S36.1: The avian immune system: A jack-of-all-trades

1Laboratoire d'Ecologie, CNRS URA 258, Université Pierre et Marie Curie, Bât. A, 7ème étage, 7 quai St. Bernard, Case 237, F-75252 Paris Cedex 05, France, fax 33 1 44 27 35 16, e mail amoller@snv.jussieu.fr; ²Dipartimento di Biologia, Sezione Zoologia Scienze Naturali, Università di Milano, Via Celoria 26, I-20133 Milano, Italy

The role of biotic and abiotic factors in determining reproductive success and ultimately population levels and regulation has a venerable history (Andrewartha & Birch 1954, 1984; Lack 1954). As concerns the biotic factors it is perhaps not surprising that intraspecific and interspecific competition and predation have played a prominent role in generation of ideas, since competition for limiting resources has been a cornerstone in evolutionary biology ever since Darwin. Furthermore, while the interactions between conspecifics or heterospecifics or between a predator and its prey can be readily seen and studied by biologists, this is not similarly the case for parasites and pathogens. In actual fact, it may be the case that an individual wins over another because the second is seriously infected with parasites. Similarly, the reason why a particular individual is killed and eaten by a predator may arise due to parasite infection (Temple 1987). Parasites have been on the rise as the main biological player in shaping the evolution of phenomena such as sex, anisogamy, sexual selection, life-history, dispersal and migration (e. g. Price 1980). Population ecology has similarly changed emphasis from competition and predation towards parasitism (Begon et al. 1986). The main reason for this change of emphasis is the realisation that parasites pose serious threats to their hosts. This is the case because small parasites have short generations and multiply much more rapidly than their hosts, and coevolutionary interactions between hosts and parasites continuously result in novel ways of parasite exploitation and host defence. Given the ubiquity of parasitism, the diversity of anti-parasite defences by hosts is not surprising. Such defences can be either behavioural, physiological or immunological. The first level of defence is avoidance of infection, while the second level comprises identification and destruction of parasites, and the third level control of a chronic infection (Hochberg 1997). Behavioural defences include avoidance of sites harbouring parasites, such as the avoidance of parasitised nests by Cliff Swallows Hirundo pyrrhonota (Emlen 1986), or avoidance of foraging sites that are contaminated by parasites (Hart 1990). Physiological responses include defences like elevated body temperatures that are detrimental to many parasites (Banet 1986), and compensation for parasite removal of host resources by increased parental effort in the case of nestlings (Johnson & Albrecht 1993) or increased ingestion in the case of adults. Finally, but most importantly, hosts may use immune responses as a direct weapon designed to deter and eliminate parasite attacks (Klein 1990; Roitt et al. 1996; Wakelin 1996). The immune system has evolved as a means of distinguishing self from non-self, and thus preventing parasites from having free access to limiting host resources.

The immune system comprises the prime way in which hosts defend themselves against parasites. General introductions to immunology are provided by Klein (1990), Roitt et al. (1996) and Wakelin (1996), while the avian immune system is described in detail by Rose et al. (1981), Toivanen & Toivanen (1987) and Davison et al. (1996). The following brief description is based on these sources. The avian immune system is comprised of two major arms of defence: the humoral and the cell mediated defence. Cells involved in the immune defence are derived from the thymus or the bursa of Fabricius. Thymic stem cells enter the thymus during early embryonic development and they subsequently differentiate into three populations of lymphocytes: T-helper, T-suppressor and T-cytotoxic cells. The function of these populations of cells in humoral and cell-mediated responses will be briefly described below. The humoral response includes the release of immunoglobulin (of which there are five different kinds) and is dependent on T-helper cells for initiation and T-suppressor cells for modulation of the response. B-cells differentiated in the bursa of Fabricius are responsible for the synthesis of immunoglobulins and antibody. Accessory cells are phagocytic and adherent cells are necessary for an optimal immune response. The humoral immunity consists of a response to a foreign substance or an antigen that directly activates a B-cell or binds to an accessory cell that activates T- and then B-cells. This is the primary immune response during which a latent phase is followed by an exponential phase characterised by a rapidly increasing level of circulating antibody, followed by a peak phase and a decline phase of antibody concentration. Each lymphocyte is capable of recognising only a single antigen. The enormous diversity of lymphocytes allows rapid reaction to a novel challenge, although many kinds of lymphocytes will never become active and proliferate because they never bind to an antigen. When an antigen binds to a few cells that can recognise it, this will result in rapid proliferation that gives rise to a sufficient number to raise an immune response in a few days. A secondary response will arise from subsequent exposure to the antigen, and this response is characterised by a short latent phase and a more rapid increase and a larger peak in antibody concentration with a subdued regression. The capacity to mount a secondary response is based on immunological memory; a process that is exploited by vaccination. The cellular basis of memory lies in the expansion of antigen-specific lymphocyte populations during the primary response, and the increased frequency of resting B- and T-cells capable of responding to that antigen in the future ensures a rapid response. Furthermore, memory B-cells make immunoglobulin earlier and have higher affinity antigen receptors than unprimed B-cells, due to clonal selection during the primary response. The bursa of Fabricius is a small dorsal diverticulum of the proctadealregion of the cloaca involved in differentiation and production of B-cells and in immunoglobulin synthesis. The bursa disappears before sexual maturity, at which age the function of the bursa is stored in circulating B-cells or B-cells stored in other immune defence organs. The cell-mediated immunity includes immunity that is independent of synthesis of antibody which are dependent on the normal development of the thymus; an immune defence organ that develops in the pharyngeal pouches but disappears before sexual maturity. Cells produced by the thymus are the three types of T-cells, but also immune responses arising in response to grafting and the delayed-type hyper-sensitivity reaction. The allograft and graft-versus-host responses, which are immune reactions to a skin transplant are derived from T-cells. The delayed-type hyper-sensitivity reaction occurs in response to an injection with an antigen and subsequent re-exposure to the antigen several weeks later. The peripheral lymphoid tissue consists of glands in which cells of thymic and bursal origin are stored. The Harderian gland is a small gland located ventral and posteromedial to the eyeball, and it contains B- and T-cells that produce antibodies. The cecal tonsil is an enlarged area of the proximal region of the caecum, which contains B- and T-cells and produces antibodies to soluble antigens. Peyer's patches appear along the intestine cranial to the ileocecal junction in young birds and contain lymphocytes beneath the epithelium. The spleen is an important immune defence organ located adjacent to the dorsal surface of the right lobe of the liver and dorsal to the proventriculus and contains vast numbers of lymphocytes. Lymph nodes have been located along the posterior tibio-popliteal and femoral veins and possess B- and T-cells. The pineal gland which is located between the cerebral hemispheres contains T- and B-cells.

The immune system is a multi-function device that is able to cope with invasion by an immense diversity of pathogens and parasites because the different cell populations are able to raise fast and efficient defences to virtually any novel challenge. The great level of sophistication of the immune system is not surprising given the multitude of ways in which parasites are able to exploit their hosts. The immune system by its mere existence is costly to produce and maintain. How large should this investment be? In other words, how much should be invested in defence to maximise the benefits and minimise the costs? This question is not readily answered because it will depend on the coevolution with parasites exploiting a particular host. Merino & Møller have presented evidence based on models of the coevolution of parasite virulence and host defence in a subsequent paper. Continuously running the immune system at a high level would entail costs additional to those of the mere activity. First, parasites would be able to adapt more quickly to the defences of a host simply by adjusting to the profile of the defence. This argument is similar to arguments concerning the evolution of pesticide resistance. Second, parasites might readily exploit the defence system of the host by hiding within immune defence organs or their products because these are protected from close scrutiny and attack by the immune system. Several severe parasites such as HIV virus and infectious bursal disease virus hide within parts of the immune system (Sharma & Rosenberger 1987). What is the evidence for costs of immune defence? The answer comes in three parts; (1) direct estimates of costs; (2) effects of activity on strength of immune responses; and (3) the immune system and auto-immune diseases. First, direct estimates of the cost of running the immune system are absent from the literature. A recent estimate for mice based on the size of immune defence organs, the production of proteins and lymphocytes, and the turnover of these components suggest that the total cost may be in the order of 20-25% of basal metabolic rate (V. Apanius pers. comm.). If this cost is realistic, it would imply that the cost of running the immune system ranks higher than the cost of brain function in mammals. At a different level, the costs of immune function are also revealed by negative genetic correlations between immune responsiveness and important fitness components such as growth in Drosophila melanogaster (Kraijeveld & Godfray 1997). Hence, a response to selection for increased immune response will result in a concomitant reduction in fitness due to reduced growth performance. Second, an emerging literature suggests that investment in other activities such as mating effort or parental effort gives rise to a reduction in immune function. Such relationships between activity and immune function may arise due to an energetic trade-off or due to one or more mediating mechanisms modulating the trade-off. Trade-offs between different activities and immune function will be discussed in detail below. Third, auto-immune diseases arise from components of the immune system attacking self rather than invading parasites (Berczi & Kovacs 1987; Golub & Green 1991). Typical auto-immune diseases in humans include multiple sclerosis, juvenile diabetes and rheumatoid arthritis. Similar diseases occur in other organisms. The cause of auto-immune diseases remains an area of great current interest. These diseases are much more common in women than in men, which is what one should expect if it is the strength of the immune system that is important, given the higher level of immune defence in women as compared to men (Berczi & Kovacs 1987). It is possible that a more diversified immune system with multiple clones of lymphocytes of similar abundance will provide a smaller threat to self than a less diversified immune system (Freitas & Rocha 1997). Auto-immune diseases are much more common in industrialised societies where the immune system rarely is challenged by a large number of parasites and pathogens that are characteristic of developing countries. Given the level of sophistication of the immune system, how can we ever measure the efficiency of the immune system of an individual or a population of individuals? Circulating levels of antibodies, concentration of lymphocytes and immunoglobulins or the size of immune defence organs such as the bursa of Fabricius, thymus or spleen will reflect the current status of infection of an individual and hence its health status. Thus measures of products of the immune system belong to two different categories, immune defence which reflects adaptations to an evolutionary history of parasite exposure, and immune response which reflects a physiological response to recent or current parasite infections. Measures of immune response reflect the physiological responses and thus the current health status of individuals. These measures do not necessarily reflect the maximum possible response for an individual because they depend on the particular parasite and the intensity of the parasite infection. Measures of immune defence thus reflect the ability of an individual to raise an immune reaction to a standardised immune defence challenge, such as that to injections with sheep red blood cells or a similar antigen. These measures are likely to reflect the maximum reaction possible for a given individual under given circumstances. Responses of birds to immune challenge tests will provide information on the ability to raise an immune response once infected. Immunocompetence is a composite measure of the ability of an individual to react to an infection. Since the immune system is so complex and multi-facetted, no single measure will suffice. A good measure of the ability of individuals to cope with parasite challenge consists of estimates of (1) the relative mass of lymphoid organs (spleen and thymus), (2) the ratio between heterophils and lymphocytes which is a general stress response measure which estimates immune changes caused by a variety of stressors, (3) response to immunisation with sheep red blood cells to obtain an in vivo measure of B- and T-cell activity, (4) response to immunisation with Brucella abortus to obtain in vivo response of B-cells independent of T-cells, (5) response to phytohaemagglutinin to obtain in vivo response of T-cells, and (6) in vitro assay of nitrous oxide production by macrophages from spleen cells (e. g. National Research Council 1992). Obviously, field ornithologists will usually only be able to obtain a few of these measures in any particular study. Interested readers should consult (Davison et al. 1996) for details.

Since different components of the immune system are involved in different tasks, we could imagine that heavy investment in one branch of the immune system would readily result in depletion of other branches. What is the evidence for trade-offs among different components of immune defence? Circulating lymphocytes are composed of clones of T-and B-cells, although the factors determining the composition of these clones remain unknown. Recent irradiation experiments in mice resulted in complete removal of all lymphocytes. Subsequent injection with different proportions of lymphocyte clones demonstrated that the proportions of the different clones did not remain constant, but shifted in ways consistent with competition among clones for a limiting resource (Freitas & Rocha 1997). The final population of lymphocytes of ca. 50 millions was reached independent of whether a single or multiple clones were used (Freitas & Rocha 1997). These findings suggest that the different clones of lymphocytes, which have different functions of recognising particular antigens, change in abundance, and that one can only dominate at the cost of another. Recently, two experiments on wild birds have revealed evidence of trade-offs among different components of immune function. Barn Swallow Hirundo rustica nestlings injected with sheep red blood cells (SRBC) to elicit a humoral immune response one week later demonstrated weaker cell mediated immune responses to a challenge with a lectin (Saino, Ninni & Møller unpublished results). This result makes sense since the SRBC challenge test elicits a systemic immune response, while the lectin antigen only consists of a local skin response. A second test using House Sparrows Passer domesticus as study organisms has attempted to identify which component of the immune system may be favoured by the hosts (Gonzalez et al. 1999). The aviary study provided evidence of a trade-off between humoral immunity as estimated by sheep red blood cell injection and cellular immunity as estimated after injection with phytohaemagglutinin, particularly under a low protein food regime. Furthermore, since the House Sparrows were infected with haemosporidian blood parasites of the genus Haemoproteus, that are challenged by the humoral component of the immune system, it was the cellular component of the immune system that was reduced under resource limitation. Ecologists have started to investigate how the efficiency of the immune system trade against investment in other activities. A central tenet in evolutionary biology is that effort spent on one activity is permanently gone and cannot be spent on other activities if energy or time is limiting (Maynard Smith 1977; Low 1978). Hence, reproductive effort is hypothesised to be traded against growth and/or maintenance, and mating effort is traded against parental effort. This simple line of argument can be taken to imply that measures of immune function should be inversely related to investment in foraging, mating or reproduction, if resources for one is traded against resources spent on another component. In accordance with this scenario, experimental manipulation of tail length in male Barn Swallows (a secondary sexual character) revealed a lower immune response to injections with sheep red blood cells (SRBC) in males with elongated tail feathers than males in two control groups and males with shortened tail feathers (Saino & Møller 1996). Furthermore, among males with elongated tail feathers there was a positive relationship between immune response and original tail length, implying that individuals in originally prime condition, and hence with large secondary sexual characters, had more resources available for raising an immune response. A second study of the Zebra Finch Taeniopygia guttata by Deerenberg et al. (1997) revealed that responsiveness to injections with SRBC was inversely related to manipulated brood size. Activity recordings of the birds showed that the reduction in immune responsiveness to the standard challenge test was proportional to the activity level of the birds. A second experiment demonstrated that training the birds simply to fly more resulted in a decrease in immune response compared to birds in a control group (Deerenberg et al. 1997). A study of foraging activity and immune responsiveness in Bumble Bees Bombus terretris also revealed a negative relationship between immune response and foraging activity (König & Schmid-Hempel 1995). The latter findings are consistent with a literature on humans and other organisms showing reduced immunocompetence under excessive activity levels (Fitzgerald 1988; Hoffman-Goetz & Pedersen 1994). In conclusion, the few studies available suggest that immune responsiveness is traded against investment in other costly activities.

The level of immune defence is not entirely a consequence of trade-offs between different activities of individuals because defence may be modulated by a number of biochemicals. The former includes general body condition and biochemicals such as carotenoids, while the latter includes various hormones such as estrogens and androgens, but also naturally occurring toxins acquired through the food. Body condition is an important determinant of immune function as demonstrated by numerous studies of humans and domesticated animals (Chandra & Newberne 1977; Gershwin et al. 1985) including poultry (Glick et al. 1981, 1983; Willis & Baker 1981) and wild birds (Lochmiller et al. 1993; Saino et al. 1997a; Møller et al. 1998c; Sorci et al. 1998). These effects of condition on immune function may arise as a consequence of the high costs of running an efficient immune system, or as a consequence of acquisition of specific nutrients that enhance immune function. Carotenoids comprise a large family of more than 600 kinds produced by plants and algae that can only be acquired through ingestion (e.g. Goodwin 1984). Many birds have yellow or red feathers or naked skin that are coloured by carotenoids, and the intensity of the coloration is presumed to directly reflect the concentration of carotenoids in the plasma during development (Fox 1976, 1979; Goodwin 1984; Gray 1996). Given that the intensity of coloration reflects the presence of rare biochemicals, signals based on carotenoids are likely to provide reliable information about the ability to acquire limiting resources. However, carotenoids may be important because they are directly involved in immune function and may thus provide information about the health status of individuals (Saino et al. unpublished manuscript). Carotenoids are associated with many functions of immunity. Beta-carotene and other carotenoids enhance T-and B-lymphocyte proliferate responses, stimulate effector T-cell function, enhance macrophage and cytotoxic T-cell capacities, increase the population of specific lymphocyte subsets, and stimulate the production of various cytokines and interleukins in humans as well as in other animals (Bendich 1989; Chew 1993). Recently, carotenoids have also been implicated in protection against neoplasia and atherosclerosis (Canfield et al. 1992). Moreover carotenoids protect animals from several kinds of experimentally induced tumours (Chew 1993). Hence, individuals that are able to acquire or metabolise carotenoids may be more healthy than others and thus enjoy enhanced mating and reproductive success. A number of biochemicals are known or supposed to reduce immune responsiveness. First, peroxides that arise from ingestion of toxins may severely suppress immune responsiveness, although this suppression can be reverted by the activity of carotenoids as oxygen radical scavengers, and protectors of lipids from peroxidation, thus decreasing immunosuppressive peroxides (Bendich 1989). A second group of immuno-suppressors are various hormones related to reproduction such as estrogens and androgens, but also corticosteroids (reviews in Grossman 1985; Folstad & Karter 1992; Hillgarth & Wingfield 1997). For example, a number of studies has suggested that testosterone has severe negative effects on immune responsiveness in a wide variety of vertebrates, and reproducing individuals thus have to trade the beneficial effects of this androgen against the costs of immune suppression. In accordance with this scenario, male birds generally have smaller immune defence organs than females, but only during the breeding season, and the degree of suppression of immune function is directly related to the intensity of sexual competition measured as the frequency of extra-pair paternity (Møller et al. 1998b). Hence, when males compete intensely among each other for access to females, males have more suppressed immune function than when there is little or no competition among males for access to copulations with female non-mates. The importance of immune system enhancers and suppressors remains little studied, and much more research is needed to allow firm conclusions about their role in mediating host immune responses during periods of hardship such as mating, reproduction and migration.

It is an inherent assumption that individuals with a superior immune response are also better able to cope with parasites and hence enjoy superior survival. A number of different studies has identified MHC haplotypes that confer resistance to infectious diseases and parasites in humans and domesticated birds and mammals (Apanius et al. 1997). Surprisingly, there is very little empirical information on this subject from free-living birds or other animals. Saino et al. (1997b) demonstrated in an experiment with the Barn Swallow that individuals with a stronger response to a challenge with sheep red blood cells were more likely to survive from one year to the next than individuals with weak responses. Similarly, a study of nestling House Martins Delichon urbica revealed that nestlings with weak T-cell responses to injections with phytohaemagglutinin has a reduced probability of survival compared to individuals with strong responses (Christe et al. 1998). Unpublished studies of House Martins, House Sparrows and Black Wheatears Oenanthe leucura provide similar evidence (Christe et al. unpublished manuscript; Sorci et al. Unpublished manuscript; Soler et al. unpublished manuscript). Hence, elevated immune responsiveness provides individuals with a survival advantage.

Given that we accept the overriding importance of parasites as a selective force in the life of birds, and that resources or specific components of the diet are limiting immune defence, individual birds have to optimise their investment in immune function relative to a host of other activities. I will briefly discuss these activities and provide readers with some central references that can provide further information. Sexual selection and mating are costly activities, particularly for individuals of the sex competing for access to individuals of the choosy sex. Hence, we should expect sexual display to result in a reduction in immune function, if time and/or energy are limiting both components. The example of the barn swallow discussed above provides direct evidence for this prediction. A recent review by Møller et al. (1998a) provides an overview of the literature on parasitism, host immune function and host sexual selection. Marlene Zuk reviews the literature on sexual selection and immune function. Once mating has been achieved, birds may start reproduction, which potentially entails costs in terms of fitness. Hence, birds have to optimise reproduction in order not to compromise immune responsiveness, as suggested by the study by Deerenberg et al. (1997). Lars Gustafsson provides an overview of this topic in a subsequent paper. Offspring of birds usually require parental care for successful development, and the amount of care provided by each of the two parents, as well as the amount of care required by the nestlings relative to what the parents are willing to give, are causes of evolutionary conflict (Trivers 1972, 1974). Immune function may lie at the heart of this conflict since individuals that provide a large share of parental care are likely to pay a cost in terms of reduced immune responsiveness. Nicola Saino provides an overview of this literature. Dispersal is the permanent movement from one site to another during post-natal independence or from one breeding site to another. Dispersal is considered to be costly because it entails the movement from known to unknown sites with their novel competitors, predators and parasites. The role of host-parasite interactions in the evolution of dispersal remains a little studied subject from a theoretical and particularly an empirical perspective (Hamilton 1986, 1993; Frank 1991; Olivieri et al. 1995). Birds demonstrate consistent patterns of dispersal with females dispersing farther than males, and natal dispersal being consistently farther than breeding dispersal (Greenwood 1980; Clarke et al. 1997). There is considerable variability among individuals in dispersal although this variability generally remains unexplained from a functional perspective. It remains an untested possibility that patterns of dispersal such as persistent differences between the sexes, or between the distance of natal and breeding dispersal, may reflect differences in host immunocompetence (Møller 1998). Bird migration has fascinated ornithologists for centuries because of the excessive use of energy for long-distance flight and the ability of individuals to return to their breeding, staging and wintering sites. It has only recently been appreciated that migratory birds differ from residents by encountering at least two very different parasite faunas; that of the breeding grounds and that of the winter quarters (Møller & Erritzøe 1998). This selects for increased investment in immune function, but also for philopatry to breeding, staging and wintering grounds because acquired immunity to specific strains of parasites will favour philopatry over dispersal (Møller & Erritzøe 1998). Population ecology deals with interactions between conspecifics, heterospecifics and the environment. While host-parasite interactions have become popular subjects for population ecologists, there is still little work done on anti-parasite defence at the level of population ecology (Lochmiller 1996). It is easy to imagine that immune responsiveness at the individual level, but also at the population level, will have important implications for the outcome of interactions like those between brood parasites and their hosts, competitors, and predators and their prey. For example, brood parasites are birds that lay their eggs in the nests of hosts that care for and rear the parasitic offspring. Since many parasites are host specific (Price 1980), and since host immune responses are to a great extent determined by resource availability, brood parasites experience a double advantage compared to the offspring of hosts simply because they suffer from few if any host specific parasites and their immune responses are elevated compared to those of their hosts (Soler et al. 1998). This could readily favour this interaction to the benefit of the brood parasite. Learning plays an important role in many activities of birds. For example, song, spatial recognition and a host of other phenomena are directly related to learning. There is increasing evidence from humans and other mammals that parasites significantly disrupt learning ability in hosts (e.g. Kavaliers et al. 1995). Is it the case that individual birds with large song repertoires have elevated learning abilities because they are healthy? Similarly, does the ability of individuals to learn spatial co-ordinates affect their ability to recover stored food, disperse or migrate? These and a number of other questions may be resolved as a consequence of studies of avian immunology, host health status and learning. Genes affecting host immune responses such as those of the major histocompatibility complex and those of the immunoglobulin heavy chain constant gene locus are extremely variable (e.g. Roitt et al. 1996). However, some host populations are genetically depauperate and show little genetic polymorphism at these loci because populations have become diminished and gone through bottlenecks (e.g. Ellegren et al. 1993). This may not be detrimental in the short run, but could threaten the long-term viability of populations. Hence, conservation may seriously depend on the ability of host populations to cope with evolving parasite populations. Island populations of birds are particularly threatened by extinction (Collar et al. 1994), and island populations are also generally characterised by little genetic variability (Frankham 1997). Thus, it is possible that the conservation status of island populations is related to their limited ability to raise immune responses against parasites. In particular, many island populations of birds suffer from the introduction of exotic species that may be carriers of parasites against which the indigenous birds have little ability to defend themselves (van Riper et al. 1986). Conservation efforts may benefit from studies of host immune function since genetically depauperate hosts may have little long-term possibility of sustained viability.

My research was supported by a grant from CNRS (ATIPE BLANCHE).

Andrewartha, H.G. & Birch, L.C. 1954. The distribution and abundance of animals. Chicago; University of Chicago Press: 782 pp.

Andrewartha, H.G. & Birch, L.C. 1984. The ecological web. Chicago; University of Chicago Press: 506 pp.

Apanius, V., Penn, D., Slev, P.R., Ruff, L.R. & Potts, W.K. 1997. The nature of selection on the major histocompatibility complex. Critical Reviews of Immunology 17: 179-224.

Banet, M. 1986. Fever in mammals: Is it beneficial? Yale Journal of Biological Medicine 59: 117-124.

Begon, M., Harper, J.L. & Townsend, C.R. 1986. Ecology: Individuals, populations and communities. Blackwell; Oxford: 876 pp.

Bendich, A. 1989. Carotenoids and the immune response. Journal of Nutrition 119: 112-115.

Berczi, I. & Kovacs, K. (eds) 1987. Hormones and immunity. MTP Press; Lancaster: pp.

Canfield, L.M., Forage, J.W. & Valenzuela, J.G. 1992. Carotenoid as cellular antioxidants. Proceedings of the Society of Experimental Biology and Medicine 200: 260-265.

Chandra, R.K. & Newberne, P.M. 1977. Nutrition, immunity, and infection. Plenum Press; New York: 246 pp.

Chew, B.P. 1993. Role of carotenoids in the immune response. Journal of Dairy Science 76: 2804-2811.

Christe, P., Møller, A.P. & de Lope, F. 1998. Immunocompetence and nestling survival in the house martin: 'The tasty chick hypothesis'. Oikos (in press).

Clarke, A.L., Sæther, B.-E. & Røskaft, E. 1997. Sex biases in avian dispersal: a reappraisal. Oikos 79: 429-438.

Collar, N.J., Crosby, M.J. & Stattersfield, A.J. 1994. Birds to watch 2: The world list of threatened birds. Cambridge; BirdLife International: 407 pp.

Davison, T.F., Morris, T.R. & Payne, L.N. (eds) 1996. Poultry immunology. Abingdon; Carfax: 463 pp.

Deerenberg, C., Apanius, V., Daan, S. & Bos, N. 1997. Reproductive effort decreases antibody responsiveness. Proceedings of the Royal Society of London B 264: 1021-1029

Ellegren, H., et al. 1993. Major histocompatibility complex monomorphism and low levels of DNA fingerprinting variability in a reintroduced and rapidly expanding population of beavers. Proceedings of the National Academy of Science of the USA 90: 8150-8153.

Emlen, J.T. 1986. Responses of breeding cliff swallows to nidicolous parasite infestations. Condor 88: 110-111.

Fitzgerald, L. 1988. Exercise and the immune system. Immunology Today 9: 337-339.

Folstad, I. & Karter, A.J. 1992. Parasites, bright males, and the immunocompetence handicap. American Naturalist 139: 603-622.

Fox, D.L. 1976. Animal biochromes and structural colors. Berkeley; University of California Press.

Fox, D.L. 1979. Biochromy. Berkeley; University of California Press.

Frankham, R. 1997. Do island populations have less genetic variation than mainland populations? Heredity 78: 311-327.

Freitas, A.A. & Rocha, B. 1997. Lymphocyte survival: a red queen hypothesis. Science 277: 1950.

Golub, E.S. & Green, D.R. 1991. Immunology: A synthesis. 2nd edn. Sunderland; Sinauer: 409 pp.

Gonzalez, G., Sorci, G., Møller, A.P., Ninni, P. & de Lope, F. 1999. Immunocompetence and condition-dependent sexual advertisement in male house sparrows (Passer domesticus). Journal of Animal Ecology (in press).

Gray, D.A. 1996. Carotenoids and sexual dichromatism in North American passerine birds. American Naturalist 148: 453-480.

Frank, S.A. 1991. Ecological and genetic models of host-pathogen coevolution. Heredity 67: 73-83.

Gershwin, M.E., Beach, R.S. & Hurley, L.S. 1985. Nutrition and immunity. Orlando; Academic Press: 417 pp.

Glick, B., Day, E.J. & Thompson, D. 1981. Calorie-protein deficiencies and the immune response of the chicken. I. Humoral immunity. Poultry Science 60: 2494-2500.

Glick, B., Taylor, R.L., Martin, D.E., Watabe, M., Day. E.J. & Thompson, D. 1983. Calorie-protein deficiencies and the immune response of the chicken. II. Cell-mediated immunity. Poultry Science 62: 1889-1893.

Goodwin, T.W. 1984. The biochemistry of the carotenoids. Vol. II. Animals. London; Chapman and Hall: pp.

Greenwood, P.J. 1980. Mating systems, philopatry and dispersal in birds and mammals. Animal Behaviour 28: 1140-1162.

Gregus, Z. & Klaassen, C.D. 1996. Mechanisms of toxicity. In: Klaassen, C.D. (ed.) Casarett and Doull's toxicology: The basic science of poisons. 5th edn. New York; McGraw-Hill: 35-74.

Grenfell, B. & Dobson, A.P. (eds). 1995. Ecology of infectious diseases in natural populations. Cambridge; Cambridge University Press: 533 pp.

Grossman, C.J. 1985. Interactions between the gonadal steriods and the immune system. Science 227: 257-261.

Hamilton, W.D. 1986. Instability and cycling of two competing hosts with two parasites. In: Karlin, S. & Nevo, E. (eds) Evolutionary processes and theory. New York; Academic Press: 645-668.

Hamilton, W.D. 1993. Haploid dynamical polymorphism in a host with matching parasites, effects of mutation/subdivision, linkage and patterns of selection. Journal of Heredity 84: 328-338.

Hart, B.L. 1990. Behavioral adaptations to pathogens and parasites: five strategies. Neuroscience and Biobehavioral Reviews 14: 273-294.

Hillgarth, N. & Wingfield, J. 1997. Parasite-mediated sexual selection: endocrine aspects. In: Clayton, D.H. & Moore, J. (eds) Host-parasite evolution: General principles and avian models. Oxford; Oxford University Press: 78-104.

Hochberg, M.E. 1997. Hide or fight? The competitive evolution of concealment and encapsulation in host-parasitoid assocations. Oikos 80: 342-352.

Hoffman-Goetz, L. & Pedersen, B.K. 1994. Exercise and the immune system: a model of the stress response. Immunology Today 15: 382-387.

Johnson, L.S. & Albrecht, D.J. 1993. Effects of haematophagous ectoparasites on nestling house wrens, Troglodytes aedon: Who pays the cost of parasitism? Oikos 66: 255-262.

Kavaliers, M., Colwell, D.D. & Galea, L.A.M. 1995. Parasite infection impairs spatial learning in mice. Animal Behaviour 50: 223-229.

Klein, J. 1990. Immunology. Oxford; Oxford University Press: pp.

König, C. & Schmid-Hempel, P. 1995. Foraging activity and immunocompetence in workers of the bumblebee, Bombus terrestris L. Proceedings of the Royal Society of London B 260: 225-227.

Kraaijeveld, A.R. & Godfray, H.C.J. 1997. Trade-off between parasitoid resistance and larval competitive ability in Drosophila melanogaster. Nature 389: 278-280.

Lack, D. 1954. The natural regulation of animal numbers. Oxford; Clarendon Press: 343 pp.

Lochmiller, R.L. 1996. Immunocompetence and animal population regulation. Oikos 76: 594-602.

Lochmiller, R.L., Vestey, M.R. & Boren, J.C. 1993. Relationship between protein nutritional status and immunocompetence in Northern Bobwhite chicks. Auk 110: 503-510.

Low, B.S. 1978. Environmental uncertainty and the parental strategies of marsupials and placentals. American Naturalist 112: 197-213.

Maynard Smith, J. 1977. Parental investment - a prospective analysis. Animal Behaviour 25: 1-9.

Møller, A.P. 1998. Dispersal, vaccination and regression of immune defense organs. American Naturalist (submitted).

Møller, A.P., Christe, P. & Lux, E. 1998a. Parasite-mediated sexual selection: Effects of parasites and host immune function. Quarterly Review of Biology (in press).

Møller, A.P., Christe, P., Erritzøe, J. & Mavarez, J. 1998c. Condition, disease and immune defence. Oikos (in press).

Møller, A.P. & Erritzøe, J. 1998. Host immune defence and migration in birds. Evolutionary Ecology (in press).

Møller, A.P., Sorci, G. & Erritzøe, J. 1998b. Sexual dimorphism in immune defense. American Naturalist (in press).

National Research Council. 1992. Biologic markers in immunotoxicology. Washington; National Academy Press: pp.

Olivieri, I., Michalakis, Y. & Gouyon, P.-H. 1995. Metapopulation genetics and the evolution of dispersal. American Naturalist 146: 202-228.

Price, P.V. 1980. Evolutionary biology of parasites. Princeton; Princeton University Press: 237 pp.

Roitt, I., Brostoff, J. & Male, D. 1996. Immunology. 4th edition. London; Mosby: 28 chapters. [NO CONTINUOUS PAGE NUMBERS]

Rose, M.E., Payne, L.N. & Freeman, B.M. (eds) 1981. Avian immunology. Edinburgh; British Poultry Science Ltd.: 441 pp.

Saino, N. & Møller, A.P. 1996, Sexual ornamentation and immunocompetence in the barn swallow. Behavioral Ecology 7: 227-232.

Saino, N., Calza, S. & Møller, A.P. 1997a. Immunocompetence of nestling barn swallows in relation to brood size and parental effort. Journal of Animal Ecology 66: 827-836.

Saino, N., Bolzern, A.M. & Møller, A.P. 1997b. Immunocompetence, ornamentation and viability of male barn swallows (Hirundo rustica). Proceedings of the National Academy of Science of USA 94: 549-552.

Sharma, J.M. & Rosenberger, J.K. 1987. Infectious bursal disease and reovirus infection of chickens: Immune response and vaccine control. In: Toivanen, A. & Toivanen P (eds) Avian immunology: Basis and practice. CRC Press; Boca Raton: 143-157.

Soler, J.J., Møller, A.P., Soler, M. & Martinez, J.G. 1998. Interactions between a brood parasite and its host in relation to parasitism and immune defence. Evolutionary Ecology (in press).

Temple, S.A. 1987. Do predators capture substandard individuals disproportionately from prey populations? Ecology 68: 669-674.

Toivanen, P. & Toivanen, A. (eds) 1987. Avian immunology. CRC Press; Boca Raton: pp.

Trivers, R.L. 1972. Parental investment and sexual selection. In: Campbell, B. (ed) Sexual selection and the descent of man, 1871-1971. Aldine-Atherton; Chicago: 136-179.

Trivers, R.L. 1974. Parent-offspring conflict. American Zoologist 14: 249-264.

van Riper, C., van Riper, S.G., Goff, M.L. & Laird, M. 1986. The epizootiology and ecological significance of malaria in Hawaiian land birds. Ecological Monographs 56: 327-344.

Wakelin, D. 1996. Immunology to parasites. 2nd edn. Cambridge; Cambridge University Press: 204 pp.

Willis, G.M. & Baker, D.H. 1981. Interaction between dietary protein/amino acid level and parasitic infection: Morbidity in amino acid deficient or adequate chicks inoculated with Eimeria acervulina. Journal of Nutrition 111: 1157-1163.