Durham Anthropology Journal Volume 13(1) ISSN 1742-2930 Copyright © 2005, R. Aspden Download a PDF version of this article

Investing in infants: Applying the biological market to the interest in other's offspring.

Robert. E. Aspden

Department of Anthropology, University of Durham

Abstract: The biological market principle (Noë et al ., 1991, Ethology , 87: 97-118) has provided an alternative theory to models of cooperation based on reciprocity. The proposed research study is an examination of two model theories of cooperative behaviour, Hamilton's kin selection theory (1964, Journal of Theoretical Biology , 7:1-52) and Noë's biological market . This paper assesses current thinking on the biological market introducing basic rules, terminology and predictions; and reviews previous literature in primates. The literature review also describes examples of the biological market that have been published since its formulation, over a decade ago. Some of these examples do not make reference to the market but are used here, as examples of the kind of situation were a market effect could be demonstrated. The production of this article comes in light of the work of my current Masters degree, studying the abundance and patterns of allogrooming in the same group of capuchin monkeys. Some of this data will be discussed to support the hypotheses and predictions made for future research. The application of the "market" in evolutionary biology, like many other economic-based ideas before, is changing the way cooperation is seen, showing equality through trade in a selfish world rather than suggesting a selfless world built on traits like altruism. The suitability of the market as an alternative to other models of cooperation is based on the fact that competition is a driving force in favour of cooperation rather than a force against it.

This paper is presented in conclusion to the author's presentation at the Department of Anthropology Postgraduate Student Conference, 2004 held in the Department of Anthropology at the University of Durham. This presentation was made to introduce both the idea of the biological market and its position within the proposed PhD research. The audience was combination of both biological and socio-cultural anthropology academics and students alike, therefore, some emphasis was made to terminology and background as well as theory and method. Some discussion is had at the end of the paper to points and questions raised in the conference itself.

1. Introduction

1.1. In most cases, models aimed at explaining the systematics of cooperation have been focused on two-player situations, in which individuals can either cooperate or defect. The model of the biological market principle (Noë et al ., 1991) reflects the natural system of group-living animals more accurately, accounting for the presence of multiple potential cooperators offering and requiring differing resources. The idea operates on a set of basic rules and reflects many aspects of human trade systems. Since its inception, studies on the biological market, from insects to primates, have produced an array of evidence to support the model and more examples can be seen in the literature that could indicate a market effect in operation. The aim of this paper is to introduce the theory of the market, the rules it follows and the terminology that it has generated and employed. This paper will focus on the theory of the biological market principle and then incorporate this idea into a proposed research study investigating the suitability of two models of cooperation for the occurrence of handling other's offspring, in capuchin monkeys ( Cebus apella ). The market principle is one model, the other is kin selection theory (Hamilton, 1964, Maynard-Smith, 1964), which is beyond the scope of this paper.

1.2. Working examples of the market effect for infant handling and allogrooming will be discussed in detail to give light on the development of this research study. Some of these examples were used in the original formulation of the market model (for example Nishida, 1983). Other examples come from literature that was produced after the market model had been created and represent examples of intraspecfic competition, one of the hypothetical situations used in the original articles of Noë et al . (1991) and Noë and Hammerstein (1994). Some examples presented have never been associated with the market idea but could demonstrate further instances of the market effect for a number of different situations. The last part of this paper will consider how this literature on the biological market model, specifically on trade of infants, could be advanced in future research. This will include development of models that predict the effect of small features within the larger market effect, an idea that Noë et al . (1991) felt to be a strength of the biological market principle.

1.3. Infant handling or "interacting with infants other than one's own" occurs in a wide variety of species. I will refrain from discussing motivation in this paper and therefore refrain from using terms like " alloparenting " or " aunting " that may suggest some form of motivation and/or relationship interest (for a discussion on motivation of caretaking see Ross and MacLarnon, 2000). I will refer to the behaviour as infant handling or access to infants to allow for the inclusion of a variety of associated behaviours and to follow currently used terminology (Manson, 1999; Henzi and Barrett, 2002). In capuchin monkeys the behaviour occurs in a variety of different forms and from a variety of different group members. Females tend to show greater interest than others but in capuchins juveniles and males do show interest to lesser degrees. Infant handling normally takes the form of carrying or holding the infant but other behaviours like nuzzling, lip-smacking, allogrooming, suckling and general pulling etc may all occur while the infant is still in contact with the mother and therefore "access to other's infants" may be more appropriate. This is especially apt for primate species that often show a lot of interest in other's infants but provide little care for them (Silk, 1999). With the associated costs/risks of others handling a mother's infant a mother should restrict access to limit harm (Maestripieri, 2001) and therefore only allow access to those who provide valuable services in return (or perhaps are genetically related).

2. The biological market

2.1. Coming straight out of economic theory, the biological market principle is built on the idea that animals may trade resources or behaviours with each other in a competitive market system were competition acts as a driving force within `classes' and cooperation occurs between different `classes'. The idea of a class system (Noë et al ., 1991) is used to describe how different group members offer different resources to others and in return desire alternative resources to be traded that they do not currently posses. It has been widely accepted for some time that in interactions such as mating systems there is an apparent market effect between differing individuals, but a formal theory with universal application was drafted much later. The term interchange was first coined by Hemelrijk (1990) to describe trade of differing resources or behaviours, while using exchange to refer to the same resource being reciprocally traded. This sets apart reciprocity and interchange and the market model itself was first evinced using interactions involving different resources. While designed around interchange, the model will cope with the effects of mutualism and exchange as features of the market. For groups of social animals, the unit of the group can be separated into the classes described because there is normally a number of individuals offering the same resource and desiring another. In cases were behaviour may be limited demographically, e.g. to females or adults only, there is likely to be a number of individuals within the group capable of performing the behaviour. This is the basis of the market system; one class wanting what the other has and interchanging a resource in a trade.

2.2. Because the individuals within one class are trading the same resource as one another in order to obtain the same alternative resource this creates market competition between these class members who compete for access to these resources. Assuming that these class members cannot exclude each other using force (violence, etc) then outbidding class rivals is the only option. Outbidding, or the need for outbidding, is affected by a simple interaction known as ` supply and demand '. Noë et al . (1991) used outbidding in an example of when there are multiple helpers at the nest. If nests are tended by multiple unrelated helpers, the work rate for each individual bird often goes up instead of down as would be expected. It would be assumed that the more birds helping would mean a shared workload. However, if these helpers gain benefits from the breeder/territory owner these helpers are in direct competition to receive these rewards. The breeder is then able to choose between helpers for the most reliable (hardworking) individual, hence, it is in the interest of the helpers to increase their work rate and out compete their rivals. This is especially the case if a territory can only support a limited number of helpers as well as the territory family. A rejected helper would also be evicted from the territory so more effort could be the difference between life and death for the helpers.

2.3. Noë et al . (1991) formalised this working hypothesis by giving the classes names in a more detailed description of the interaction. Individuals labelled Type-A ("Auctioneers") have a resource wanted by the type-B individuals ("Bidders"). In a simple model type-Bs outnumber type-As (two type-Bs and one type-A) and the resource held by type-A is limited. All type-Bs cannot, therefore, access the resource. If the resource held by type-A has a fixed value it is likely the the alternative resource can be traded to a specific threshold. When both individuals are trying to trade for the resource the type-A can playoff the two type-Bs against one another. In this instance both type-Bs must increase their work rate to out compete the other. The competition between the two type-Bs is effective because the type-A is able to exert some degree of partner choice over the counterparts. This has been exemplified in game theoretical models that consider more than two-player systems. It is the move from two-player systems to multiple-player systems that breaks open a competitive market model (see Noë and Hammerstein, 1995, Box 2, page 337). Competition and partner choice also have some degree of an effect over the decision to enter a cooperative act.

2.4. The previous models used commonly in biological studies of cooperation consider only the two-players involved in the interaction. The Prisoner's Dilemma (Axelrod and Hamilton, 1981) and the theory of reciprocal altruism (Trivers, 1971) consider only two players already locked in a cooperative situation. In both these models each player has two options: to cooperate or to defect. The idea of partner choice is not considered; these models rely on partner-fidelity (Bull and Rice, 1991). In the biological market model, because individuals are in direct competition with each other for access to other individuals holding resources the most basic model always contains three players. Now if we consider the type-A player involved in a game theoretical situation with two type-Bs there are three options: to cooperate with the first type-B, to defect from the first type-B and cooperate with the second type-B, or defect completely. Each option has specific costs and benefits similar to those imposed on the players of the previous two models, however, the option to interact with an alternative player influences the overall cost of defecting for the type-A. Furthermore with the threat of type-A defecting to interact with the other type-B the first type-B is forced to make a larger offer before being allowed to cooperate with the type-A. Assuming offers cannot be withdrawn once they are made these offers are honest signals of desire or commitment to interact. This effectively limits, and in some cases, eliminates cheating (Noë and Hammerstein, 1995). Because partner choice is not an option in the models of the prisoner's dilemma or reciprocal altruism, without regulatory systems (monitoring etc) that protect against cheating, there is always an option for players to cheat meaning competition works against cooperation.

2.5. The case above explains how rival demand can mean greater competition and an increased work effort; the effect of supply has similar consequences. If the number of resource holders goes down in number, i.e. the type-As become even fewer, or the amount of available resource goes down the competitive type-Bs would suffer from a greater level of competition for access to these increasingly rare resources. There is a difference between numbers of resource holders and the amount of resources in certain cases, for example monopolisable fruit sources, as an individual may be able to hold onto multiple `units' of a resource so there may be more resource than resource holders or if one resource holder is able to completely monopolise a resource there may be more potential resource holders than available resources. This relative power for the type-As means an increased ability to exert partner choice. If the resource availability decreases this should enable the type-A to request more of the rendered "commodity" in trade for the resource on offer and from any individual it wishes to interact with. The idea of monopolisation is another important feature of the market model.

2.6. Resource holding power (RHP) is analogous to dominance in that it is the ability of an animal to keep hold of a resource (Noë et al ., 1991). Dominance is the normal way RHP is exerted in group-living animals but other factors like coalition formation in primates, for example, also affects the RHP of an individual. Possession should also limit access to resources, as it should not be possible for non-possessors to take a resource, by force, from a current possessor. Possession could come in many forms including respect of ownership, if animals recognise prior ownership of resources (Kummer, 1991). Possession may be due to the resource type, which may not be tangible or accessible by force. For example if the commodity being traded is a form of behaviour then it cannot be forced out of an individual (with the exception of submissive behaviours). Even if another animal is dominant and can forcibly take resources, the benefits of taking small resources from subordinates may not significantly outweigh the risks involved in obtaining them if this involves fighting, etc (for example see Mathy and Isbell, 2001). Furthermore, if the risks involved in then holding onto the resource oneself are also high it may not benefit an animal to first fight to obtain these resources, particularly if long-term possession is the only way to benefit from the resource (Kummer, 1991). In light of these various points a basic model of the "market effect" can be generalised in the form of a few basic rules.

2.7. When a collaboration is formed there is potential for a market effect only when (quoting from Noë and Hammerstein, 1994, page 6):

"It is plausible that pairs are formed according to the preference of at least one party."

"Partners cannot, as a rule, be forced to hand over their commodity without their consent."

"Competitors [(same class members)] cannot be excluded by force from contacts with the opposite class, leaving outbidding each other as the most likely form of competition."

"Cheating, that is changing the value of the commodity after forming the pair is not possible."

2.8. While the model was originally designed to explain the asymmetries of relationships through interchange and power imbalance in group-living animals the model predicts reciprocal exchange for the same behaviour as well as interchange of different resources. Hemelrijk and Luteijn (1998) successfully highlighted the idea that competition and reciprocity are inversely correlated. They stipulated that in female-bonded groups males can be seen as a resource worth competing over and because females normally out-number males females have to compete to access males. They further demonstrated that reciprocation of allogrooming reduced as the group sex ratio decreased, i.e. when the numbers of females and males became more even, females compete less for access to males. Allogrooming, therefore, becomes more reciprocal because it is not being interchanged for tolerance. Similarly, a link between competition and reciprocity was demonstrated by a negative correlation between allogrooming reciprocity and aggressive displacement behaviour. If females suffer from lower competition, services such as allogrooming need not be rendered against resource access, a finding also supported by Barrett et al . (1999, 2002) and Henzi et al . (2003). In Henzi et al . (2003) exchanging allogrooming in periods of low competition was found to be an effective strategy for maintaining social relationships with powerful partners, with whom interchange (for tolerance, etc) would secure relationships in periods of increased competition.

2.9. Exchange (in reciprocity) can still be accounted for under the rules of the market, using supply and demand, as described above. When supply of the resource is sufficient for all group members wishing to obtain it this removes competition within and between classes so resources can be exchanged like for like. The benefit of maintaining social relationships, through reciprocity, is gained later when both partners require resources that the other is offering, as shown by Henzi et al . (2003). The Hemelrijk and Luteijn study above is not a perfect example of the market model because the type-B female is servicing the type-A for tolerance in periods of higher competition while both type-A female and type-B female are directly competing for access to the same resource (mating males), however, I use the principles they describe to support the explanation of exchange within the market (competition driving interchange versus reciprocity, Hemelrijk and Luteijn, 1998). Since the original papers on market models, a number of examples have been demonstrated to support and test the theory and have developed a number of smaller models to adapt the basic rules of the market to biological systems. This literature is used to support the market model and its applicability to the current research topic.

3. The case of infant access

3.1. The interchange trading of allogrooming for access to another's infant was just one of a number of examples used to model the market effect in Noë and Hammerstein (1994). Since then, work by Henzi and Barrett working on Chacma baboons ( Papio cynocephalus ursinus ) has provided a working example of the principle (Henzi and Barrett, 2002). Grooming was interchanged for access to a mother's infant and three significant features of the data supported the market model. Firstly, mothers were unlikely to engage in reciprocal sessions of grooming with individuals who handled the mother's infant (demonstrating interchange). Secondly, handling of infants most often occurred immediately after cessation of a grooming bout directed from "potential handler" to mother (instantaneous trade). Thirdly, and described as most critically important by Henzi and Barrett (2002), the fluctuation of supply and demand created a market effect that when more infants were available grooming rates directed towards these mothers were significantly lower than when few infants had been available. Di Bitetti (1997) also showed that, in capuchin monkeys ( Cebus apella ), infant presence correlated with a significant increase in received allogrooming but did not invoke the market model as an explanatory theory.

3.2. I would highlight two other features of interest from Henzi and Barrett (2002) that support the market idea. Firstly, in mother-mother dyads interchange trading was replaced with reciprocal exchange of grooming-for-grooming, and secondly, when females became mothers (gave birth) rates of received allogrooming went up compared to when these mothers had no infant. The reciprocal nature of mother-mother dyads is predictable based on the fact that the other resource being interchanged for allogrooming is infants, which both mothers have themselves. Possessing her own offspring would remove the need for a mother to access another mother's infant (depending on motivation). Two mothers are not interchange trading and, therefore, reciprocally exchange allogrooming. The fact that received allogrooming goes up for individual females when they have an infant correlates allogrooming behaviour directly with the presence of infants. The suggestion has been made that mothers even receive higher amounts of allogrooming when they are actually holding their infant opposed to when they are not (Matsumura, 1997). This immediate trade may make trading easier, removing the need for bookkeeping (as seen in reciprocal altruism) and reduce the chances of free-riding or defection. Securing within-bout reciprocity (or in this case interchange) is a feature achieved by animals varying clique size (Henzi et al. 1997), an adjustment that correlates with competition, and as they suggest, supports the idea of a tradable commodity within a market effect (see also Barrett et al ., 1999).

3.3. Manson (1999) studied capuchin monkeys ( Cebus capucinus ) to see how infants affected female-female bonding. Individuals who handled infants were likely to engage in more social activities with the mother, both allogrooming her and forming coalitions. Manson (1999) did not invoke the market model and alternatively suggested that infant handling was used to test bonds between "close associates". This idea implies that infant handling would indicate, to each member of the `tested' relationship, the level of willingness of the other to invest in the relationship. The use of infant handling in bond-testing would preclude an interchange of the behaviour for allogrooming, although the data Manson presented would possibly support this connection. Retesting the data in Manson (1999) and Di Bitetti (1997, above) invoking the supply and demand effect could provide further evidence to support the market model.

4. Further interchange of allogrooming

4.1. The market effect has been demonstrated for trading of allogrooming with a number of other resources besides access to infants. de Waal (1997) showed a market effect in the interchange of allogrooming for tolerance at feeding sites. Interestingly he did not precisely invoke the market model as described by Noë et al . (1991) and Noë and Hammerstein (1994) but referred to his own " service economy " were a number of different resources were traded "reciprocally" and "exchanged" for each other (actually meaning interchanged as described above). The results he describes in his paper (de Waal, 1997) accurately reflect a market model and market effect. Each analysis performed reflect the services he examines being interchanged for each other with one trade being made dependent upon the other. Other effects like third-party influence were ruled out so that the only remaining conclusion was a direct trade of allogrooming for food. de Waal (1997, page 379) suggests that grooming alters an individual's " food-getting success (FGS)"; those that groom are more successful at feeding sites. He describes a " grooming effect " defined as "the difference in FGS between food trials preceded by grooming of the possessor by the nonpossessor and trials not preceded by such grooming" (381).

4.2. While the recipients of food sharing slightly increased the amount of allogrooming given to benefactors, these benefactors (sharing food) were less likely to groom their beneficiary than other individuals who had not received food. It can be interpreted that this result indicates an interchange of allogrooming with tolerance in some relationships and in others, a more reciprocal nature. This result was matched by the fact that while both reciprocal allogrooming relationships and reciprocal food sharing were positively correlated this result was non-significant. The final piece of evidence from this paper is the effect of dominance. When subordinates had monopolised food clumps even dominant individuals waited to be offered food or to be allowed to take some from the clump, often making the outstretched-arm food-gesture. This suggests that there is a degree of not only partner-choice shown in rejection/resistance behaviour but also that any individual who holds the resource is capable of exerting some degree of RHP.

4.3. The situation described in Palombit et al . (2001) for female chacma baboons and their male `friends' has potential for a perfect example of the biological market, with clear market effect and variation in supply and demand. However, the results described show that not all situations are suited for the application of the market model. New alpha-male chacma baboons are often infanticidal when entering into a group. To guard against this effect females solicit help from male `friends' with whom allogrooming relationships have been established. Females can form new friendships with males who have already got a relationship with another dependent female. The fact that these females are in direct competition for the protection of these friendly adult males and that they service these relationships with allogrooming has the potential for a market effect were these females would have to outbid each other for the attention of the adult male. However, as stipulated by the rules of the biological market (see above, Noë and Hammerstein, 1994), the effect of outbidding is only an effective strategy if within-class rivals cannot exclude each other from accessing members of the other class type (rule three, above).

4.4. Palombit et al . (2001) describe how dominant females are able to exclude subordinates from accessing male friends and so when a dominant female forms a friendship with a male who is currently a friend of a more subordinate female, the dominant female is able to take over access to the male. These dominant females do not increase attacks on subordinates however they are able to displace and supplant the lower-ranked rival. An increase in supplanting of rivals is thought to indicate a reduction in tolerance of subordinate competitors, however, Palombit et al . stipulate that more data is necessary to test this hypothesis against the other alternatives of avoidance by the subordinate or a natural increase in rate of interactions (supplants) as a result of increased proximity between the females. One point of interest is that dominant females increase their effort to maintain proximity to male friends once subordinate females begin to associate with these males, which was interpreted to mean that dominant females must invest more to maintain their current affiliations. More research of this correlation could reveal a market effect while the exclusion of subordinates by dominants breaks the rules of the market model. Dominance is a confounding variable that has been painful for those investigating the market (Henzi et al ., 2003).

4.5. Noë et al . (1991) and Noë and Hammerstein (1994, 1995) emphasise that the biological market is an alternative to reciprocity and two-player game theoretical situation. They acknowledge that none of these models will universally relate to all situations but infer that these different hypothetical models should be applied in different situations. The model of the biological market would appear to apply to the last example above but shows the importance of the rules for modelling in behavioural ecology. Without more data or reanalysing the data of Palombit et al . (2001) the biological market could not be applied to a situation that appears in the first instance to be the definitive example of the model itself. The suggestion that dominant females raise effort to maintain affiliations when other subordinate females are present indicates a potential market effect however the reverse situation breaks down with no apparent RHP for subordinates who appear incapable of maintaining possession of male friends. More data is required to provide working examples of this kind of effect, were in one direction RHP and a market effect can be seen but subordinates suffer the social costs of their low ranking.

4.6. The examples to follow demonstrate how subordinates can turn their position to an advantage. In Nishida (1983) three male chimpanzees, in the Mahale mountains of Tanzania, were observed to compete amongst each other for either alpha-status or solicitation of coalitionary support. Two males (the alpha and gamma) were seen to regularly attack the third (beta) male. Grooming bouts were repeatedly seen between the alpha-male and gamma-male and also between the gamma-male and beta-male. The gamma-male changed coalitions a number of times, and this was apparently following allogrooming bouts with the other two rivals. Nishida reported that a dominant male (either the alpha or beta-male) was only capable of securing total dominance with coalitonary support from the third (gamma-) male; a market effect for allogrooming of the gamma-male interchanged for his support in coalitions. Nishida (1983) reports that the gamma-male was able to maximise his own copulatory success when the status of his two superiors was most unstable. The market effect appears to be exemplified in this situation were one type-A with strong RHP was able to playoff the two type-Bs to maximise his own self-interest by benefiting the more committed partner.

4.7. A similar situation may have been presented in Manson et al . (1999). In most cases allogrooming was recorded to go up the hierarchy, however in one period of observations for the group "SR-LV-1996" allogrooming was directed down the hierarchy. Manson et al . interpreted this to be a trade of grooming from dominant females for coalitionary support from subordinates in a period of great unrest, when two females were in constant rivalry for the alpha-position. While the market model was not applied to this situation, the fact that at other times allogrooming, in the same group, went up the hierarchy suggests a market effect may have produced the inverse allogrooming distribution; females having to increase their effort to secure the alliance of their coalitions when they are needed. If, as in Nishida (1983), these two females of in Manson et al . (1999) were competing for support from the same subordinate(s) the increase of work effort suggests an attempt to outbid each other. An interesting point to note is that these females are competing for coalitions because they are not able to dominate each other through violence, therefore fitting with the rules of the biological market they are not able to exclude each other through violence from accessing the other class, hence outbidding becomes their only option. As stated above, RHP is not always akin to dominance because coalitonary support can raise the dominant status of an individual that is physically weaker than those it is dominant to. In this situation dominant status is the resource being held and coalitionary support is being rendered in the interchange of allogrooming.

5. Developing the market model for infant access

5.1. The methodology of the study that follows is currently being researched and therefore is beyond the scope of this paper. This part of the paper will highlight the points of the biological market that are of interest to study in the capuchin monkey group being used already for research on cooperation. The work on access to infants by authors such as Henzi and Barrett (2002) and other studies of biological markets ignore or separate the effect of dominance and kinship from the market model. It is my interpretation that a biological market should be able to include biological variables such as dominance and kinship. Henzi et al . (2003, page 933) stated "that high-rank may enable a female to override market value (Henzi and Barrett, 2002)". I would argue that this "override" is not, in the strictest sense, a bypassing of the market effect but an additional value that must be considered by traders in a market. It may be that for the example of kinship that being related to the trader means less of a value may be rendered for services. It has long been known that relatives exchange between one another (reviewed in Chapais, 2001, Silk, 2002) and that they may gain preferential access to resources. Kin selection was the first explanation of why some services or resources are given without reciprocity from one individual to another. However, it is also well known that kin selection does not explain all occurrences of cooperation or apparent altruism; individuals give to non-relatives and resources are often traded or reciprocally exchanged amongst relatives.

5.2. The proposed investigation aims to test data on access to infants and infant handling in a group of capuchin monkeys ( Cebus apella ) against two models of cooperative theory. Firstly, the behaviour will be tested in a biological market traded for allogrooming and secondly examined for a preference for kinship. A further step will be to examine whether the two models can be combined in a joint theory. Two possible combinations can be developed; an " either/or " theory or an " collective " theory. The " either/or " theory is that kin gain access to related infants, etc, without trading for allogrooming were as non-kin must provide allogrooming for accessing unrelated infants. Alternatively, the " collective " theory would be that all individuals operate within a biological market and must trade allogrooming with the mother for access to her infant, however, when dealing with relatives the amount of grooming rendered for access to the infant is offset against the degree of relatedness.

5.3. A simple example of this latter combination would be the selling of a car. To the general public the may be sold for £5000, however when the owner's brother shows interest it may be sold to him for £4500 because of the effect of being related. The relatedness value may in this instance be equal to the £500 reduction in price of the car. The trade of money for the car is still made but the value of money rendered against the purchase of the car is less due to the added value of being related. In the case of the proposed research, it would mean that relatives may still need to barter for access to the mother's infant but because the individual is a relative does not have to render as much of a service as a non-relative. This is based on the idea that being related means an individual is less likely to harm the infant so the mother may be less nervous in handing over the infant. Inclusive fitness may motivate the individual to inspect and care for related infants but the individual may still need to convince the mother of their intentions by allogrooming her. Individuals commit to relationships through behaviour and demographic input. Dominance would be analogous to kinship affecting the market model in a similar way; and is an important consideration for this investigation.

5.4. In some instances, dominants may gain access to resources without trade at all. This " collective " model could explain this in that the value of dominance outweighs the value of the resource or the RHP of the possessor. It may not be worth fighting dominant individuals to maintain possession and in the case of infants, not fighting or resisting may result in return of the infant with reduced stress or physical harm. This would produce the idea of a threshold limit were individuals must decide whether to hold onto resources and barter or allow them to go without trade. This threshold is a concept not a fixed value. For the example of the car above, the £500 reduction for kin is not an exact value of the worth of that brother it is a proportional representation of a combination of kinship and value of the resource. This idea is highly complex and may be too complicated for modelling into the biological market. It would effectively mimic reality much closer making every situation unique. This is because assessment of the threshold would be dependent on all individuals in the interaction (partner-choice, partner-type), both the resources being traded and then inclusion of all other rules of the market such as supply and demand.

6. Supporting evidence

6.1. This proposed research comes at the end of a current Masters degree investigating distribution of allogrooming amongst female capuchin monkeys at La Vallee des Singes primate park, France. Two variations in behaviour observed in the current work have highlighted the potential for using this group in the study of biological markets and the use of captive animals allows for an experimental- type approach to investigating biological markets and kin selection. The two main points of evidence are the difference in allogrooming between the two observation periods and more specifically the difference in allogrooming received by the females from juveniles between the two periods of observations. The study was conducted over two 6-week periods and collected 161.5 hours worth of data on 5 adult females (average per female per observation period = 16.15 hours ± 1.18). In September 2002 all females had a dependent infant of less than 6 months of age while in the second study period, April 2003, all infants were older than 6.5 months and semi-independent moving and playing in the trees without mothers.

6.2. There was significantly more total allogrooming in September 2002 (S'02) than in April 2003 (A'03). In S'02 average grooming rate for a female = 330 grooming seconds per hour (gs/hr), SD = 185; in A'03, grooming rate = 163 gs/hr, SD = 105. This total includes all dyads in the group, not only focal females. Further analysis is necessary for individual differences so only totals are presented here. The case of allogrooming between adult females and juveniles also produced an interesting result. What is interesting is the variation in alloG that juveniles directed towards adults when adults had dependent infants. Juveniles received 50.2 minutes of allogrooming in S'02 and 43.8 in A'03 (0.7 mins/hour and 0.5 mins/hour respectively). Juveniles groomed adult females for 20.2 minutes in S'02 compared to only 7.5 minutes in A'03 (0.3 mins/hour and 0.09 mins/hour respectively). While these totals are much smaller than those for allogrooming between adult females this result is indicative of a greater use of allogrooming in S'02 than in A'03 for juveniles grooming adults. The low amount of allogrooming between juveniles and adult females is exaggerated because no focal samples were made of the juveniles so all data presented comes from focal samples of only the adult females. This also means that no statistical inference can be made from this dataset about interest in infants from juveniles.

6.3. While infant handling (holding and carrying) was rare, adult females, especially, showed a lot of interest in infants and exhibited other infant-directed behaviour like nuzzling, licking, smelling, allogrooming, allosuckling and lip-smacking. The behaviour of lip-smacking was unique to being performed only by an individual inspecting an infant; it never occurred between two non-infants and was never seen from an infant towards a non-infant. These factors demonstrate the importance of including all infant-directed behaviours besides simply carrying or holding, as part of data collection, because of the abundance of alternative inspection behaviours that may be performed as well as or instead of carrying. The data in this current research also supports this suggestion, in that allogrooming was more abundant in S'02 than in A'03, however, all females had their own infant, infant carrying of non-offspring was rare and other inspection behaviours were common. Motivation has not been discussed in this paper, however, note here that if motivation is inspecting infants and not handling them this may still contribute to a market effect in interchange of allogrooming for access. This will be considered when analysing mother-mother dyads (as in Henzi and Barret, 2002).

7. Discussion

7.1. Two points are worth considering further. The first is: Why do non-human primates have to groom the mother to gain access to her infant compared to the human situation were baby-sitters are paid to look after the infants? Baby-sitting should be considered as a different issue to infant handling in primates. Humans show interest in infants, similar to primates, without the motivation of being paid for this interest. Furthermore, consider the argument that the case of baby-sitters is a different market situation. Baby-sitters, in this instance, are the type-A class and are able to playoff the type-Bs (parents) against one another to secure the best wage, easiest job, etc, for themselves. The baby-sitter is providing a rare commodity (that cannot be taken by force) while the parents want to be able to leave their children in the hands of someone responsible. Another point to consider is that in the case of young children becoming baby-sitters, they very rarely look after non-relatives, often show desire to look after their younger siblings and are normally not paid by their own parents for this care. Alternative views may be that the human case is unique due to economic (monetary) systems were services are paid with a uniform currency of set value or that adult humans live in a biological and cultural system that demands they leave their children to perform certain essential tasks in their lives and therefore require assistance from a baby-sitter.

7.2. The second point concerns the function of allogrooming as a service. As part of my current research I am investigating the hygienic value of allogrooming and have previously argued that only hygiene or hedonism could really provide a service within the biological market that benefits the recipient of such services. Allogrooming could be compared to autogrooming to establish the amount of allogrooming necessary to provide such a hygienic function. However, as argued by Sade (1966), most of the hygiene maintenance is provided through allogrooming not autogrooming so I would argue against comparing the abundance of the two together. This is especially so if allogrooming also serves other functions of more social application: proximity, coalition formation and anti-agonistic behaviour. The potential for multiple functions of allogrooming complicates any comparison of the two behaviours because it means allogrooming and autogrooming do not serve the same, sole function. I would also suggest, further, hesitance in completely discounting the effect of any social function in the biological market. If only "friends" handle infants improving social proximity and social relationships through allogrooming is an easily achievable tactic, with benefits to both groomer (performer) and groomee (recipient).

8. Conclusion

8.1. The evidence in this paper supports the biological market principle (Noë et al, 1991). Motivation, in terms of function for allogrooming and likewise for infant handling (infant access), requires discussion to ascertain the reasons behind any market. The use of capuchin monkeys in a experimental approach in a captive situation should allow for a number of different variables to be controlled including supply and demand through changing group dynamics using contraception, etc. The current aim is to expand on the success of work by Henzi and Barrett (2002) and develop further functional models to facilitate or simulate variation in the biological market as found in natural group-living populations. As argued before these models may not have a universal application but could help answer some questions about variation in cooperation and competition at the level of the individual. I believe that studying access to infants in primate species using the biological market principle (and also kin selection) would give further insight into the influence of evolutionary systems on social dynamics of non-human primates and may relate to similar systems of modern humans.