| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Psychology, University of Maryland, College Park, Maryland 20742
2 Department of Biology, Georgia State University, Atlanta, Georgia 30303
* To whom correspondence should be addressed. E-mail: jherberholz{at}psyc.umd.edu
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionLiterature Cited |
|---|
Abbreviations: DI, dominance index
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionLiterature Cited |
|---|
In crustaceans as in other animals, conflicts over these resources often lead to the formation of social hierarchies characterized by animals of dominant and subordinate status. Once social ranks are established among members of the group, dominant and subordinate crayfish show clear differences in their behaviors, and distribution of resources is arranged through less violent interactions (Edwards and Herberholz, 2005).
Crayfish are characterized by their tendency to form social hierarchies through dyadic agonistic encounters even in environments that are free of resources. Bovjberg (1953) measured rank-order formation in aquaria free of any resources (except space) in groups of four similar-sized crayfish and found that a dominance order is established within a few days, remaining stable for 2 weeks. Lowe (1956) measured social rank orders in groups of four and eight crayfish in featureless aquaria and reported that hierarchies with four ranks were stable for several weeks, whereas in groups of eight animals the lower ranks were less well differentiated. In a study of hierarchy formation in groups of four similar-sized crayfish introduced to resource-free aquatic arenas, the highest ranked individuals were clearly distinguished from the other three ranks by the greater number of initiated attacks, which led to more wins (Copp, 1986). Issa et al. (1999) reported the emergence of "superdominant" animals among groups of five juvenile crayfish that were also tested in a resource-free environment. This form of dominance hierarchy with one superdominant (the largest or most aggressive one) and four lower ranked crayfish was persistent throughout the duration of the grouping.
Several studies measured aggressive interactions, both in intraspecific and interspecific pairs or groups of crayfish, in the presence of vital resources such as shelter and food (Capelli and Hamilton, 1984; Ranta and Lindström, 1992; Peeke et al., 1995; Figler et al., 1999; Usio et al., 2001; Bergman and Moore, 2003). However, most studies that analyzed agonistic elements displayed during pair-wise encounters of crayfish were conducted in featureless aquaria that provided no resources besides space (Bruski and Dunham, 1987; Figler et al., 1995; Rutherford et al., 1996; Guiasu and Dunham, 1997; Tierney et al., 2000; Goessmann et al., 2001; Herberholz et al., 2001, 2003; Schapker et al., 2002; Song et al., 2006). It was found that early stages of an encounter between well-matched pairs of crayfish are marked by an aggressive escalation that can include strikes and grappling with the claws, and bouts of tail-flips (Rubenstein and Hazlett, 1974; Bruski and Dunham, 1987; Edwards and Herberholz, 2005). At some point, fighting is interrupted by an abrupt change in the agonistic behavior of one animal as it switches from aggressive to submissive behaviors. This switch marks the change in the social status of the animal and identifies the new subordinate (Herberholz et al., 2001, 2003). After the formation of social ranks, dominant animals display a dominant posture, spend more time in shelter construction, respond aggressively to tactile stimulation, and continue to initiate most attacks. Conversely, subordinates display a submissive posture, suppress burrowing, retreat in response to tactile stimulation, and escape from the dominants’ attacks (Guiasu and Dunham, 1998; Herberholz et al., 2003; Song et al., 2006). The lingering effects of social status in dominants and subordinates vary, depending on the presence or absence of the opponent and reinforcement of social status (Yeh et al., 1996; Herberholz et al., 2003; Song et al., 2006).
Food is an important and sometimes limited resource for crayfish, and fights are of higher intensity in the presence of a more valuable food resource (Bergman and Moore, 2003). Hungry crayfish also engage in more fights than satiated animals (Hazlett et al., 1975) and escalate fights more rapidly (Stocker and Huber, 2001). Capelli and Hamilton (1984) found that aggressive interactions were strongly reduced when food and shelters were plentiful.
Many of the previously mentioned studies were conducted under the assumption that the rapid formation of a dominance hierarchy among crayfish serves to determine the order and extent of access to future resources. The assumption was based on examples from the literature demonstrating that preferential access to resources is determined by social rank in several animal species. However, whether dominant status is in fact correlated with control of resources varies for each species and each kind of resource, and inverse relationships between social dominance and access to resources (especially food) have also been reported (Strum, 1982; Craig, 1986; Gerald, 2002; Knowles et al., 2004). To our knowledge, this has not been explicitly tested in crayfish. Gherardi and Cioni (2004) analyzed agonistic behavior during interspecific fights between pairs of crayfish in a resource-free environment and again after resources (food or shelter) were provided. They showed that the crayfish species that dominated in the absence of any resources was also more aggressive and won more fights in the presence of a resource. Since the crayfish interacted in pairs, however, rank orders were not established and the authors did not intend to correlate individual social ranks with access to the resources.
We allowed groups composed of three juvenile crayfish of the same species to establish a dominance hierarchy and recorded the social ranks of all individuals before and after a single food resource was provided. We also measured the time that differently ranked animals spent near or in contact with food after it was introduced to the test arena.
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionLiterature Cited |
|---|
Thirty size-matched crayfish (size: mean 2.6 cm ± SD 0.2 cm; 2.2–3.0 cm, measured from rostrum to telson) were assigned to 10 experimental groups. The maximal size difference between members in any of the 10 groups was 0.3 cm, and the average size difference among members of all groups was 0.13 ± 0.07 cm. Members of each group were tested for physical intactness and marked with small dots of different colors on their carapaces. All experiments were conducted during the day at about the same hours.
Three animals were simultaneously introduced to a test arena (15 x 30 x 20 cm) with gravel floors (height: 2.5 cm, crushed coral chips of 4–6 mm in size) and filled with water (height: 10 cm). The walls were covered with white paper to prevent any visual distraction. The arena was separated into two parts by an opaque divider. Agonistic interactions among the three crayfish took place in the larger compartment (60% of total space). The behavior of all three animals was videotaped for 30 min with a video camera (Canon XL-1) mounted on a tripod above the aquarium and connected to a TV monitor (Panasonic 2010-Y).
After the 30-min period, for each group one piece of chicken liver (weight range: 11.9–12.1 g) was placed into the smaller separated compartment (40% of total space) of the tank. The animals had no previous experience with chicken liver as a food source. The piece of liver was large enough that crayfish of the size used in our study were unable to move it, and it was in a chunk that did not float or diffuse. The divider was subsequently lifted, and the behavior of all animals was then recorded for another 50 min. Since we were unable to determine how much food each animal consumed, we analyzed the length of time spent near (within one body length of) the food and in direct physical contact with it. All recordings were stored in mini-DV format. Agonistic behaviors were measured using single-frame analysis on a TV monitor (Panasonic 2010-Y), and durations were measured using the internal clock of the video camera. Aggressive behaviors were scored as "approach" (moving forward toward an opponent with lowered claws), "threat display" (spread of claws), and "attack" (launching forward with claws open and an elevated posture). Submissive acts were scored as "retreat" (walking away from an approaching or attacking opponent) and "escape" (using rapid tail-flips to move away from an approaching or attacking opponent). Behaviors displayed during the second phase of the experiment (with the food present) were always analyzed first so that the social rank of the animals was unknown at the time of analysis. The dominance index (DI) for each animal was calculated as (total number of recorded aggressive acts / [total number of recorded aggressive acts + total number of recorded submissive acts]) x 100.
If applicable, data are presented with means and standard deviation. Since the majority of data did not pass normality and equality tests, we used nonparametric statistical tests (SPSS version 12.0) for dependent data (sample size >2: Friedman test; sample size = 2: Wilcoxon signed rank test). The Spearman rank test was used for correlation analysis.
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionLiterature Cited |
|---|
0.01; Wilcoxon signed rank test: P 0.05 for all three variables). Within groups, the differences in DIs were clearly more pronounced between alphas and betas (mean: 29.9 ± 14.1%; min: 8.9%, max: 51.8%) than between betas and gammas (15.2 ± 10.3%; min: 4.2%, max: 33.4%). In 7 of 10 groups, alphas achieved DIs that were at least 20% higher than the DIs of betas. However, this was true only in 3 of 10 groups for the DIs of betas compared with the DIs of gammas (Fig. 1A). Animals of different ranks did not differ in body size (alphas: 2.6 ± 0.3 cm; betas: 2.6 ± 0.2 cm; gammas: 2.6 ± 0.3 cm; Friedman test: P = 0.968).
|
0.05). The reduction in aggressive behavioral acts was smallest in alpha animals (53.1%; n = 10) as compared with both the betas (66.3%; n = 10) and gammas (65.9%; n = 10). Surprisingly, previously assigned alphas now displayed 26.9 ± 21.4 aggressive acts but not one single submissive act during the 50 min when food was available (Fig. 1B). Therefore, the DI for each alpha animal now was 100.0 ± 0.0% (n = 10). Former betas displayed 11.8 ± 7.7 aggressive acts and 9.0 ± 7.4 submissive acts, which led to an average DI of 53.8 ± 21.7% (min: 0%, max: 75.0%; n = 10); and former gammas displayed 8.4 ± 6.7 aggressive acts and 9.9 ± 6.9 submissive acts in the presence of food. The average DI of gammas now was 47.1 ± 16.7% (min: 20.0%, max: 75.0%; n = 10). The increase in DIs for alphas after food was made available was significant (Wilcoxon signed rank test: P 0.05), while the decrease in DIs for betas and gammas was not (Wilcoxon signed rank tests: P = 0.139 and P = 0.333, respectively). Although all alpha animals retained their highest ranking in the presence of food, rank reversal was observed between former betas and gammas in 2 of 10 groups (20%). Two medium-ranked animals dropped down to the lowest rank, while their formerly lower ranked opponents achieved medium positions in the order when food was present (Fig. 1B). When compared with the resource-free period, the difference between the DIs of alphas and betas increased (46.2 ± 21.7%, n = 10), while it decreased between the DIs of betas and gammas (6.7 ± 28.8%; n = 10) when food was present.
We also measured the time spent by all individuals near (within one body length) or in direct physical contact with the food during the 50 min (3000 s) when food was present (Fig. 2). As shown in Figure 2A, animals that were assigned as alphas during the resource-free testing period spent 222.8 ± 203.5 s (7.4% of total time) within a body length of the food and 1576.7 ± 711.6 s (52.6% of total time) in direct physical contact with the food. Betas spent 139.0 ± 203.9 s (4.6% of total time) within a body length of the food and 1145.5 ± 389.3 s (38.3% of total time) touching the food. Finally, the lowest ranked animals (gammas) spent 177.5 ± 130.5 s (5.9% of total time) within a body length of the food but only 582.3 ± 379.1 s (19.4% of total time) in contact with the food. The differences in times spent within one body length of the food did not differ significantly among animals of different ranks (Friedman test: P = 0.273), but we found significant differences for times spent in physical contact with the food (Friedman test: P 0.01; Fig. 2A). The highest ranked animals (alphas) spent more time in contact with the food than the medium-ranked animals (betas) although the difference was marginally not significant (Wilcoxon signed rank tests: P = 0.059). The lowest ranked animals (gammas) spent significantly less time in contact with the food than the alphas (Wilcoxon signed rank tests: P 0.05) or betas (Wilcoxon signed rank tests: P 0.05). In 7 of 10 groups, the alpha animals spent substantially more time touching the food than beta animals did (difference: 694.4 ± 384.7 s; n = 7). In three groups, the betas spent slightly more time than alphas in contact with the food (difference: 183.0 ± 51.8 s; n = 3). Gammas, the animals ranked lowest according to the initial resource-free interaction period, spent less time in physical contact with the food than did all animals that were ranked higher. Figure 2B underlines the finding that higher ranked crayfish contacted the food for longer periods than lower ranked animals. This is evidenced by a significant positive correlation between the DIs of all animals and the time spent in contact with the food (Spearman correlation coefficient: 0.570, P 0.01; n = 30).
|
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionLiterature Cited |
|---|
We allowed previously isolated (socially naïve) juvenile crayfish to interact agonistically for 30 min in a resource-free environment and subsequently presented them with an attractive food resource. It is unlikely that fighting and hierarchy formation in the resource-free environment was an artifact of the three crayfish placed in a confined space. Although crayfish do not live naturally in highly organized social groups, they do occur in large densities in their natural habitats. Such densities reported under natural conditions exceed the density used in our study (Nyström, 2002). Moreover, crayfish can chemically recognize the social status of conspecifics, which reduces aggression during repeated agonistic encounters (Zulandt Schneider et al., 2001).
We found that social ranks established in the featureless environment predicted the relative durations of access to food after it was made available. Although we were unable to determine how much food the animals consumed, time spent in physical contact with the food is a clear measurement of accessibility to the resource.
Higher ranked animals spent more time in direct contact with the food than did lower ranked animals. Position in the rank order, measured as dominance index, and time spent touching the food were positively correlated. Interestingly, the highest ranked animals (alphas) consolidated their position in all groups, completely suppressing any submissive behavioral patterns after food was supplied. This might be an indication for "taking possession" and acknowledging the value of the resource by the dominant animals, which in turn reduces their willingness to retreat from it. It is also interesting to note that dominant, intermediate (beta), and subordinate (gamma) crayfish displayed no difference in the amount of time spent in the vicinity of the resource. Obviously, moving around within one body length of the resource did not exceed the threshold for eliciting active defense of the resource by the higher ranked animals. In addition, animals of all ranks did spend time in contact with the food; no animal was completely excluded from accessing the resource. The most likely explanation for this lies in the size of the food. We chose a relatively large (and heavy) piece of liver as the food resource to prevent individual crayfish from carrying it away. As a consequence, animals of different ranks would sometimes access the food from opposite sides, which allowed lower ranked animals to temporarily contact the food unnoticed by the higher ranked animals.
In the presence of the food resource, no rank reversals occurred between alphas and animals of lower rank, and rank reversal between intermediate and lowest ranked animals happened in only 20% of all cases. This shows that the previously established rank orders remained relatively stable throughout the experiments. However, the difference in dominance indices of betas and gammas was reduced after the food was introduced, indicating a decreased distinction between these lower ranks. In contrast, the differences in dominance indices of alphas versus betas increased with resource availability, thus suggesting the presence of one "superdominant" crayfish in each group. Whether this is in fact related to the introduced food resource is unclear, since emerging "superdominants" have also been described in studies were space was the only resource present (Copp, 1986; Issa et al., 1999).
The significant reduction in overall agonistic displays in the presence of the food can be explained by (i) the larger spatial area available, (ii) the time assigned for food inspection and "handling" (Corkum and Cronin, 2004; Baird et al., 2006), and (iii) the typical decrease in aggressiveness over time in groups or pairs of crayfish (Issa et al., 1999; Herberholz et al., 2001, 2003). However, the highest ranked animals did not retain their position simply because the overall number of interactions decreased. In fact, alpha animals maintained a high level of aggressiveness in the presence of the food, as indicated by the smaller reduction in aggressive behavioral acts as compared with both the betas and gammas. This probably contributed to their status consolidation during this period.
In each experiment, all three crayfish were introduced to the tank at the same time and then allowed to interact for only 30 min before the food was provided immediately afterward. Since the environment was new to all animals, an effect of prior residence on hierarchy formation was not measured. In lobsters, escape behavior decreased and aggression increased with time of residence (Cromarty et al., 1999), but in pairs of juvenile crayfish, size rather than prior residence was correlated with success in obtaining and defending a resource (Figler et al., 1999). Thus it would be interesting to test how previous experience with the environment or with specific resources affects hierarchy formation in groups of crayfish.
In our study, we measured access to food after a relatively short interaction period. We found that this was sufficient to establish a social rank order that determined subsequent access to an immediate resource. As crayfish live together for longer periods, agonistic behaviors change, and aggression is reduced (Issa et al., 1999). Consequently, access to the resource may be differently distributed among members of a social group that interact for longer periods. Ranta and Lindström (1992) showed that unevenly distributed food evokes agonistic interactions among groups of crayfish housed in nursery ponds. They found that after 2 nights, larger members of the group had evicted smaller animals from shelters that were closer to a single point of food distribution. Conversely, if lower ranked animals are denied access to food for long periods, as they become hungrier they may become more aggressive and more willing to challenge higher ranked animals (Hazlett et al., 1975; Stocker and Huber, 2001). Clearly, the formation and maintenance of the hierarchy depends on many factors, including shared time in the same environment; type and availability of the resource; and previous experiences with conspecifics, environment, and resource.
We found that short-term grouping of equal-sized juvenile crayfish is sufficient to lead to the formation of a dominance hierarchy with distinct social ranks. The order of these ranks, quickly established in an environment where no immediate access to resources was provided, was a clear indicator of the length of time subsequently spent in contact with an introduced food resource. Higher ranked animals had more access to the food than did lower ranked animals, which shows the benefit of superior social rank and why it is worth fighting for it, even in the absence of immediate reward other than procurement of personal space.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
Literature Cited |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionLiterature Cited |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. J. Farca Luna, J. I. Hurtado-Zavala, T. Reischig, and R. Heinrich Circadian Regulation of Agonistic Behavior in Groups of Parthenogenetic Marbled Crayfish, Procambarus sp. J Biol Rhythms, February 1, 2009; 24(1): 64 - 72. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
