirren, Bischoff, Saladin and Richner2007; Choisy and de Roode, Reference Choisy and de Roode2010). Plastic changes in parasite virulence can in turn cause plastic changes in host resistance.

Human-induced global warming is increasingly recognized as a potential cause of increased virulence of parasites, as temperature often acts as an ‘on-off’ switch for the expression of virulence genes (Konkel and Tilly, Reference Konkel and Tilly2000; see Marcogliese, Reference Marcogliese2001; Mitchell et al. Reference Mitchell, Rogers, Little and Read2005). This is currently contributing to the worldwide bleaching of corals – the loss of endosymbiotic zooxanthellae – through effects of temperature on the expression of virulence factors in bacteria that inhibits photosynthesis of the zooxanthellae (reviewed in Rosenberg and Ben-Haim, Reference Rosenberg and Ben-Haim2002). For instance, the current decline of the coral Oculina patagonica is caused by global warming activating the expression of several virulence factors in the coral-bleaching bacterium Vibrio shiloi (reviewed in Rosenberg and Falkovitz, Reference Rosenberg and Falkovitz2004).

Similarly, various forms of human-induced pollution alter virulence and resistance by causing physiological changes in parasites and hosts. Chemical pollutants, in particular, can impair immune responses of hosts by altering the production, longevity, structure and function of the cells and proteins of the immune system (Schuurman et al. Reference Schuurman, Frieke Kuper and Vos1994; Banerjee, Reference Banerjee1999; Blakley et al. Reference Blakley, Brousseau, Fournier and Voccia1999; Jorgensen, Reference Jorgensen2010). For example, chemical compounds that impair the production of the mucus that protects the gills of fishes can hamper resistance, as the mucus contains compounds that trap and inhibit the growth of microorganisms (see Shephard, Reference Shephard1994; Bols et al. Reference Bols, Brubacher, Ganassin and Lee2001; Alvarez-Pellitero, Reference Alvarez-Pellitero2008 and references therein). Petroleum hydrocarbons, for instance, impair mucus production in cod, which increase the prevalence of gill parasites, such as ciliates and monogeneans (Khan and Kiceniuk, Reference Khan and Kiceniuk1988; Khan, Reference Khan1990). Another chemical pollutant that hampers the functioning of the immune system in a range of fishes and amphibians is the herbicide atrazine (Rohr and McCoy, Reference Rohr and McCoy2010). It triggers the degeneration of macrophages in mullets (Biagianti-Risbourg, Reference Biagianti-Risbourg1990), lowers the respiratory burst activity of circulating phagocytes in rainbow trout (Rymuszka et al. Reference Rymuszka, Siwicki and Sieroslawska2007), and reduces the abundance of T cells in a number of species (e.g. Christin et al. Reference Christin, Gendron, Brousseau, Ménard, Marcogliese, Cyr, Ruby and Fournier2003; Christin et al. Reference Christin, Menard, Gendron, Ruby, Cyr, Marcogliese, Rollins-Smith and Fournier2004; Rymuszka et al. Reference Rymuszka, Siwicki and Sieroslawska2007). A meta-analysis found the herbicide to reduce 77% of the immune responses that were included in the analysis (Rohr and McCoy, Reference Rohr and McCoy2010). On the other hand, the herbicide has positive effects on the immune system of some species, through complex interactions between the herbicide and physiological functions (e.g. Chakrabarty and Banerjee, Reference Chakrabarty and Banerjee1988; Sures, Reference Sures2006; Sures, Reference Sures2008).

Anthropogenic disturbances can influence host resistance also indirectly, through trade-offs between resistance and other fitness functions (Martin et al. Reference Martin, Hopkins, Mydlarz and Rohr2010). For instance, Adamo and Lovett (Reference Adamo and Lovett2011) showed that increasing temperatures because of global warming increases reproductive output, mass gain and the activity of the immune system in the cricket Gryllus texensis. Because of physiological trade-offs between reproduction and disease resistance, the increased effort reduces resistance to some parasites. This suggests that global warming could cause crickets to become more susceptible to some parasites.

Evolutionary changes in virulence and resistance

Human activities that cause longer-term changes in the environment can induce evolutionary changes in virulence and resistance if genetic variation in the direction of selection exists, or is introduced through gene flow or mutations. The influence of human activities on the evolution of parasite virulence is especially pronounced at farms where the density of hosts is high, as high host density favours higher parasite virulence (Murray and Peeler, Reference Murray and Peeler2005; Mennerat et al. Reference Mennerat, Nilsen, Ebert and Skorping2010). In aquacultures, the density of one species can even be a thousand times higher than in the wild (Pulkkinen et al. Reference Pulkkinen, Suomalainen, Read, Ebert, Rintamaki and Valtonen2010). An example of a parasite that has evolved more virulent strains is the salmon louse Lepeophtheirus salmonis at salmon farms (Mennerat et al. 2012). After the rapid growth of aquaculture, the more virulent strains have amplified and spread to wild salmonid populations, where they have had adverse effects on the natural stocks (Krkošek et al. Reference Krkošek, Ford, Morton, Lele, Myers and Lewis2007; Costello, Reference Costello2009). Another example of the evolution of more virulent strains is the bacterial fish parasite Flavobacterium columnare, which frequently causes severe parasite outbreaks at fish farms (Kunttu et al. Reference Kunttu, Valtonen, Jokinen and Suomalainen2009; Pulkkinen et al. Reference Pulkkinen, Suomalainen, Read, Ebert, Rintamaki and Valtonen2010).

A classic example of human-induced coevolution between host resistance and parasite virulence is the release of the myxomatosis poxvirus in Australia to control the population growth of the introduced European rabbit Oryctolagus cuniculus. The most lethal strain of the virus, originating from America, was released during the 1950s. This caused dramatic mortality among the rabbits, with over 99% of the infected rabbits dying. Within 10 years, a less virulent strain of the virus had established itself, which killed only 50% of infected rabbits (Fenner and Marshall, Reference Fenner and Marshall1957; Marshall and Fenner, Reference Marshall and Fenner1960). The higher survival of the infected rabbits enhanced the transmission of the less virulent strain, which in turn selected for enhanced host defence (Massad, Reference Massad1987). This led to the evolution of resistant rabbits, which in turn selected for higher virulence (Best and Kerr, Reference Best and Kerr2000).

For parasite virulence and host defences to evolve, they need to be heritable and express genetic variation in the direction of selection. Declining populations often suffer from the loss of genetic variation. This can hamper the evolution of host defences against parasites and further worsen the probability of survival of the populations. For instance, several threatened vertebrate taxa show less variation in the genes of the major histocompatibility complex, MHC, which is involved in parasite recognition (e.g. Hedrick et al. Reference Hedrick, Parker, Miller and Miller1999, Reference Hedrick, Parker, Gutiérrez-Espeleta, Rattink and Lievers2000; Giese and Hedrick, Reference Giese and Hedrick2003). An example of this is captivity inbred winter-run chinook salmon Oncorhynchus tshawytscha, which are less resistant to three species of parasites than outbred populations (Arkush et al. Reference Arkush, Giese, Mendonca, McBride, Marty and Hedrick2002). This is particularly evident for salmon that are homozygous for the MHC genes. Surprisingly, wild Chinook populations show high allelic diversity for the MHC genes compared with other genes, despite also being inbred (Garrigan and Hedrick, Reference Garrigan and Hedrick2001).