Juvenile hormone and aggression in honey bees.

We determined whether defense by individual bees against non-nestmates in honey bees (Apis mellifera) is correlated with their juvenile hormone (JH) titers, which are known to vary developmentally and seasonally. We bioassayed winter and summer bees for aggressive and non-aggressive individuals. Bees in winter could not be distinguished by task group, but bees in summer were segregated into nurses and guards. JH titers were correlated with aggressive behavior at two levels. First, winter bees and summer nurses, known to have lower JH titers, both showed less aggression toward foreign bees than did summer guards. Second, aggressive individuals had significantly higher JH titers than did non-aggressive bees within each colony. Inter-colonial variation in aggressiveness was maintained during summer and winter, suggesting a genetic basis for these differences. An alarm pheromone test further substantiated the existence of inter-colonial differences. We found significant variation in JH titers among different colonies, but this variation was not significantly associated with colony-level aggressiveness. The correlation between JH and levels of aggressiveness within a colony suggests a regulatory role for JH, but variation among colonies involves factors other than JH.

Interfamily variation in comb wax hydrocarbons produced by honey bees.

The hydrocarbons of honeybee comb wax vary significantly between colonies. This variation is explained in part by genetic (familial) differences among colonies. Even though significant differences in wax hydrocarbons exist among families, there is a high level of consistency within and among families in a correlation analysis, indicating structural constancy in comb wax. The significance of these results in interpreting the potential role of comb wax in the nestmate recognition system of the honeybee is discussed.

Kin discrimination within honey bee (Apis mellifera) colonies: An analysis of the evidence.

Compelling evolutionary arguments lead to the prediction that honey bee workers should discriminate between supersisters and half-sisters within colonies. We review the theoretical support for discrimination during swarming, queen rearing, feeding, and grooming. A survey of the data that tests whether such discrimination takes place shows that, despite substantial effort in a number of laboratories, there is no conclusive evidence for intracolony discrimination in any of the postulated contexts. The strongest suggestive data is in the critical context of queen rearing, but flaws in experimental design or analysis make the best available tests inconclusive. We present new data that shows that cues exist on which discriminations can be made among adult workers in nestmate recognition interactions and in feeding interactions, but our data does not differentiate between subfamily recognition and recognition associated with color phenotypes. We conclude that while selection may favor discrimination between supersisters and half-sisters, as a practical matter such discriminations play no role, or only a minor role, in the biology of the honey bee.

Honeybee nestmate recognition: Effects of queen fecal pheromones.

Previous work has shown that queen honeybees,Apis mellifera, produce waxy esters composed of 8-14 carbon acids and 6-14 carbon alcohols in their feces. We tested these esters for effects on nestmate recognition; 11 of the 12 esters tested significantly modified the recognition characteristics of worker honeybees. Pairwise tests showed that workers can discriminate between at least some pairs of queen esters and that workers can discriminate between a queen ester and hexadecane (another known nestmate recognition cue). These results suggest that a queen may use the esters to enable workers to recognize her or to scent-mark her colony.

The behavioral genetics of colony defense in honeybees: genetic variability for guarding behavior.

Guard honeybees stand at the entrance of colonies and facilitate the exclusion of nonnestmates from the colony. In this study, we examined the hypothesis that genetic variability among individuals in colonies might explain variability in guarding activity. To do this, we cross-fostered honey bees between colonies with high-defensive responses and colonies with low-defensive responses in alarm pheromone tests. Individuals from high-defensive colonies were more likely to guard in their own colonies (controls) than cross-fostered bees from low-defensive colonies. Cross-fostered high-defensive bees also were more like to guard in low-defense colonies. These results support the hypothesis that interindividual differences in guarding behavior are at least partially under genetic control. A positive correlation between number of guards and response to alarm pheromone demonstrates a link between behaviorally separated components of the overall defensive response.

Comb wax mediates the acquisition of nest-mate recognition cues in honey bees.

Honey bees, Apis mellifera, acquire nest-mate recognition cues from wax, the predominant material used in nest construction. Exposure of a newly emerged worker bee to wax-comb substrate significantly reduced the acceptability of that worker to sister bees. Cues acquired from the comb provided colony-specific information about the identity of worker bees; moreover, the effect of comb exposure has been previously shown to override individually produced cues. Food odors (anise oil), when dissolved in paraffin wax, affected worker-recognition characteristics but food odors did not affect these characteristics when fed to bees in sugar candy. Paraffin wax alone did not affect the recognition cues of bees, showing that the wax can be a neutral medium for the transmission of cues. The wax comb in the colony and the hydrocarbon outer layer of the bee cuticle may be a continuous medium for any hydrocarbon-soluble substances used by honey bees in nest-mate recognition; if so, a mechanism by which environmental cues are acquired by honey bees is provided.

Kin recognition in social bees.

Kin recognition in social insects has become a central issue in sociobiology because studies of the recognition abilities of social insects provide a test of kin selection theory. W.D. Hamilton(1) formalized kin selection theory by showing how individuals can gain fitness by increasing the reproductive output of relatives (kin). The social interactions of individuals, or groups, should be influenced by the genetic structure of the population. The ability to recognize kin can increase the adaptive value of social behavior by modulating it according to genetic relationship. From this, the specific prediction emerges: if individuals can distinguish among others with which they interact on the basis of the degree to which they are related, then behavior should be biased preferentially toward more closely related reproductive individuals.

Kin discrimination by worker honey bees in genetically mixed groups.

We tested the hypothesis that in a genetically mixed assemblage of worker honey bees, individual workers would behave differently toward unfamiliar sisters than toward unfamiliar nonsisters. Groups of worker honey bees of mixed genetic composition were assembled by collecting pupae from separate colonies and placing the worker bees together on eclosion. A total of 10 workers, 5 from each of two kin groups, were used to form each group. When the workers were 5 days old, a worker of one of the two kin groups was introduced into the mixed group. This worker had previously been held in a group of its sisters, without contact with queen or nonsister bees. The interactions with the introduced bee indicate that in a mixed kin group, individual workers learn the composite identity of the group and do not attack unfamiliar bees differentially on the basis of kinship. However, kinship does influence the total number of interactions in which an introduced bee engages when placed in a genetically mixed group; bees interacted significantly more often with sisters than with nonsisters. There was a trend for bees to be involved in more feeding interactions with sisters. This finding indicates an ability of a bee to learn and use its own cues. In mixed groups, each bee maintains its genotypically correlated identity; the bees' odors do not comingle into a "group" or "gestalt" odor. The significance of these results is discussed in light of the genetic structure of natural colonies of honey bees.

Individual recognition and learning of queen odors by worker honeybees.

A honeybee queen is usually attacked if she is placed among the workers of a colony other than her own. This rejection occurs even if environmental sources of odor, such as food, water, and genetic origin of the workers, are kept constant in laboratory conditions. The genetic similarity of queens determines how similar their recognition characteristics are; inbred sister queens were accepted in 35% of exchanges, outbred sister queens in 12%, and nonsister queens in 0%. Carbon dioxide narcosis results in worker honeybees accepting nonnestmate queens. A learning curve is presented, showing the time after narcosis required by workers to learn to recognize a new queen. In contrast, worker transfers result in only a small percentage of the workers being rejected. The reason for the difference between queens and workers may be because of worker and queen recognition cues having different sources.

Behavioral control of workers by queens in primitively eusocial bees.

Queens of Lasioglossum zephyrum, a primitively eusocial bee, are considerably more active than workers. The queen's behavior stimulates worker activities; removal of the queen results in a marked reduction in activities of other bees. The queen not only activates workers but also directs them by a primitive recruitment behavior suggestive of tandem running of highly eusocial ants.

Agonistic behavior in the German cockroach, Blattella germanica.

Agonistic behavior in Blattella germanica is delineated with an emphasis on fighting techniques and population factors affecting aggressiveness. Male-male, male-female, and male-male encounters are not significantly different in either level of aggressiveness or frequency. Female-female carrying oothecae are, however, more aggressive than other female-female. The mean intensity of aggression increases as population density is increased, but number of contacts per individual per unit time remains constant. Defense of specific, territories does not occur and aggressive interactions are most common during the dark portion of the photocycle, when many of the cockroaches are observed to be foraging.