A geometrical model for the effect of interference on food intake

Interference competition is often due to kleptoparasitism (food stealing). In which case, the attack distance, the distance over which one animal attacks another in an attempt to steal food, determines to a large extent the competitor density range over which interference significantly affects the intake rate of foraging animals.We develop a simple model of kleptoparasitism containing three parameters: attack distance, the density of foraging animals and a single dimensionless parameter α which summarizes the non-geometrical aspects of the interference process. Dominant and subdominant animals are not considered separately. The model predicts that the average intake rate will decrease exponentially with animal density and that a measure of the strength of interference depends on attack distance squared.The simple model is compared with a much more detailed individual-based foraging model from the literature. Simulated average intake rates are indeed well approximated by an exponential decrease with competitor density. Also the measure of interference behaves in the way expected from the simple model. By explaining the shape of the relationship between intake rate and animal density, the simple model provides insight into the behaviour of the detailed behavioural model.Insight into the role of geometry is important in the interpretation of field results and in the further development of detailed foraging models.     

Do fledglings choose wisely? An experimental investigation into social foraging behaviour

Many cooperative bird species have an extended period of post-fledging care. Despite the fact that this period of care can last up to several months, it remains a relatively understudied stage of chick development. This period, when young are actively begging but highly mobile, provides an opportunity for young to maximise the amount of care they receive by selectively choosing particular adults to beg from. In pied babblers Turdoides bicolor (a cooperatively breeding passerine), fledglings closely follow foraging adults and beg for food regularly (a behavioural interaction termed social foraging). Using a combination of natural observations and experimental manipulations, we found that fledgling pied babblers preferentially socially forage with adult care-givers who have high foraging success, since this results in young receiving more food. By supplementally feeding adults to artificially increase their foraging success, we increased the proportion of time that fledglings chose to socially forage with them, confirming that fledglings are selectively choosing dyadic interactions with the best adult foragers. These results indicate that pied babbler fledglings are sensitive to and can respond to short-term changes in adult foraging success, enabling them to maximize their nutritional intake, a behavioural adjustment that has long-term benefits in this system.     

Kleptoparasites: a twofold cost of group living for the colonial spider, Metepeira incrassata (Araneae, Araneidae)

Several species of kleptoparasitic and araneophagic spiders (Araneae: Family Theridiidae, Subfamily Argyrodinae) are found in colonial webs of the orb-weaving spider Metepeira incrassata (Araneae, Araneidae) from Mexico, where they steal food and/or prey upon their spider hosts. Census data from natural M. incrassata colonies reveal that the incidence of these species increases with colony size. This pattern may reflect the presence of several other orb-weaving spiders, each with their own kleptoparasitic species, invading larger M. incrassata colonies. As the number of these associated spiders increases, so does the density and number of Argyrodinae species in M. incrassata colonies, suggesting that associated spiders might reduce their own kleptoparasite load by building their webs within M. incrassata colonies. This represents a twofold cost to M. incrassata, as a field enclosure experiment revealed that a primarily kleptoparasitic species (Argyrodes elevatus) may reduce prey available to their hosts, but a kleptoparasitic/araneophagic species (Neospintharus concisus) inflicts high mortality upon M. incrassata. However, the cost of kleptoparasitism and predation by these species may be offset in part for M. incrassata individuals in large colonies by certain defensive mechanisms inherent in groups, i.e., “attack-abatement” and “selfish herd” effects. We conclude that increased occurrence of kleptoparasitic and/or predatory Argyrodinae spiders is a consequence of colonial web building and is an important potential cost of group living for colonial web-building spiders.     

Predator type, not body condition, influences positioning within whirligig groups

The anti-predator responses of an animal may depend on the species of predator attacking. For example, ground-dwelling monkeys respond with different warning calls in response to attacks from hawks from above vs attacks from leopards at the side. In grouping animals, it has usually been assumed that the best anti-predator behavior against most predators is to move toward the center of a group. However, little work has been done to study positioning of individuals in response to different predators and whether it might be influenced by gender or body condition. An optimality model is developed that gives predictions for how the needs of predator avoidance, drafting, and foraging could lead to one optimal location within the group. We then conducted a laboratory experiment with individually marked whirligig beetles (Coleoptera: Gyrinidae). We exposed them to an aerial (bird) or aquatic (fish) attack and determined changes in their distance to the front of groups at different speeds of water. We found that female whirligigs occupied the front of a group in response to a fish attack in fast water, but the back after an aerial bird attack, while the reverse was true for males. In slow water, different results were found; in response to a bird attack, females were more likely to be found in the front compared with males. Body condition was not correlated to the front/back position of either males or females before or after attacks in our study.     

Age and sex influence marmot antipredator behavior during periods of heightened risk

Animals adjust their antipredator behavior according to environmental variation in risk, and to account for their ability to respond to threats. Intrinsic factors that influence an animal’s ability to respond to predators (e.g., age, body condition) should explain variation in antipredator behavior. For example, a juvenile might allocate more time to vigilance than an adult because mortality as a result of predation is often high for this age class; however, the relationship between age/vulnerability and antipredator behavior is not always clear or as predicted. We explored the influence of intrinsic factors on yellow-bellied marmot (Marmota flaviventris) antipredator behavior using data pooled from 4 years of experiments. We hypothesized that inherently vulnerable animals (e.g., young, males, and individuals in poor condition) would exhibit more antipredator behavior prior to and immediately following conspecific alarm calls. As expected, males and yearlings suppressed foraging more than females and adults following alarm call playbacks. In contrast to predictions, animals in better condition respond more than animals in below average condition. Interestingly, these intrinsic properties did not influence baseline time budgets; animals of all ages, sexes, and condition levels devoted comparable amounts of time to foraging prior to alarm calls. Our results support the hypothesis that inherent differences in vulnerability influence antipredator behavior; furthermore, it appears that a crucial, but poorly acknowledged, interaction exists between risk and state-dependence. Elevated risk may be required to reveal the workings of state-dependent behavior, and studies of antipredator behavior in a single context may draw incomplete conclusions about age- or sex-specific strategies.     

Foraging behavior in early Atlantic cod larvae (Gadus morhua) feeding on a protozoan (Balanionsp.) and a copepod nauplius (Pseudodiaptomussp.)

Food limitation is likely to be a source of mortality for fish larvae in the first few weeks after hatching. In the laboratory, we analyzed all aspects of foraging in cod larvae (Gadus morhua Linnaeus) from 5 to 20 d post-hatching using protozoa (Balanion sp.) and copepod nauplii (Pseudodiaptomus sp.) as prey. A camera acquisition system with two orthogonal cameras and a digital image analysis program was used to observe patterns of foraging. Digitization provided three-dimensional speeds, distances, and angles for each foraging event, and determined prey and fish larval head and tail positions. Larval cod swimming speeds, perception distances, angles, and volumes increased with larval fish size. Larval cod swam in a series of short intense bursts interspersed with slower gliding sequences. In 94% of all foraging events prey items were perceived during glides. Larval cod foraging has three possible outcomes: unsuccessful attacks, aborted attacks, and successful attacks. The percentage of successful attacks increased with fish size. In all larval fish size classes, successful attacks had smaller attack distances and faster attack speeds than unsuccessful attacks. Among prey items slowly swimming protozoans were the preferred food of first-feeding cod larvae; larger larvae had higher swimming speeds and captured larger, faster copepod nauplii. Protozoans may be an important prey item for first-feeding larvae providing essential resources for growth to a size at which copepod nauplii are captured. Received: 20 April 1999 / Accepted: 12 January 2000     

Effects of aggregation on feeding and survival in a communal wren

Summary Foraging by a social wren, Campylorhynchus nuchalis (Troglodytidae), in a tropical savanna habitat is not enhanced by aggregation. Data for marked individuals show that solitary foraging results in a higher capture rate than foraging near others. We find no evidence of imitative foraging, as individuals actively avoid successful foragers following a capture and successful foragers do not restrict their search to recently productive stations or techniques. Captures are seldom temporally clumped, and clumping is probably not pronounced enough to favor imitation. Juveniles show no greater tendency to respond to captures of others, or to succeed in foraging in a group, than do adults. Aggregation is probably disadvantageous for foraging because of dispersed, scarce, cryptic, and noneruptive prey and because of the searching technique of these foliage-gleaning insectivores. If predator avoidance is enhanced by aggregation, it does not result in either increased survival or increased foraging efficiency in large groups, even by juveniles.     

Do yellow-bellied marmots respond to predator vocalizations?

We conducted four experiments to determine whether yellow-bellied marmots, Marmota flaviventris, discriminate among predator vocalizations, and if so, whether the recognition mechanism is learned or experience-independent. First, we broadcast to marmots the social sounds of coyotes, Canis latrans, wolves, Canis lupus, and golden eagles, Aquila chrysaetos, as well as conspecific alarm calls. Coyotes and eagles are extant predators at our study site, while wolves have been absent since the mid-1930s. In three follow-up experiments, we reversed the eagle call and presented marmots with forward and reverse calls to control for response to general properties of call structure rather than those specifically associated with eagles, we tested for novelty by comparing responses to familiar and unfamiliar birds, and we tested for the duration of predator sounds by comparing a wolf howl (that was much longer than the coyote in the first experiment) with a long coyote howl of equal duration to the original wolf. Marmots suppressed foraging and increased looking most after presentation of the conspecific alarm call and least after that of the coyote in the first experiment, with moderate responses to wolf and eagle calls. Marmots responded more to the forward eagle call than the reverse call, a finding consistent with a recognition template. Marmots did not differentiate vocalizations from the novel and familiar birds, suggesting that novelty itself did not explain our results. Furthermore, marmots did not differentiate between a wolf howl and a coyote howl of equal duration, suggesting that the duration of the vocalizations played a role in the marmots’ response. Our results show that marmots may respond to predators based solely on acoustic stimuli. The response to currently novel wolf calls suggests that they have an experience-independent ability to identify certain predators acoustically. Marmots’ response to predator vocalizations is not unexpected because 25 of 30 species in which acoustic predator discrimination has been studied have a demonstrated ability to respond selectively to cues from their predators.     

Age-dependent vigilance in winter aggregations of cooperatively breeding white-winged choughs (Corcorax melanorhamphos)

Summary Winter aggregation of cooperatively breeding white-winged choughs (Corcorax melanorhamphos) was examined to elucidate the advantages for different group members. Adults have higher vigilance burdens than juveniles when foraging in their regular group. Aggregation of groups in winter enables the older members to reduce vigilance and increase the time spent foraging. Vigilance was lowest and foraging time highest in aggregations of 20 individuals but the size of cooperatively breeding groups did not exceed fifteen. Despite increased foraging time, choughs did not consume more food items when aggregating.     

A forest monkey’s alarm call series to predator models

Some non-human primates produce acoustically distinct alarm calls to different predators, such as eagles or leopards. Recipients respond to these calls as if they have seen the actual predator, which has led to the notion of functionally referential alarm calls. However, in a previous study with free-ranging putty-nosed monkeys (Cercopithecus nictitans martini), we demonstrated that callers produced two acoustically distinct alarm calls to eagle shrieks and leopard growls, but both alarm calls were given to both predators. We can think of two basic explanations for this surprising result, a methodological and theoretical one. Firstly, acoustic predator models may not always be suitable to test alarm call behaviour in primates, sometimes causing uncharacteristic behaviour. Secondly, referential alarm calling may not be a universal feature of primate alarm call systems. Considering the methodological and theoretical importance of these possibilities, we conducted a follow-up study using life-sized leopard, eagle, and human models on the same population and compared the resulting vocal responses to those given to acoustic predator models. We compared the alarm call series given to each of these predator model types and found a considerable degree of consistency suggesting that the mode of presentation did not affect anti-predator calling strategies. However, evidence for audience effects on calling behaviour was inconclusive. While it appears that predator class is reliably encoded by different call series types irrespective of the mode of presentation, observations of these same call series given in non-predatory contexts indicate that predator class is unlikely to be the relevant organising principle underlying the alarm-calling behaviour in this species. We conclude by offering an alternative, non-referential, account of the alarm-calling system exhibited by this species. Electronic supplementary material The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

The paradox of risk assessment: comparing responses of fathead minnows to capture-released and diet-released alarm cues from two different predators

Summary. Many aquatic prey are known to use chemical alarm cues to assess their risk of predation. In fishes, such alarm cues can be released either through damage of the epidermis during a predatory attack (capture-released) or through release from the predator feces (diet-released). In our study, we compared the importance of capture- versus diet-released alarm cues in risk assessment by fathead minnows (Pimephales promelas) that were na?ve to fish predators. We utilized two different fish predators: a specialized piscivore, the northern pike (Esox lucius) and a generalist predator, the brook trout (Salvelinus fontinalis). Handling time of pike consuming minnows was much shorter than for trout consuming minnows, likely resulting in less epidermal damage to the minnows during attacks by pike. In accordance with this, minnows showed a less intense antipredator response to capture-released cues from pike than capture-released cues from trout. This represents a paradox in risk assessment for the minnows as they respond to the specialized piscivore, the more dangerous predator, with a less intense antipredator response. In contrast, the minnows showed a stronger antipredator response to the specialized piscivore than to the generalist when given diet cues. This work highlights the need for researchers to carefully consider the nature of the information available to prey in risk assessment.     

Spatial position and feeding success in ring-tailed coatis

The location of an animal within a social group has important effects on feeding success. When animals consume quickly eaten food items, individuals located at the front edge of a group typically have greater foraging success. When groups feed at large clumped resources, dominant individuals can often monopolize the resource, leading to higher feeding success in the center of the group. In order to test these predictions, behavioral data relating foraging success to within-group spatial position were recorded from two habituated groups of ring-tailed coatis (Nasua nasua) in Iguazu, Argentina. Foraging success did not fit expected patterns. When feeding on small ground litter invertebrates, coatis had the same foraging success at all spatial positions. This pattern likely resulted from an abundance of invertebrates in the ground litter. When feeding on fruit, individuals in the front of the group had greater feeding success, which was driven by the relatively quick depletion of fruit trees. Dominant juveniles were often located in the front of the group which led to increased access to food. This resulted in higher feeding success on fruits but simultaneously increased their risk of predation. Although groups typically became more elongated and traveled faster when feeding on fruit, it did not appear that the coatis were drastically changing their spacing strategies when switching between the two food types. Paradoxically, spatial position preferences during invertebrate foraging appeared to be driven by fruit trees. Because fruit trees were encountered so frequently, juveniles ranging at the front edge of the group during invertebrate foraging were the first to arrive at fruit trees and thus had higher foraging success. This study demonstrates the importance of how food patch size and depletion rate affect the spatial preferences of individuals.     

Colonial breeding and nest defence in Montagu's harrier (Circus pygargus)

We assessed whether colonial breeding allows individuals to decrease their investment in predator defence, by presenting decoys of owls, foxes and crows to Montagu's harrier, Circus pygargus. Decoy detection increased with colony size, as did the number of individuals mobbing the decoy. The number of mobbers was greater for predators potentially risky for the adults (owl or fox) than for non-dangerous predators (crow). Recruits (breeding neighbours, fledglings and non-breeders) were present a lower percentage of the time, and attacked and alarm called less frequently than tested individuals. Nevertheless, the overall attack rate on the predator increased with the number of mobbers. When the size of the mobbing group increased, individuals were more likely to attack predators that represented a risk for adults, but did so less intensively and with a lower frequency of close dives. Thus, coloniality decreased the individual costs of defence in terms of risk taken, whilst enhancing defence efficacy. Birds alarm called more intensively when presented with dangerous predators than with the crow. The number of recruits significantly increased with increasing alarm rate of the tested individuals, even when taking colony size into account. Furthermore, the alarm rate of the tested birds also had a significant effect on the proportion of recruits that engaged in attacks against dangerous predators but not against the crow. The higher recruitment and attack rates for dangerous predators were thus apparently modulated through alarm calling. We discuss whether tested birds may manipulate recruits' behaviour to lessen their own risk.     

Chick-a-dee call variation in the context of “flying” avian predator stimuli: a field study of Carolina chickadees (<Emphasis Type="Italic">Poecile carolinensis</Emphasis>)

Chick-a-dee calls function in social organization in Poecile (chickadee) species. Recent field and aviary studies have found that variation in chick-a-dee calls relates to the type or proximity of avian predator, or level of threat. Earlier studies on calls in the context of predator stimuli have typically used stationary and perched predator models. For chickadees and other small songbirds, more frequently detected and more dangerous avian predatory stimuli are flying predators. In the present study, we tested whether simulated flying avian predator and control models influenced chick-a-dee calling behavior of wild Carolina chickadees, Poecile carolinensis. At 20 independent field sites, chickadee subjects were presented with wooden models that were painted to resemble either a predatory sharp-shinned hawk (Accipiter striatus) or a blue jay (Cyanocitta cristata) and that were made to “fly” down a zip line near a feeding station chickadees were using. The note composition of chick-a-dee calls was affected by both the flight of stimuli and type of model. Call variation in this flying predator context suggests interesting similarities and differences with experimental findings with congeners. Finally, increased production of certain notes to the flying of both model types provides support for a “Better Safe than Sorry” strategy. When costs of alarm calling are low but costs of discriminating potentially serious threats may be extremely high, individuals should err on the side of caution, and alarm call to any potentially threatening stimulus.     

Predator-specific alarm calls in Campbell's monkeys, Cercopithecus campbelli

One of the most prominent behavioural features of many forest primates are the loud calls given by the adult males. Early observational studies repeatedly postulated that these calls function in intragroup spacing or intergroup avoidance. More recent field experiments with Diana monkeys (Cercopithecus diana) of Taï Forest, Ivory Coast, have clearly shown that loud male calls function as predator alarm calls because calls reliably (1) label different predator classes and (2) convey semantic information about the predator type present. Here, I test the alarm call hypothesis another primate, the Campbell's monkey (C. campbelli). Like Diana monkeys, male Campbell's monkeys produce conspicuous loud calls to crowned hawk eagles (Stephanoaetus coronatus) and leopards (Panthera pardus), two of their main predators. Playback experiments showed that monkeys responded to the predator category represented by the different playback stimuli, regardless of whether they consisted of (1) vocalisations of the actual predators (crowned hawk eagle shrieks or leopard growls), (2) alarm calls to crowned hawk eagles or leopards given by other male Campbell's monkeys or (3) alarm calls to crowned hawk eagles or leopards given by sympatric male Diana monkeys. These experiments provide further evidence that non-human primates have evolved the cognitive capacity to produce and respond to referential labels for external events.     

Predation affects alarm call usage in female Diana monkeys (Cercopithecus diana diana)

Diana monkeys produce acoustically distinct calls to a number of external events, including different types of predators. In a previous study, we found population-wide differences in male alarm call production in Taï Forest, Ivory Coast, and on Tiwai Island, Sierra Leone, mostly likely originating from differences in predator experience. In Taï Forest, leopards (Panthera pardus) are common but on Tiwai Island they have been absent for decades, while the predation pressure from crowned eagles (Stephanoaetus coronatus) has been similar. To further evaluate the impact of predator experience, we here analyse the vocal behaviour of female Diana monkeys in both habitats. Female Diana monkeys produce predator-specific alarm calls, alert calls and contact calls in response to predators, suggesting that their calls serve in a broader range of functions compared to males. Results showed that females produced the same call types at both sites, despite the differences in predator fauna. Regarding call usage, leopard alarm calls were extremely rare on Tiwai Island, but not in Taï Forest, whereas we found no differences in eagle alarm call production. When comparing response latencies, Tiwai females were slower to respond to both predators compared to Taï females. Finally, we found no habitat-specific acoustic differences in the alert and predator-specific alarm calls, but significant differences in frequency-based parameters of contact calls. Overall, our results suggest that ontogenetic experience can affect primate vocal behaviour in both usage and acoustic structure but that the way in which particular call types are affected may be closely linked to function.     

Ontogenetic differences in chemical alarm cue production determine antipredator responses and learned predator recognition

How individuals assess, respond and subsequently learn from alarm cues is crucial to their survival and future fitness. Yet this information is not constant through time; many individuals are exposed to different predators throughout their life as they outgrow some predators or move to habitats containing different predators. To maximise overall fitness, individuals should discriminate between different cues and respond and learn from only those that are relevant to their current ontogenetic stage. We tested whether juvenile spiny chromis, Acanthochromis polyacanthus, could distinguish between chemical alarm cues from conspecific donors of different ontogenetic stages and whether the cue ontogenetic stage of the cue donor affected the efficacy of learning about predators. Juveniles displayed a significant antipredator response when conditioned with juvenile chemical alarm cues paired with predator odour but failed to respond when conditioned with predator odour paired with either adult alarm cues or with saltwater. Subsequently, individuals only recognised the predator odour alone as a threat when conditioned with juvenile alarm cues. This demonstrates that prey may be highly specific in how they use information from conspecific alarm cues, selectively responding to and learning from only those cues that are relevant to their developmental stage.     

Threat-sensitive predator avoidance in damselfish-trumpetfish interactions

Summary Predatory threat can vary during a predator-prey interaction as an attack escalates or among predators at different times. A Threat-sensitivity hypothesis is presented which predicts that prey individuals will trade-off predator avoidance against other activities by altering their avoidance responses in a manner that reflects the magnitude of the predatory threat. This hypothesis was tested in the field by presenting prey (threespot damselfish, Stegastes planifrons) with models of foraging predators (Atlantic trumpetfish, Aulostomus maculatus). During a presentation, damselfish displayed progressively stronger avoidance as predator models were brought nearer; response waned rapidly once predator models passed overhead. Larger predator models and those oriented in a strike pose evoked stronger avoidance reactions than smaller and non-attacking models, intermediate responses were evoked by size and orientation combinations that were intermediate in threat, and habituation was more common to weakly-threatening presentations. Smaller damselfish showed stronger avoidance of models than did larger damselfish. Nonavoidance activities, such as feeding and territorial defense, were curtailed during presentations or were more common during weakly threatening presentations. Approaches to the models, equated with mobbing, were more common among large damselfish, again reflecting degrees of vulnerability among different size prey individuals. These initial results indicate that damselfish threatened by predators respond in a graded manner that reflects the degree of threat posed by the predator, in accordance with the Threat-sensitivity hypothesis.     

Anti-predator behavior of group-living Malagasy primates: mixed evidence for a referential alarm call system

Many mammals warn conspecifics with alarm calls about detected predators. These alarm calls are either functionally referential, urgency based, or they can have multiple functions, including predator deterrence. The taxonomic distribution of these alarm call systems is uneven, with primates providing the best-known examples for a functionally referential system and rodents most examples of an urgency-based system. Reports of different alarm call systems in lemurid primates prompted us to examine the anti-predator behavior of two additional lemur species. In an experimental field study we exposed adult redfronted lemurs (Eulemur fulvus rufus) and white sifakas (Propithecus verreauxi verreauxi) to playbacks of vocalizations of their main aerial and terrestrial predators, as well as to their own alarm calls given in response to the presentation of these predators. We scored the subjects' immediate behavioral responses, including alarm calls, from video recordings made during the first minute following a playback. We found that both species gave specific alarm calls only in response to raptor playbacks and the corresponding alarm calls, whereas calls given in response to carnivores and the corresponding alarm calls were also observed in other situations characterized by high arousal. Other behavioral responses, such as gaze and escape directions, corresponded to the hunting strategies of the two predator classes, suggesting that the corresponding vocalizations were categorized correctly. These two lemur species, which represent different families, have therefore independently evolved a mixed alarm call system, characterized by functionally referential calls for diurnal raptors, but not for carnivores. Electronic supplementary material to this paper can be obtained by using the Springer LINK server located at http://dx.doi.org/10.1007/s00265-001-0436-0 Electronic Publication     

Ontogenetic changes in alarm-call production and usage in meerkats (<Emphasis Type="Italic">Suricata suricatta</Emphasis>): adaptations or constraints?

In many species, individuals suffer major mortality in their first year because of predation. Behaviours that facilitate successful escape are therefore under strong selection, but anti-predator skills often emerge gradually during an individual’s early development. Using long-term data and acoustic recordings of alarm calls collected during natural predator encounters, we aimed to elucidate two largely unsolved issues in anti-predator ontogeny: (1) whether incorrect predator assignment is adaptively age-appropriate, given that vulnerability often changes during development, or whether age-related differences reflect true mistakes made by immature individuals; and (2) the extent to which the development of adult-like competence in alarm-call production and usage is simply a function of maturational processes or dependent upon experience. We found that young meerkats (Suricata suricatta) were less likely to give alarm calls than adults, but alarmed more in response to non-threatening species compared to adults. However, stimuli that pose a greater threat to young than adults did not elicit more calling from young; this argues against age-related changes in vulnerability as the sole explanation for developmental changes in calling. Young in small groups, who were more likely to watch out for predators, alarmed more than less vigilant young in larger groups. Moreover, despite similarities in acoustic structure between alarm call types, calls appeared in the repertoire at different rates, and those that were associated with frequently encountered predators were produced relatively early on. These results indicate that experience is a more plausible explanation for such developmental trajectories than maturation.