1 General introduction

↓1

Few organisms are as well studied as the honey bee (Apis   mellifera L.). It is the sole subject of research in several university departments and research centres and has several journals devoted specifically to it. This is partly attributable to its economic importance in agricultural systems, but also because of its unique biology and relative ease of access and manipulation for study.

The evolution of co-operation in eusocial insects has for decades been a central topic in evolutionary biology. The key trait of eusocial (“truly social“) species is that most of the colony members (workers) give up their own chances of reproduction and help raise the offspring of nestmates, which are highly fecund (sexuals or reproductives).

Eusociality according to Wilson (1971) is defined by the following three characteristics: 1) the partition of reproduction among the colony members, with sterile or subfertile workers and highly fecund sexuals or reproductives, 2) overlapping adult generations and 3) co-operative brood care. All ants and termites, some bees and wasps, ambrosia beetles (Kent and Simpson, 1992), aphids (Aoki, 1987; Benton and Foster, 1992), thrips (Crespi, 1992), shrimps (Duffy, 1996) and some naked mole-rats (Sherman et al., 1991; Jarvis et al., 1994) have been found to fit this definition.

↓2

In eusocial insects, for instance, we have to consider that differences in relatedness within colonies of social Hymenoptera are likely to create a variety of conflicts and cooperation between colony members. Furthermore, kin selection has been widely accepted by the scientific community. Insect societies have long served as a useful model with regard to these issues in evolutionary biology, i.e. the level at which natural selection operates.

Kin recognition can be defined as the ability of an individual to distinguish kin from non-kin or to differentiate between different classes of kin (Hepper, 1986). Such recognition appears to be common in the animal kingdom (Hepper, 1991), by using a widely involved mechanism to recognize kin not previously encountered, and has been demonstrated in monkeys (Wu et al., 1980), a social spider (Evans, 1999), a sweat bee, (Greenberg, 1979), ants (Jutsum et al., 1979; Mintzer, 1982; Lenoir, 1984), paper wasps (Allen et al., 1982; Pfennig et al., 1983; Hepper, 1986; Gamboa, 2004) and the honey bee (Breed, 1981 and 1983; Page and Erickson, 1984; Getz and Smith, 1986; Page et al., 1989; Carlin and Frumhoff, 1990). Additionally, among mammals, most work on kin recognition has involved rodents: mice, rats, squirrels and voles (Gadagkar, 1985).

Kin selection can also explain the evolution of sociality and cooperation among individuals. Inclusive fitness theory, for the first time, provides a framework for explaining the way that the level of relatedness between individuals can compensate the costs associated with altruism and can influence the occurrence of social behavio u r (Hamilton, 1964).

↓3

Hamilton (1963; 1964a, b) was the first to develop kin selection and viewed it as a far-reaching important evolutionary principle. His “kin selection theory” states that individuals can transmit copies of their genes not only directly through their own reproduction, but also indirectly, by favouring the reproduction of kin. Altruistic behaviour should be favoured if the ratio of the costs (c) accruing to the donor of the altruistic act to the benefits (b) gained by the beneficiary is lower than the relatedness (r) of the recipient to the donor of the altruistic behaviour (Hamilton’s rule, c/b < r). Here, benefit means the enhanced production or survival of offspring by the beneficiary, whereas cost means the number of offspring lost by the altruist.

Kin selection theory also implies the occurrence of potential kin conflicts, because, in contrast to the cells of an organism, nestmates are not genetically identical (Ratnieks and Reeve, 1992; Keller and Reeve, 1999). Hence, kin selection predicts a dynamic equilibrium between co-operation and conflict, depending on, for example, the genetic composition and size of a colony, the benefits and costs of group membership, and the benefits and costs of selfish behaviour and policing (Keller and Chapuisat, 1999).

Social insect species, for example, are particularly interesting models to study altruism, because the haplodiploidy that characterizes Hymenoptera, the order of ants, bees and wasps, can lead to high levels of relatedness between individuals. However, insect colonies show a great variability in their social organization and this results in highly variable levels of relatedness among colony members (Zinck et al., 2009). However, recently, Nowak et al., (2010) have speculated that higher relatedness is not necessarily the best option for inclusive fitness and that, therefore, relatedness is irrelevant for eusociality.

↓4

Polyandry (females mating with several males) occurs commonly and is a widespread phenomenon in social Hymenoptera (Page, 1986; Ross, 1986; Have et al., 1988; Arnqvist and Nilsson, 2000; Jennions and Petrie, 2000; Crozier and Fjerdingstad, 2001; Maklakov and Lubin, 2006); this has important consequences for reproductive conflict and cooperation among colony members. For a honey bee, the colony consists of a single queen, tens of thousands of sterile female workers and usually a few hundred drones (Winston, 1987). The honey bee (A.  mellifera) is a particularly important model organism in studies of social cooperation and conflict, because of the genetic variation within a colony (Oldroyd and Fewell, 2007). In general, colony honey bees normally comprise 10-20 patrilines (Estoup et al., 1994; Robinson et al., 1994; Arnold et al., 1996; Oldroyd et al., 1997). Workers within the same patriline share an average coefficient of relatedness of 0.75 (full-sister) and among patrilines 0.25 (half-sister) (Ratnieks and Reeve, 1991; Châline et al., 2003; Châline and Arnold, 2005). Furthermore, A. mellifera  workers are able to discriminate the degree of relatedness to themselves of larvae (Visscher, 1986; Noonan, 1986). Because of relatedness differences, workers can potentially gain a threefold increase in inclusive fitness if they can induce a full-sister instead of a half-sister to head a new colony (Visscher, 1998). Occasionally, this involves preferentially rearing queens from related larvae (Breed, 1983; Page and Erickson, 1984; Page et al., 1989; Carlin and Frumhoff, 1990). For this, worker bees possess the sensory capabilities and behavioural responses that would enable them to maximize their individual inclusive fitness through nepotism in queen rearing. This is an important aspect in kin selection theory. However, several studies with regard to A. mellifera have been unable to establish nepotism in queen rearing. Less related larvae exhibit a similar or even higher acceptance for queen rearing than do larvae with a higher degree of relatedness (Woyciechowski, 1990; Breed et al., 1994).

Hence, queens' production in these societies involves numerous group decisions with respect to foraging, nest maintenance and reproduction. This is because the inclusive fitness of the whole colony depends strongly on decisions made during this process (Tarpy and Gilley, 2004). Therefore, colonies show a variety of complex behaviours that far transcend those of the individual colony member and which cannot be accounted for by any apparent central control or simple hierarchical structure (Hölldobler and Wilson, 1990).

The purpose of this work has been to focus on whether the acceptance of larvae for queen rearing is also affected by any factor over and above genetic relatedness. Only a few studies have shown a significant preference for related larvae (Page and Erickson, 1984; Noonan, 1986; Visscher, 1986; Tarpy and Fletcher, 1998; Koeniger et al., 1996; Mohammedi and Le Conte, 2000). However, all available studies have revealed a significant preference towards related larvae. Nevertheless, most of these studies involved larvae which had hatched in non-related colonies of origin, so that recognition might not have been based on kin recognition, but rather on colony odour, which is partly environmental and can be differentiated by worker bees. Consequently, a special experimental design has been created to exclude any influence of colony odour of larvae to be reared as queens. Consequently, we used age-standardized larvae that had hatched and were reared under standardized conditions outside of their hives.

↓5

In this thesis, I first examine and create a methodological basis for the study of differential prenatal maternal investment shown in honey bees, by determining the repeatability of egg weight measurements and by identifying the optimum age of eggs for measuring differences in maternal investment. Then, I speculate that the initial chance of a larva related to the individual weight of the egg from which it originated will affect its later performance. Consequently, rearing queens from larger eggs might help to optimize colony fitness. This is perhaps more important than selection attributable to relatedness to a special subgroup.

Finally, queen rearing results from a colony decision, which involves many workers; consequently the lack of significantly proving nepotism may be due to the fact that relatedness-driven kin preference of individuals cannot be transferred into a colony decision. Therefore, I tested also by DNA fingerprinting the relatedness of larvae to be reared as queens and the nursing worker bee, which initiates the queen rearing process, in order to determine the influence of reaction towards kin.


© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.
DiML DTD Version 4.0Zertifizierter Dokumentenserver
der Humboldt-Universität zu Berlin
HTML generated:
05.12.2014