Introduction
Egg dumping is known with several names such as, intraspecific brood parasitism, conspecific brood parasitism, intraspecific nest parasitism and they all refer to the even when a female bird lays her eggs in another female' s nest, that belongs in the same species, and does not take part in caring for the eggs and later the young. Another much less commonly used name for this behavior is intraspecific brood amalgamation ( Eadie et al., 1988) which is considered to be value-free. In other words, parasitism is thought to imply that the behavior is solely beneficial to the parasite and costly to the host, which might not necessarily be the case in general. In this review, I will be examining the kind of species that exhibit this behavior, what determines it and why. Also, I will be looking at some of the methods that researches use to study this unique reproductive behavior.
Species
In one of the first review studies of intraspecific brood parasitism, Yoram Yom-Tov (1980), presents one of the earliest attempts to catalog the avian species that exhibit this unique behavior. Mostly based on crude and ineffective methods of identifying parasitism in birds, the results present a mere 53 species of birds, most of which belong to the order of the Anseriformes. Other orders mentioned are the Galliformes, Passeriformes and Columbiformes. However, in the next 20 years, it became apparent how unappreciated is intraspecific brood parasitism and Yom-Tov (2001) provides an updated list of the known brood parasitic bird species. In this review, the results identified a total of 234 avian species, with the Anseriformes representing the largest group, 74 species, and the Passeriformes with 66 species.
The literature that I am reviewing focused mostly on the following species, goldeneye ducks (Bucephala clangula), cliff swallows (Petrochelidon pyrrhonota), maned ducks (Chenonetta jubata), canvasbacks (Aythya valisineria), lesser snow geese (Chen Caerulescens Caerulescens), bar-headed geese (Anser indicus), European starlings (Sturnus vulgaris), American coots (Fulica americana), wood ducks (Aix sponsa), swallows (Hirundo rustica), barnacle geese (Branta leucopis) and lastly the redhead duck (Aythya americana) (Ahlund and Andersson, 2001; Briggs, 1991; Brown and Brown, 1998; Forslund and Larsson, 1995; Lank et al., 1989; Lyon, 1993; Moller, 1987; Sandell and Diemer, 1999; Semel and Sherman, 2001; Sorenson, 1991; Stouffer and Power, 1991; Weigmann and Lamprecht, 1991). All the above species exhibit intraspecific brood parasitism.
Out of all these species, the majority of them were either colonial or semi-colonial, such as the cliff swallow, the bar-headed goose, the European starling, the lesser snow goose and others. In addition to their social grouping, another important characteristic is the state of the young chicks after hatching. There was a total of five species that exhibited precociality, the goldeneye, canvasbacks, bar-headed goose, American coots, and wood ducks and a total of six species that exhibited altriciality, cliff swallows, maned ducks, lesser snow goose, European starlings, swallows, and redhead duck. Finally, the type of parenting exhibited by the species is also very crucial when discussing intraspecific brood parasitism and there were a total of 7 species with biparental care for their youngs and only 3 species with uniparental care for their young (Ahlund and Andersson, 2001; Briggs, 1991; Brown and Brown, 1998; Forslund and Larsson, 1995; Lank et al., 1989; Lyon, 1993; Moller, 1987; Sandell and Diemer, 1999; Semel and Sherman, 2001; Sorenson, 1991; Stouffer and Power, 1991; Weigmann and Lamprecht, 1991). As I will examine in greater detail later in my review, all the above characteristics play a crucial proximate role in the existence of intraspecific brood parasitism.
Methods
As I mentioned above, it was not until rather recently that the true extend of intraspecific brood parasitism's occurrence was realized. This can be mainly attributed to the fact that it is a lot harder to be detected since the host's and the parasite's hatchlings are largely undistinguishable. In order to overcome this problem of detecting intraspecific brood parasitism researchers have adopted an array of techniques that can assure reliable results.
The most widely used method to identify parasitism involved daily nest checking. The vast majority of the articles mentioned the appearance of two or more eggs in a nest per day as one of the criteria that determined whether or not the nest was parasitized (Briggs, 1991; Brown and Brown, 1998; Lank et al., 1989; Lyon, 1993; Moller, 1987; Sandell and Diemer, 1999). Even though this is based on the fact that most females birds do not lay more than one egg per day, it does not take into account the fact that parasites are known to remove eggs prior to laying their own (Sandell and Diemer, 1999) and that hosts are known to reject parasitic eggs from their clutch (Lyon, 1993). Also, the appearance of eggs before or after the host female initiated and ended laying its own clutch was considered parasitism (McRae, 1998; Sandell and Diemer, 1999). All the above methods involved a lot of frequent nest visiting, which could result in nest desertion by the birds because of the increased disturbance. One way of avoiding this would be adopting time-lapse photography as part of the observation efforts (Sorenson, 1993).
Some of the less used methods of identifying parasitism involve gel electrophoresis and DNA fingerprinting, mostly because of their high degree of difficulty to be performed. Gel electrophoresis could be used to identify any protein polymorphisms that render a young as parasitic (Yom-Tov, 1980) and DNA fingerprinting could not only identify parasitic young but also assign them to their true mother (Petrie and Moller, 1991). Both of these methods, however, would involve a great deal of nest visiting which would once again possibly deter the females' behavior. One very interesting method that was used by Stouffer and Power (1991), was injecting females with tetracycline chelates with calcium ions which caused tissue that was composed of calcium, including egg-shells to fluorescent. In doing so, they manage to identify the parasitic eggs among a clutch. Finally, I believe that the most effective but also the most difficult and tedious method would be daily and observations of nests either directly or using time-lapse photography because events such as egg rejection by the host, egg removal by the parasite can be identified and be accounted for.
Proximate Causations
In order to fully understand intraspecific brood parasitism, researchers set out to determine which are the ecological and the physiological factors that might affect its occurrence. However, these are just proximate factors, in the sense that they did not fully explain the evolutionary significance of the behavior but could only be associated with it.
First of all, the determinant that is most commonly associated with intraspecific brood parasitism is the visibility of the females' nests. In other words the easier it is for the female to spot another female's nest the more likely it is that they will exhibit brood parasitism. For instance, in McRae's (1998) study female moorhens were seen to mainly parasitize the nests of their close neighbors. Maned ducks (Briggs, 1991) and swallows (Moller, 1987) were also seen to behave in similar manners. Also in their reviews Eadie et al. (1988) and Beauchamp (1997) both mention that conspecific brood parasitism happens at higher frequencies in species that nests in cavities and thus could more easily found. Along these same lines, the density of a bird population has also been associated with this behavior; the denser an avian population the more probable it is that brood parasitism occurs. According to Sorenson (1993), the intraspecific brood parasitism that he observed in a canvasback population could be attributed to high density of the population. In most of the studies and reviews that I read, coloniality of a species was also a very good determinant of IBP (Lank et al., 1989; Beauchamp, 1997). In general, the easier it is to detect the bird’s nest in a population of a certain species, the higher the chance of parasitism occurring.
Another determinant that has been supported by most of the research is the condition of the young right after hatching, in other words whether a species is precocial or altricial. Ahlund (2005) mentions in his study on brood parasitism in the common goldeneye duck, that self-feeding, precocial, young might make intraspecific brood parasitism more feasible mostly by reducing the cost to the host of rearing some other female's eggs. Less costly for the host would be mean that the hosts would need to adopt less defense mechanisms against the parasites, assuming that the presence of the parasites eggs does not decrease the host eggs' hatching success, which is the case in cliff swallows (Brown and Brown, 1998). In their review, Rohwer and Freeman (1989) also determined that intraspecific brood parasitism is much more common amongst birds with self-feeding young.
Lastly, the studies of Sorenson (1993) and Weigmann and Lamprecht (1991) both identified a relation between the intraspecific brood parasitism occurrence and the amount of young females (usually 2 years old females) in the general population of breeding females. The more the young females, the more prevalent parasitism is. A possible explanation as to why this is the case is the fact that younger female's are usually more inexperienced and less likely to initiate their own nest.
Hypotheses
Epiphenomenon Hypothesis
The epiphenomenon or otherwise known as the side-effect hypothesis was first proposed by Semel and Sherman (2001) in their paper about intraspecific parasitism in wood ducks. What this hypothesis is saying is that intraspecific brood parasitism has not specific adaptive significance but it is just the result of another unrelated behavior. They are basing this hypothesis on the wood duck's high degree of fidelity to a nesting place that have used in previous years. In other words, females tend to prefer the nests that they have used in the past and if these are occupied when they return from migration, they will not be deterred from laying their own eggs too. As a result two females are laying their eggs in the same nest until eventually they encounter each other and fight for the nest. One of them will leave, leaving its eggs in the nest which is how 'parasitism' occurs according to Semel and Sherman (2001). Also, it is important to note that in this case host and parasite did not differ in terms of their nesting and egg-laying behaviors.
Although, it might be the case in this particular wood duck population, there is very little evidence of the side-effect hypothesis in other cases. For instance, the behavioral observations of Weigmann and Lamprecht (1991) showed that there are distinct differences between the two groups behavior. Parasites in their study rarely visit nests on non-laying days, they do not visit the same nest repeatedly, they do not nest where she already laid which indicate that they purposely trying to parasitize another females nest. Ahlund (2005) also notes that the behavior is tactical and not accidental since there are behavioral differences between the two groups. This hypothesis could account for some of the parasitism that occurs but certainly not for the behavior as a whole. Also, the very large number of different species that exhibit this behavior could not be explained just by this epiphenomenon hypothesis (Lyon and Eadie, 2008)
Kinship Hypothesis
This hypothesis suggests that it might be close relatedness between the host and the parasite that facilitates the evolution and persistence of intraspecific brood parasitism (Semel and Sherman, 2001). If the host is incubating and caring after the young of one of its close relatives, then the cost of rearing those young is not as large, mostly because in doing so the host is increasing its own inclusive fitness (Lyon and Eadie, 2008). This hypothesis could also be related with determinant of nest proximity and population density. The empirical observations related to this hypothesis have had mixed results. Semel and Sherman (2001) in their observations about wood ducks, detected that the ducks would actively avoid to parasitically lay in their kin's nest, which hypothesized was because of the possible negative effects of parasitism on the host's fitness. On the other end of the scope, Andersson and Ahlund (2000) determined using egg albumen sampling that host and parasite are often related and the higher the degree of relatedness the greater the number of eggs the parasite lays. The significance of this hypothesis still remains to be explored.
Best-of-bad-job Hypothesis
The best-of-bad-job hypothesis can be divided into two sub-hypotheses, the restraint hypothesis and the constraint hypothesis. On the one hand, the restraint hypothesis refers to the cases when the females, in an attempt to avoid the high costs of incubation and caring of the young in a very unfavorable environment, avoid initiating their own nest and restrain their reproductive efforts by passing on these costs to the parasite (Briggs, 1991). On the other hand, the constraint hypothesis suggests that the females resort to parasitism because environmental or phenotypical limitations, for instance because of nest loss because of predation or natural disaster (Sorenson, 1993), or because of inability to obtain the required resources due to competition (Forslund and Larsson, 1995; Lyon, 1993; Weigmann and Lamprecht, 1991), or because of the female's poor physical condition after a long migration (Lank et al, 1989). All these are reasons that would render initiating a nest extremely costly (Lyon and Eadie, 2008; Sorenson, 1991).
The consensus concerning the reproductive success of pure parasites (non-nesting females) is that it is far less successful that both that of pure nesters and that of nesting parasites (females that adopt both strategies) (Lank et al., 1989; Lyon and Eadie, 2008; Sorenson, 1991), mostly because of factors such as egg-transfer inefficiency, insufficient egg-production and reduced hatching success, usually because of bad timing of the parasitic attempt (Lank et al., 1989; Sorenson, 1993). With this in mind, researches hypothesize that parasitizing is alternative that could provide some, even if minimal, reproductive success to females that are otherwise 'doomed' of the conditions that I mentioned above. Weigmann and Lamprecht (1991) noticed actually that parasitizing was actually more successful than avoiding breeding in general. If this hypothesis holds true, then it is predicted that the females might have decreased annual fecundity because of parasitism's lower reproductive success but at the same time the manage to increase their life-time reproductive success by ensuring their survival until conditions improve, especially if we are referring to the restraint hypothesis (Lank et al., 1989; Lyon and Eadie, 2008).
Mixed-Strategy Hypothesis
The mixed-strategy hypothesis explains the cases that females not only parasitize but also initiate their own nest and clutch. Researchers explain this attempt by the females to overcome some of the constraints of brood and clutch size and as a result manage to increase their annual fecundity to numbers greater than both the pure nesters and pure parasites (Ahlund and Andersson, 2001). Moorhens for example are limited by the amount of chicks the can rear per nest and in order to avoid this limitation the adopt a mixed strategy and parasitize females with previous breeding experience and manage to lay an average of 30% more eggs than non-parasites (McRae, 1998). The American coots in Lyon's (1993) study exhibit similar behaviors. It is also suggested that contrary to the best-of-a-bad job hypothesis, the individuals that adopt both reproductive strategies tend to be of higher quality with a larger amount of resources, which is reflected by their survivorship (Brown and Brown, 1998).
Flexible Life-History Hypothesis
This hypothesis, which has not been given nearly as much attention, is based on the fact that, as Sorenson (2001) puts it, "When the conditions an animal faces are variable, selection should favor 'yes-if' genes, which allow animals to 'decide' on appropriate level of reproductive investment for any given set of conditions". According to him, the females at the beginning of each season will have to choose between nesting, parasitizing, combining both strategies or completely abstaining from breeding and they will base their, certainly unconscious, 'decision' on the environmental conditions with which they are faced. For instance if conditions are well and favorable then females will choose to invest the most in their reproductive effort and probably adopt a mixed-strategy (Lyon and Eadie, 2008; Sorenson, 1991). This hypothesis also takes into consideration the fact that there have been no professional life-long parasites but instead most females will alternate their reproductive tactic depending on the year (Lyon and Eadie, 2008). The key idea of this hypothesis is flexibility and the ability of the birds' to respond to environmental and social cues and adapt their effort according to those. One limitation of this hypothesis is, however, the fact that not all species have been associated with all for reproductive tactics. In other words, there are species whose parasites are almost always non-nesters (Sandell and Diemer, 1999) and others that are only nesters (Brown and Brown, 1998).