What color morph male of side blotched lizard has the most success forming strong pair bonds in western side blotched lizard mating behavior?

Evolution of a New Ovulation Character in House Finches

Within 36 years (36 generations) a whole population of house finches (Carpodacus mexicanus) in Montana, at the northern limit of species' range, evolved a novel ovulation trait, a maternally determined sex-specific ovulation sequence: about 90% of the first-laid eggs are females and ca. 80% of the second-laid eggs are males (Badyaev et al. 2006). The extremely short period of the time during which the evolution of the trait occurred and the fact that it took place within the range of species, that is, under conditions of sympatry and gene flow, excludes the possibility of involvement of gene mutations, gene drift, or genetic recombination.

Such evolutionary changes are related to specific fluctuations in the levels of the pituitary prolactin (PRL) and androgens in the circulation of mothers during oogenesis. Levels of these hormones, which are ultimately regulated by the bird's brain, influence segregation of sex chromosomes during the first meiotic division, thus determining the genetic sex of the egg; oocytes that grow during high androgen levels develop into male eggs and those that happen to grow during high PRL—into female eggs (Badyaev et al., 2005). Hence, it is a neurally controlled change in the patterns of release of these hormones by mothers, rather than any changes in genes or DNA, that determines the evolutionary change of the sex-biased egg laying.

Regulation Through Sex-Biased Egg Laying

Another way of differential deposition of maternal factors in eggs has been recently observed in finches. While the mother modulates the level of steroids in blood, it simultaneously exhibits a sex-biased egg-laying order, thus leading to their differential sex-specific allocation in eggs.

A very young, 36-year-old population of house finches (Carpodacus mexicanus) in Montana, at the northern limit of species’ range, exhibits maternally determined sex-specific ovulation sequence: about 90% of the first-laid eggs are females, and about 80% of the second-laid eggs are males. This differential ovulation according to the sex was not observed in house finches of the same species in the southern extreme of the range, in Arizona. Differences between the two northern and southern populations of house finches were observed also in the temporally differential allocation of yolk in Montana but not Arizona (Badyaev et al., 2006). Let us remember that ovulation in vertebrates is under strict neurohormonal control.

Emerging virus infections

Small Passeriformes play a minor role in the spreading of the emerging viral diseases West Nile virus (WNV) and avian influenza (AI).

The American crow (Corvus brachyrhynchos) plays the most obvious role as a reservoir of the mosquito-transmitted WNV in the US. The ability of the invading NY99 strain of WNV to elicit an elevated viraemia response in Californian passerine birds is critical for the effective infection of Culex mosquitoes. Of the bird species tested, Western scrub jays, Aphelocoma coerulescens, produced the highest viraemia response, followed by house finches, Carpodacus mexicanus, and house sparrows, Passer domesticus. Most likely, few mourning doves (Zenaidura macroura) or common ground doves (Columbina passerine) and no California quail, Callipepla californica, or chickens would infect blood-feeding Culex mosquitoes. All Western scrub jays and most house finches succumbed to infection (Reisen et al 2005). House finches and English sparrows are competent hosts for both West Nile and St Louis encephalitis viruses and frequently become infected during outbreaks. Although mortality rates were high during initial infection with West Nile virus, prior infection with either virus prevented mortality upon challenge with West Nile virus (Fang & Reisen 2006).

Infections with influenza virus have been reported in finches and in imported mynahs. An avian influenza A virus of the subtype H7N1 was isolated in summer 1972 from a single free-living siskin (Carduelis spinus Linnaeus, 1758). Additional cases of morbidity or mortality were not observed in the area were the sick siskin was found. The virus induced following experimental inoculation of chicken embryos a high rate mortality (mean death time approximately 24 hours). This virus was considered as a highly pathogenic avian influenza A virus. Canaries that were housed in the same room with the siskin were accidentally exposed by contact to the sick siskin, which resulted in virus transmission followed by conjunctivitis, apathy, anorexia and a high rate mortality (Kaleta & Hönicke 2005). The role of passerines in the spread of avian influenza A is, however, negligible. A total of 543 migrating passerines were captured during their stopover on the island of Heligoland (North Sea) in spring and autumn 2001. They were sampled for the detection of avian influenza A viruses (AIV) subtypes H5 and H7, and for avian paramyxoviruses serotype 1 (APMV-1). For virus detection, samples were taken from (a) short-distance migrants such as chaffinches (Fringilla coelebs, n=131) and song thrushes (Turdus philomelos, n=169), and (b) long-distance migrants such as garden warbler (Sylvia borin, n=142) and common redstarts (Phoenicurus phoenicurus, n=101). Virus detection was done on conjunctival, choanal cleft and cloacal swabs. In none of the tested samples was AIV detected. Six out of 543 birds (1.1%) were found to carry non-pathogenic and lentogenic strains of APMV-1. This indicates that the passerine species examined in this study may play only a minor role as potential vectors of APMV-1 (Schnebel et al 2005). In another study 413 free-living migrating passerines from 37 different species were caught in the autumn of 2004 in Slovenia and cloacal swabs taken and processed by RT-PCR and virus isolation. Only one sample from a common starling (Sturnus vulgaris) by RT-PCR was positive for AIV, but negative for H5 and H7 (Račnik et al 2007).

Evidence for West Nile virus infection in white storks and their potential role as a host reservoir

The three factors that are critical to the ability of a bird species to act as a host reservoir for an arthropod-borne virus are firstly the ability of the virus to replicate in that bird species to give a viremia, including the duration and titer resulting from infection, secondly the generation of immunologically naive individuals that could become infected and hence serve as hosts, and thirdly a high survival rate after infection (Carrara et al., 2005). There is little information on the WNV viremia in the white stork, although there is extensive data from experimental infection of other avian species. Thus, for example, Reisen et al. (2006) reported a variable viremic response to WNV in house sparrows (Passer domesticus) and house finches (Carpodacus mexicanus) compared to white-crowned sparrows (Zonotrichia leucophys) and that infection rates of Culex tarsalis females were low at bird titers < 6.5 log10 plaque forming units (pfu) / ml. Although not susceptible to disease, inoculation of chickens using both infected mosquitoes and subcutaneous infection resulted in a transient viremia but at a low level of < 4.0 log10 pfu/ml, insufficient to infect mosquitoes (Langevin et al., 2001; Komar et al., 2003). A wide-ranging study inoculating a range of captive-bred and wild-caught avian species with WNV demonstrated that infection in many species, but particularly Passeriformes, resulted in a four to five day viremia with titers as high as 10 log10 pfu/ml (Komar et al., 2003). Such levels would be sufficient to infect mosquitoes. These and other studies suggest that a species susceptible to WNV infection would have a transient viremia of > 7 log10 pfu/ml lasting between one and six days from infection. The second factor, the appearance of susceptible individuals, is driven primarily from bird breeding. White storks have up to four chicks per brood, which if they survive to fledging will provide a major component of the southerly migrating bird flocks. As immunologically naïve individuals, it is these that are most likely to become infected. However, the birth rates of large birds such as storks is much lower than those of rodents, which serve as host reservoirs for Venezuelan equine encephalitis virus (VEEV) for example (Carrara et al., 2005). Thus birds are unlikely to represent a constant reservoir of infection unlike rodents which breed all year in tropical regions.

The adaptation of arboviruses to replicate at elevated temperatures facilitates utilization of new host species, for example birds (Brault, 2009), which may be important in range expansion. Of considerable interest is the increased replication efficiency of different WNV strains in vertebrate hosts at elevated temperatures. Thus, replication of a WNV strain from Kenya in cell culture was reduced 6,500-fold at 44°C relative to levels at 37°C, while replication of the strain of WNV originally introduced into North America (NY99) was only reduced by 17-fold (Kinney et al., 2006). Kinney et al. (2006) suggest that the ability of the natural temperature-resistant strain, NY99, to replicate at high temperatures could be important in the increased avian virulence of the NY99 genotype. Thus, mean body temperatures of WNV NY99-infected American crows ranged between 40.5 and 44.5°C (Kinney et al., 2006) potentially giving the NY99 strain a clear replicative advantage in transmission within that bird population.

4.1 Diurnal Shedding Patterns

For intestinal coccidia to be transmitted from one host to the next, they have to be shed in the faeces of their current host. Moreover, the shedding of oocysts by passerine birds has a strong circadian pattern that appears to be a ubiquitous phenomenon. To our knowledge, in all passerines tested to date, oocyst shedding peaks in the afternoon or evening. In our review of the literature, evidence of a circadian rhythm was found for English house sparrows (P. domesticus; Boughton, 1933), cliff swallows (Petrochelidon pyrrhonota; Stabler and Kitzmiller, 1972), canaries (Serinus canarius; Box, 1977), chaffinches (Fringilla coelebs; Gryczyńska et al., 1999), house finches (Carpodacus mexicanus; Brawner and Hill, 1999), dark-eyed juncos (Junco hyemalis; Hudman et al., 2000), greenfinches (C. chloris; Brown et al., 2001), starlings (Sturnus vulgaris; Dolnik, 2002), scarlet grosbeaks (Carpodacus erythrinus; Dolnik, 2002), reed warblers (Acrocephalus scirpaceus; Dolnik, 2002), willow warblers (Phylloscopus trochilus; Dolnik, 2002), blackbirds (Turdus merula; Misof, 2004), serins (Serinus serinus; López et al., 2007), garden warblers (Sylvia borin; López et al., 2007), regent honeyeaters (Xanthomyza phrygia; Morin-Adeline et al., 2011) and green-winged saltators (Saltator similis; Coelho et al., 2013). In some instances, the difference in oocyst shedding can vary from detection of no oocysts whatsoever shed in the morning to several hundreds of thousands in the afternoon (Brawner and Hill, 1999). Acknowledging the variation in circadian rhythms of oocyst shedding is important, otherwise studies that do not account for this temporal variation may draw spurious conclusions.

Multiple hypotheses have been suggested to explain the diurnal patterns of coccidian oocyst shedding. The synchronization of oocyst discharge may be necessary to increase the concentration of oocysts within faeces to levels that are sufficient to ensure transmission to the next host, if successful transmission is dosage dependent (Dolnik et al., 2011). Furthermore, oocyst shedding may be timed to enhance transmission by increasing the number of sporulated oocysts so as to coincide with a higher density of hosts during periods of active feeding activity, as is suggested to occur in the case of the blackbird (T. merula; Martinaud et al., 2009). A peak in afternoon shedding may also be an adaptation to increase oocyst survival under external environmental conditions. Strong support for the validity of this last-mentioned theory is that oocysts experimentally exposed to sunlight had a significantly lower survival rate than those left in shade; 1 hour of sunlight decreased oocyst survival by approximately 50% (Martinaud et al., 2009). These authors found that oocysts were more susceptible to UVB than UVA radiation (Martinaud et al., 2009). Their findings are also congruent with the observation that diurnal periodicity in oocyst discharge persists in hosts inhabiting regions at high latitudes (Dolnik et al., 2011). Despite long hours of daylight, UV radiation levels still fluctuate greatly throughout the day, peaking around midday and declining in the afternoon (Dolnik et al., 2011). Coupled with an absence of diurnal patterns in the activity of snow buntings, these authors reasoned that circadian oocyst shedding was most likely an adaptation to survive conditions outside the host. Martinaud et al. (2009) came to the same conclusion but, in addition, stressed that the number of viable oocysts present in the subsequent morning as a result of afternoon shedding was also likely to be the result of adaptive timing to coincide with the feeding activity of blackbirds.

Constraints and Costs – The Handicap Principle

Despite the potential benefits that signalers stand to gain by misleading receivers, the majority of signals seem to convey at least some reliable information. In other words, stable honesty is the usual outcome of the arms race between signaler and receiver. How can such honesty persist in the face of the evolutionary conflict between the two?

Honesty may endure if it is physically impossible for a signaler to produce a misleading signal. In some cases there is a material link between the signal and some underlying aspect of the signaler’s state, such that the former is constrained to provide information about the latter. For instance, many animals display bright coloration attributable (in part at least) to carotenoid pigments. Since most cannot synthesize carotenoids themselves but must acquire them as part of their diet, the intensity of carotenoid coloration is inescapably linked to (past) foraging success. An individual that fails to acquire sufficient carotenoids through foraging simply cannot manufacture the pigments needed for vivid color. Studies of the house finch (Carpodacus mexicanus), for instance, a species in which males display patches of carotenoid pigment on the crown, breast, and rump, have shown that females prefer redder males and that vivid coloration reflects nutritional state at the time of molt. Since molting precedes mate choice by some time, redder males may not necessarily be in better condition when they are actually chosen as partners, but plumage coloration was correlated with nest attentiveness and overwinter survival, suggesting that brighter males do make, on average, better fathers.

While a few signals may be ‘unfakable’ because of such physical constraints, most are not. An individual’s choice of one threat display or another, for instance, is not physically constrained by its fighting ability or motivation, nor is the length of a bird’s tail inescapably tied to its value as a mate. To account for widespread honesty, therefore, an explanation is required that is applicable to situations in which the relationship between the signal and the underlying state of the signaler is open to behavioral or evolutionary change.

Amotz Zahavi was the first biologist to propose such a general explanation for honesty. He suggested that a signal could provide reliable or ‘honest’ information about the quality of the signaler if it were costly to produce. Consider, for instance, a population in which males reliably advertise their value as mates by means of some sexual display. The correlation between signal and mate value would seem vulnerable to disruption by the spread of a deceitful mutant that adopts the display typical of the most desirable males, regardless of its own quality. Suppose, however, that the signal is costly to produce but that superior males can more easily bear these costs than can inferior males. Under these circumstances, it might pay an inferior male to refrain from signaling, even though it would be physically able to do so, because the costs involved would outweigh the benefits to be gained. At the same time, for superior males who can more easily bear the costs of display, production of the signal would yield a net benefit. Under these circumstances, honesty can persist.

While Zahavi initially focused chiefly on sexual displays, he has subsequently argued that the handicap principle is applicable to many and perhaps all forms of communication. Costly signals could serve to advertise fighting ability, for instance, and thereby to deter rivals, or could deter predators by advertising ability to escape or fend off attack. Costly contributions to group activities could, in a similar way, advertise social status, provided that only dominant, high-quality individuals can afford to invest time and energy in such pursuits. In short, receivers in most contexts can obtain reliable information about signaler quality by attending to displays that are costly to produce.

Experimental Evidence for a Role of Testosterone as a Proximate Regulator of Mating Systems

So far, we looked at comparative data to investigate whether there is a relationship between testosterone concentrations and mating system. But is there experimental evidence that testosterone is a proximate factor for the expression of different mating strategies so that mating systems can be manipulated using testosterone treatment? Once again, most investigations have been done in birds. Testosterone implants increased the likelihood of males to become polygynous or show extra-pair behavior in song sparrows (Melospiza melodia), white-crowned sparrows (Zonotrichia leucophrys), dark-eyed juncos (Junco hyemalis), starlings (Sturnus vulgaris), mallards (Anas platyrhynchos), and red grouse (Lagopus lagopus), but not in pied flycatchers (Ficedula hypoleuca), house finches (Carpodacus mexicanus), spotless starlings (Sturnus unicolor), red-winged blackbirds (Agelaius phoeniceus), and blue tits (Cyanistes caeruleus), although male blue tits with testosterone implants showed a greater interest in interacting with females other than the one they were paired with. A potential confound for studies that investigate the effect of testosterone on the genetic mating system is that treatment with testosterone may lead to a shutdown of the internal production of testosterone and sperm. This is a pharmacological effect that should be kept in mind when comparing the within- and extra-pair fertilization success of testosterone-treated males with controls.

Additional evidence for an effect of testosterone in the proximate control of mating strategies comes from male side-blotched lizards (Uta stansburiana) that exist in several color morphs related to mating tactics. Adult males with an orange throat have high levels of testosterone, are highly aggressive, and defend large territories that overlap with the territories of multiple females. Adult males with a blue throat have intermediate levels of testosterone, are less aggressive, and defend small territories overlapping with those of few females. Adult males with a yellow throat express low levels of testosterone, do not defend territories but mimic females, and sneak copulations. Throat color has a genetic basis and is influenced by hormone levels during development. But blue- and yellow-throated males implanted with testosterone during adulthood start to defend territories which are as large as those of orange-throated males.

Whereas testosterone seems to activate mating strategies in adult side-blotched lizards, organizational effects of testosterone play a major role in a closely related species, the tree lizard Urosaurus ornatus. Also, tree lizards come in several color morphs: males with an orange-blue dewlap are aggressive and defend territories, whereas males with an orange dewlap are nonaggressive and do not establish territories. Males with high levels of testosterone and progesterone during ontogeny develop into the orange-blue morph which become territorial and defend the home range of several females. Males with low levels of testosterone and progesterone during development turn into the orange morph, which follow a sneaker or nomadic strategy. Adult circulating levels of testosterone do not differ between orange-blue and orange tree lizards and – unlike in side-blotched lizards – mating strategies are fixed and cannot be manipulated via testosterone implants. Thus, the major hormonal effects on mating strategies seem to occur at different times during the life history of tree- and side-blotched lizards. Organizational effects prevail in tree lizards, whereas side-blotched lizards remain plastic and mating strategies can be manipulated with hormones during adulthood. In summary, these data suggest that testosterone may facilitate polygynous mating strategies in some species, but since this is not universally the case, testosterone alone cannot explain mating decisions.

An important factor that has been largely neglected in the discussion of proximate factors for the control of mating strategies is the female part: it always takes two to tango and it takes even more to become polygamous or promiscuous. Thus, treatment with testosterone may increase the propensity of males to seek additional mates, but this does not mean that females of all species, and under all circumstances, are ready to accept these offers. There is a large body of ecological literature discussing factors that might influence the decision of females to become the secondary mate of a polygynous male. Such factors include for example male quality, territory quality, availability of unpaired males, etc. But we know very little about the physiology of such females. Potential hormonal factors that lead to or prevent the formation of a pair bond between females and males are discussed in the next section.

3.4 Differential Regulation of Yolk and Circulating Levels

Groothuis and Schwabl (2008) distinguish three potential ways in which levels of hormones in the circulation and the yolk may be linked. First, yolk hormones may simply reflect levels in the circulation regulating female reproductive physiology and behavior. According to this ‘physiological epiphenomenon’ hypothesis, yolk and circulating levels would be positively correlated. Secondly, the ‘flexible distribution hypothesis’ assumes that hormones may be allocated either to the circulation or to the yolk, resulting in a negative correlation between circulating and yolk levels. Third, the ‘independent regulation hypothesis’ suggests that both production and distribution can be regulated, so that circulating and yolk levels are not necessarily correlated. Only the last mechanism would allow an unconstrained evolution of separate adaptive effects on mother and offspring. Actual measures of levels in the circulation and the yolk find a positive correlation (domestic canary (Schwabl, 1996a); house finch (Carpodacus mexicanus) (Badyaev et al., 2005)), no correlation (domestic canary (Tanvez, Beguin, Chastel, Lacroix, & Leboucher, 2004); domestic pigeon (Columba livia f. domesticus) (Goerlich, Dijkstra, Schaafsma, & Groothuis, 2009)), or a negative correlation (black-backed gull (Larus fuscus) (Verboven et al. 2003); house finch (Navara, Siefferman, Hill, & Mendonça, 2006)) between circulating and yolk hormone levels, suggesting that independent regulation is at least potentially possible (for a detailed discussion see Groothuis & Schwabl, 2008). Finally, in house sparrows (Passer domesticus) (Mazuc, Bonneaud, Chastel, & Sorci, 2003) there were opposite correlations at different levels: both yolk and circulating levels increased with breeding density, but within females a negative correlation between yolk and circulating levels was found. When the gonadal axis of female dark-eyed juncos (Junco hyemalis) was stimulated by GnRH during different phases of the reproductive cycle (Jawor et al., 2007), circulating T increased only in the week before egg laying and, although yolk and circulating levels were not correlated, the magnitude of the response to GnRH correlated with yolk T levels, suggesting that the responsiveness of the reproductive axis influences steroid levels in both mother and offspring. Perhaps the strongest evidence for a differential regulation is the fact that relative levels of different hormones differ strongly between circulation and the yolk (Groothuis et al., 2005b). Estradiol and CORT are found only in very low concentrations in the egg, but in relatively high concentrations in the circulation, and the ratio of P4 to different androgens differs between maternal circulation and yolk.

Overall, this indicates that various factors contribute to hormone levels in the circulation and the yolk and that there is no simple relationship. A direct comparison between plasma and yolk concentrations is also not easy because of the different dynamics of circulating concentrations, which can fluctuate rapidly, and yolk concentrations, which accumulate over several days. Therefore, single blood samples taken at specific stages during egg laying may not reflect the accumulation of hormones in the yolk. Further, plasma levels are the product of all follicles, whereas levels in each follicle may depend mostly upon production in the follicle itself and less by adjacent follicles. Finally, as mentioned before, conclusions from implants demonstrating transfer to the egg are difficult to draw, since circulating levels are generally increased above the physiological range, which could result in a transfer to the yolk that would normally not occur. However, even in the case that hormone levels in the mother and egg correlate, this does not rule out potential specific adaptive roles for these hormones both in the circulation and in the yolk: if external factors are functionally relevant for both mothers and offspring, correlated levels of hormones in circulation and yolk may carry useful information for both.

Processes that could affect relative levels in the circulation and the yolk are differences in solubility between hormones, presence of converting enzymes in the yolk, differential degradation of hormones in the circulation (liver) and yolk, and specific binding proteins or binding of steroids to lipids that are transported into the egg (Groothuis & Schwabl, 2008). In birds, androgens are bound to corticosteroid-binding globulins (CBGs) and to albumin since there is no specific protein binding sex steroids. Binding globulins therefore can regulate hormone levels, and circulating CORT levels may influence free T levels or vice versa by competing for CBG (Swett & Breuner, 2008). Several proteins that bind THs are thought to be involved in active transport of these hormones into the yolk (McNabb & Wilson, 1997).

1 Carotenoid Pigmentation

Carotenoid pigments are responsible for the red and yellow colors seen in many vertebrate and invertebrate animals (Goodwin, 1950). These pigments come from food, rather than being produced by the metabolism; thus, although the distribution of carotenoids on the organism’s body is generally inherited (Brush and Siefried, 1968), the intensity of coloration depends on the diet (Goodwin, 1950; Brush, 1978). Carotenoid-pigmented characters, therefore, are condition-dependent, not absolute, and they may play a special role in sexual selection (Kodric-Brown and Brown, 1984; Andersson, 1986).

A study of house finches (Carpodacus mexicanus) in Michigan (Hill, 1990) suggests that females may pay close attention to these characters, and argues for increased examination of carotenoid-based secondary sex characters. Male house finches vary considerably in the intensity of the reddish coloration on the breast and head feathers, even within age classes (Hill, 1990). In both field and laboratory work, Hill (1990) showed that females preferred the reddest of four males, whether the coloration occurred as part of natural variation in intensity or was the result of experimental lightening and reddening of the feathers using hair dyes (Fig. 1).

Two implications of this research are relevant here. The first is that females used a condition-dependent trait to make mating decisions, one that indicates a certain level of competence or quality on the part of the male. Condition-dependent traits, because they are costly, prevent “cheating” on the part of a male with bright colors but low fitness. In the context of interest, “fitness” is resistance to parasites. The resistance itself, which is favored by selection, is genetically fixed, not facultative. The trait indicating this fitness, however—bright plumage coloration in this case—is only expressed when its bearer is in good enough condition to do so.

This brings us to the second interesting implication, which is that carotenoids and carotenoid-pigmented structures are particularly affected by parasites. Poultry scientists and breeders have known for many years that skin and comb color in fowl were pigmented with carotenoids obtained in feed, and that the amount of carotenoid pigment observed was related to factors such as fecundity (Palmer and Kempster, 1919a,b,c; Stone et al., 1971). In chickens infected with coccidia, protozoan parasites that can cause severe juvenile mortality, coloration of these carotenoid-dependent tissues was lighter than in uninfected birds, even though the diseased individuals received the same amount of dietary carotenoids as healthy chickens (Ruff et al., 1974). The mechanism by which pathogens interfere with carotenoid metabolism is not fully understood, but the phenomenon appears to be true for parasites other than coccidia (Bird, 1952; Henderson, 1951). In red jungle fowl, females preferred males with more brilliantly colored eyes and combs, both carotenoid-dependent traits and both influenced by nematode gut parasites (Zuk et al., 1990a). Several studies of fishes have shown the influence of carotenoid-derived skin coloration on female choice, with females often preferring the more intensely colored male (Kodric-Brown, 1989; Endler, 1983).

The interaction of these factors of disease and diet to produce condition-dependent ornamental traits is an extremely interesting area and one that deserves more study. If females use a condition-dependent trait such as a carotenoid-pigmented skin patch or feather color as a basis for mate choice, but these characters can only be produced by parasite-resistant males who also obtain the appropriate diet, the females may be getting both a healthy male with so-called “good genes” and one who either knows where to forage, has a good territory, or otherwise possesses the capability of transforming carotenoids in the environment into signals (Fig. 2).

It remains to be seen if these two contributors, foraging ecology and parasite resistance, have a compensatory or perhaps additive relationship. Will a male who is exceptionally good at foraging but still susceptible to parasites, for example, be as successful in attracting females as one who is above average in both capacities or one who is exceedingly healthy but not as good at obtaining carotenoids in the diet? At this stage, the nature of the relationships in Fig. 2 can only be speculative, but the potential exists for new links between such previously disparate topics as diet choice and mate choice, with parasites acting as a go-between. In any event, it is important to note that the operation of this mechanism by no means obviates the operation of Hamilton’s and my model; it presents a way in which sexual ornaments may be influenced by parasites, but not the reasons behind the process occurring.

Bacterial Diseases

Avian mycobacteriosis is a common problem in passerines. It is most commonly caused by Mycobacterium avium, M. intracellulare, and M. genavense, although other opportunistic mycobacterial species may also cause disease.35 The disease is characterized by granulomatous inflammation in any organ, but the gastrointestinal (GI) tract, liver, spleen, bone marrow, and lung are common sites of infection.5 A vascular form of the disease characterized by aortitis and cardiopulmonary arteritis has been described and anecdotally seems to be a relatively more common presentation in fairy bluebirds (Irena puella) as well as in other passerine species.16,25 Birds affected by avian mycobacteriosis usually present with a marked leukocytosis and nonspecific signs of illness such as lethargy, weakness, anorexia, and chronic weight loss. Diagnosis is aided through identification of acid-fast positive rods in cytologic or histologic specimens, although culture is required for definitive diagnosis. Polymerase chain reaction (PCR) testing may also aid in identification and speciation of some mycobacterial organisms. The site of infection and intermittent shedding of bacteria may make antemortem diagnosis challenging. Recent epidemiologic studies suggest that avian mycobacteriosis is more likely an opportunistic pathogen acquired from the environment, rather than a pathogen that is directly transmitted from bird to bird.37,45 As treatment is generally unrewarding and not recommended in zoologic settings because of the zoonotic potential, management of the disease is aimed at prevention and reducing stress and other causes of immunosuppression in passerines.

Mycoplasma gallisepticum is the causative agent of mycoplasmal conjunctivitis in passerines. The disease is primarily associated with wild house finches (Carpodacus mexicanus), although many other passerine species are reported to be affected.22 Some passerines such as American goldfinches (Spinus tristis) may be subclinically infected and may serve as potential reservoirs.8 Clinical signs consist primarily of conjunctivitis that presents as periocular swelling and upper respiratory tract exudate that may form a dried crust on the head and face. Affected birds may lose vision and then starve because of inability to acquire food. Transmission occurs both vertically and horizontally, through direct contact, aerosolization, and fomites.22 The presence and style of bird feeders have been shown to affect the transmission of the disease in wild bird populations.12 Mycoplasma sturni has been found in association with acute conjunctivitis with focal mucosal ulceration in several species of wild passerines.22 However, the bacteria may also be found in individuals without clinical disease, so its role in the disease remains uncertain.

Avian chlamydiosis is caused by Chlamydophila psittaci and affects a wide range of avian hosts, including passerines. The disease may cause acute mortality or nonspecific signs of illness such as fluffed feathers, lethargy, weakness, anorexia, and abnormal droppings. Signs of upper respiratory involvement such as oculonasal discharge and conjunctivitis may also be present. Marked leukocytosis, hepatomegaly, and splenomegaly are diagnostic features that may suggest chlamydiosis, although other etiologies should also be considered. Diagnosis is often achieved with antigen testing such as PCR or immunofluorescent antibody (IFA) testing on blood samples, conjunctival/choanal/cloacal swabs, or tissues. C. psittaci is a zoonotic disease causing influenza-like clinical signs in humans. Therefore, routine Chlamydophila sp. screening and quarantine measures are warranted for most bird collections, particularly those that feature walk-through aviaries.

Salmonellosis is a significant disease of wild and captive passerines.6,19 Many species and serotypes of Salmonella have been isolated from birds, with Salmonella typhimurium being most often associated with clinical disease. Clinical signs include diarrhea and nonspecific signs of disease. Disease severity may be variable and intermittent, including subclinical infections. Septicemia may occur with bacterial lesions occurring in multiple organ systems. Necropsy findings include weight loss, hepatomegaly, splenomegaly, and necrotic inflamed pale foci in the GI tract and other affected organs. Treatment may be difficult and may result in the creation of subclinical carriers. Outbreaks are frequently associated with bird feeding stations, so efforts should be taken to minimize fecal–oral contamination. Yersiniosis, caused by Yersinia pseudotuberculosis, causes similar clinical signs and gross lesions in passerines and may be differentiated through bacterial cultures.

Other bacterial pathogens that are also associated with passerines include Campylobacter jejuni, which is reported to cause GI signs and nestling mortality in some passerines, including the Gouldian finch. Many species of passerines may be subclinical carriers, however, and serve as a source of infection for other species. Society finches (Lonchura domestica) with subclinical disease have been implicated as sources of infection when used to foster Gouldian finch nestlings.32 Avian cholera, caused by Pasteurella multocida, in passerines is most often associated with cat bites.36 Without rapid treatment, septicemia may result and be rapidly fatal. Erysipelas, caused by Erysipelothrix rhusiopathiae, has been associated with disease in wild passerines, including the endangered Hawaiian crow (Corvus hawaiiensis).48 Few clinical signs and gross lesions are present because of the rapidly fatal septicemia that results from infection.47 Although they usually do not demonstrate clinical disease, passerines are known to play a significant role in the life cycle of many Borrelia spp.26