Reinforcement of male mate preferences in sympatric killifish species Lucania goodei and Lucania parva

Reinforcement occurs when reduced hybrid fitness leads to the evolution of a stronger prezygotic isolation. Populations sympatric with closely related species, where hybridization occurs, are predicted to have stronger mate preferences than allopatric populations. The reinforcement of male mate preference is thought to be rarer than the reinforcement of female preference, but this inference may be biased by the lack of studies on male preference. We tested male mate preferences from sympatric and allopatric populations of two closely related species of killifish: Lucania goodei and Lucania parva. We found that sympatric males had greater preferences for conspecific females than allopatric males. Furthermore, conspecific preferences in allopatric populations were weakest when these populations were geographically distant (>50?km) from those of heterospecifics. Our data suggest that reinforcement has contributed to male conspecific preference and speciation in Lucania.     

Sensory ecology in a changing world: salinity alters conspecific recognition in an amphidromous fish, Pseudomugil signifer

Group fission and fusion processes are driven by state dependence, risk and the availability of information from others. Yet the availability of information changes under different environmental conditions, thus aiding or inhibiting group formation and maintenance. Chemical cues provide information on the location of individuals and can act as a mechanism for individuals to group together, although they can be greatly affected by environmental conditions. Using a flow channel, we studied how one shoaling fish species, the Pacific blue-eye (Pseudomugil signifer), responds to conspecific chemical cues (CCCs) in different environmental conditions (salinities). This species lives in estuarine environments, ranging in salinity from fresh to fully marine. P. signifer responded to CCCs in freshwater but not in saltwater. Furthermore, P. signifer did not respond to saltwater with CCCs added from freshwater. It took significantly longer for fish in saltwater, than in freshwater, to locate and join a shoal when only CCCs from the shoal were present. Finally, fish formed more cohesive shoals in freshwater than in brackish or saltwater. These results suggest that these fish do not rely on chemical cues in saltwater to locate conspecific shoals. Furthermore, the reduced amounts of these cues in saltwater may inhibit the maintenance of tight shoal structures. We suggest that fish utilise different sensory modalities in fresh or saltwater in order to locate one another, or the social structure of these groups is fundamentally different between these two water types. The importance of this study in relation to understanding how animals utilise and change different sensory modalities in varying environmental conditions is discussed.     

Variation in male sailfin molly, preference for female size: does sympatry with sexual parasites, drive preference for smaller conspecifics?

Male sailfin mollies (Poecilia latipinna) are sexually parasitized by gynogenetic Amazon mollies (Poecilia formosa). In areas of sympatry, Amazon mollies are frequently larger than female sailfin mollies. In sympatry, selection may favor males that prefer smaller conspecific mates (avoid mismating with Amazon mollies), or selection may favor males that prefer larger conspecific mates (higher fecundity). To explore this potential species and mate-quality recognition conflict, we examined male preference variation across populations. Males from one sympatric population showed stabilizing preference functions, whereas in another sympatric population, males showed directional preference functions. Variation across sympatric populations may be related to the length of time of co-evolution with Amazon mollies. In the allopatric populations, we found flat preference functions. Variation in male preferences could have important ramifications for the maintenance of Amazon mollies, as well as for the evolution of female size.     

Association patterns of sailfin mollies (Poecilia latipinna): alternative hypotheses

Individuals may associate with each other due to a variety of selective forces, such as intra- and intersexual selection, and conspecific recognition. Previous studies have concluded that mate choice governs association behavior in polygynous species of fish. I examined whether mate choice underlies the preference for larger individuals by examining preference for association (time spent in proximity to a fish) not only between opposite-sex individuals but also between same-sex individuals of the live-bearing sailfin molly (Poecilia latipinna). Males and females from three size classes were tested with a large and a small object fish of the same and opposite sex. Females preferred to associate with larger over smaller males. Males also preferred to associate with larger over smaller females, as expected. The same female and male test fish also preferred to associate with larger over smaller fish of the same sex. Moreover, females demonstrated no significant difference in their strength of preference (large–small) when offered males or females. The same held true for males. When males and females were subsequently tested with one large male and one large female, females tended to prefer large males while males showed no significant preference for association based on sex. In another experiment, females were tested with a large female and a small male, and significantly preferred the former. These findings suggest that association patterns may have arisen under a variety of conditions, such as predation pressures, shoaling behavior, and associative preference behavior. The assumption that association behavior is a uniformly sufficient predictor of mate choice in fish needs to be re-examined for P. latipinna and other species. Received: 6 November 1998 / Received in revised form: 12 May 1999 / Accepted: 12 May 1999     

Interdemic variation in mixed-species association patterns: common diurnal primates of Kibale National Park, Uganda

We used interdemic variation in the tendency to form mixed-species groups to examine the costs and benefits of association among the primates of Kibale National Park, Uganda. A year-long survey of six sites revealed that the amount of time that the five common diurnal primates [red colobus (Procolobus tephrosceles), black-and-white colobus (Colobus guereza), redtail monkeys (Cercopithecus ascanius), blue monkeys (Cercopithecus mitis), and grey-cheeked mangabeys (Lophocebus albigena)] spent in mixed-species groups varied dramatically among sites. In many cases, the proportion of time that species associated was positively related to their densities. By using detailed behavioral observations of redtail monkeys and red colobus made over 4 years (2660 h) at four sites, we were able to reject the null hypothesis that associations occur by chance for only one of four sites. However, a correlative approach exploring the costs and benefits of association suggests that ecological variables do influence association patterns. We found that redtail monkeys and red colobus overlapped in diet (19.2% of their foraging effort) and traveled further when in mixed-species groups than when alone. Having demonstrated this, we examined the applicability of the ecological constraints model for predicting the proportion of the time spent in mixed-species groups based on food availability. For this analysis we concentrated on red colobus from the site with 35 months of observation and demonstrated that their tendency to be in mixed- species groups was related to food availability. We used two methods to examine if mixed-species associations function to decrease predation risk. First, chimpanzees are known to prey heavily on red colobus, but rarely kill other primates. The time red colobus spent in mixed-species groups was correlated to chimpanzee density, but it was not for the other monkey species, suggesting that mixed-species groups serve to decrease predation risk. Second, when red colobus groups contain more infants and are presumably at the greatest risk of predation, they form mixed-species groups most often. These results demonstrate that the costs and benefits of mixed-species associations vary dramatically over small spatial and temporal scales. If such variation is generally the case, then studies conducted at different locations or different times could easily highlight the importance of difference selective agents in favoring mixed-species associations. Received: 10 February 1999 / Received in revised form: 16 September 1999 / Accepted: 2 October 1999     

Preferences of six leaf beetle species among qualitatively different leaf age classes of three Salicaceous host species

Summary. As Salicaceous plants produce new leaves for a prolonged period of time, they expose a wide range of differentially aged leaves to herbivores during the growing season. In this work, I show that young leaves of three Salicaceous species, Populus tremula L., Salix phylicifolia L. and S. pentandra L., contain more nitrogen than conspecific old leaves. In P. tremula and S. pentandra young leaves also contained more low-molecular weight secondary compounds, phenolic glucosides. Leaves of S. phylicifolia did not contain phenolic glucosides in detectable amounts. Furthermore, in P. tremula and S. pentandra young leaves contained less polymeric digestability-reducing phenolics, condensed tannins, than old leaves. In S. phylicifolia, higher concentrations of condensed tannins were found in young leaves. In laboratory feeding trials with six leaf beetle species, young leaves of the studied plants were invariably preferred in all tested herbivore × host species combinations. In particular, it is remarkable that three leaf beetle species with known different overall relationships to phenolic glucosides equally preferred more glucoside-containing young S. pentandra leaves over conspecific old ones. Four beetle species were found to prefer young leaves of S. phylicifolia despite the higher content of condensed tannins in young leaves. These results indicate that the general preference of leaf beetles for young leaves of Salicaceous plants probably does not primarily result from variable distribution of secondary compounds. Apparently, the preference for young leaves is fundamentally due to variation in leaf nutritive traits, such as nitrogen content. Received 9 February 2001.     

Flexible habitat selection and interactive habitat segregation in the marine congeners <Emphasis Type="Italic">Idotea baltica</Emphasis> and <Emphasis Type="Italic">Idotea emarginata</Emphasis> (Crustacea,Isopoda)

Habitat segregation among competing species is widespread yet very little is know how this is achieved in practice. In a case study, we examined short-term effects of conspecific and congeneric density on habitat selection in two competing marine isopod species, Idotea emarginata and Idotea baltica. Under semi-natural conditions in large outdoor cylindrical tanks (4 m high; volume 5.5 m³), animal groups of different size and composition had the choice between a set of relevant habitat samples (surface-floating seaweed, the water column, seaweed on the bottom). Habitat selection in both I. baltica and I. emarginata proved to be largely independent of conspecific density (level of intraspecific competition). In single-species treatments, both species showed a similar and stable pattern of distribution, with a clear preference for seaweed on the bottom. In mixed-species treatments (MST), however, the species were largely separated by habitat. While the distribution of I. emarginata was completely unaffected by the mere presence of interspecific competitors, habitat selection of I. baltica changed notably when I. emarginata was present. The habitat use patterns observed in MST conformed to those realized in geographical areas where the two species overlap in distribution: I. emarginata is dominant among decaying seaweed on the sea floor, and I. baltica is the dominant species among surface-floating seaweed. Our findings suggest that habitat segregation between the two species is essentially interactive, resulting from rapid decision-making of I. baltica with respect to habitat selection. The underlying mechanism is discussed. I. emarginata is highly superior to I. baltica in interference competition and rapidly eliminates the latter from one-habitat systems which do not allow I. baltica to escape from this interaction. In more natural, heterogeneous environments, however, I. baltica seems to be able to coexist with the superior competitor due to its broader habitat niche, flexibility in habitat selection, and a behavioural disposition to avoid normally preferred habitats when these are occupied by I. emarginata.     

Asexuals looking for sex: conflict between species and mate-quality recognition in sailfin mollies (Poecilia latipinna)

When two closely related species are sympatric the process of species recognition (identifying conspecifics) and mate-quality recognition (increased fitness benefits) can yield a conflict when heterospecifics resemble high-quality conspecifics. Conflict in species versus mate-quality recognition may serve as a possible mechanism for the persistence of unisexual, gynogenetic Amazon mollies (Poecilia formosa). Amazon mollies require sperm from closely related species (e.g., sailfin mollies, P. latipinna) to start embryogenesis but inheritance is strictly maternal. When choosing mates, male sailfin mollies from populations sympatric with Amazon mollies may rely on traits indicating species identity rather than those indicating mate quality. Conversely, males from allopatric populations may rely more on traits indicating mate quality. Previous work has found that male sailfin mollies in sympatry exhibit a significantly greater mating preference for female sailfin mollies over Amazon mollies compared to males in allopatry. In addition, male sailfin mollies prefer to associate with and produce more sperm in the presence of larger conspecific females, which are more fecund. We hypothesized that male sailfin mollies experience a conflict in species recognition and mate-quality recognition in the presence of Amazon mollies that are relatively larger than female sailfin mollies. To test this hypothesis, we paired males from sympatric and allopatric populations with a larger Amazon molly and a smaller female sailfin molly. We scored the number of mating attempts that males directed to conspecific and heterospecific females. Males in most sympatric and allopatric populations demonstrate no clear preference for conspecifics. In addition, we found some evidence for a difference in mating preference between allopatric and sympatric populations with males from allopatry showing a greater heterospecific mate preference. These results indicate a conflict between species and mate-quality recognition. In sympatry this conflict may contribute to the persistence of gynogenetic Amazon mollies.     

Disentangling the effects of group size and density on shoaling decisions of three-spined sticklebacks (<Emphasis Type="Italic">Gasterosteus aculeatus</Emphasis>)

Many animals live in groups most of their life. One function of this behaviour is an increased predator protection whereas larger groups provide better protection than smaller ones. A causal explanation is that due to a higher number of shoal members the individual risk of being predated will decrease (“dilution effect”). Additionally, shoaling leads to increased predator confusion. This “confusion effect” can be strengthened by an increased group density, which often correlates with group size. Many studies found that individuals prefer the larger of two groups. However, whether this preference is due to a larger group size or because of an increased density of the larger group remained unclear. To disentangle these factors we gave three-spined sticklebacks (Gasterosteus aculeatus) the choice between shoals of (1) different group size and density, (2) different group size, but equal density and (3) equal group sizes, but different densities. As expected, test fish preferred the larger and denser shoal over the smaller, less dense one. This preference was lost when shoal size differed but density was kept constant. When shoal size was equal but density differed, test fish preferred the less dense shoal. However, this was only the case when test fish chose between two relatively dense shoals. On the other hand, when overall density was low, test fish did not discriminate between shoals of different densities. This result may be explained in terms of predator avoidance. The results show that shoaling preferences might not always be influenced by a higher number of group members but also by the density and cohesiveness of the respective groups. An erratum to this article can be found at     

Space use in lions, <Emphasis Type="Italic">Panthera leo</Emphasis>, in the Selous Game Reserve: social and ecological factors

Habitat qualities, such as food supply or access to refuges, often influence home-range size. Furthermore, such qualitative differences usually lead to conspecific competition over space, which can be an important factor in determining the distribution of individuals within populations. In carnivores, patterns of resource dispersion are hypothesized to determine home range-size and group size. But in contests over space (or other resources), larger groups usually dominate smaller ones, and group size should therefore also affect home-range size. Here I describe the space use of lions, Panthera leo, in the Selous Game Reserve, Tanzania, and ask whether space use is related to pride size, habitat, or relatedness. Home ranges varied in size, but size showed no correlation to number of adult females in the pride or to habitat type. Lions exhibited a significant preference for riverine and short-grass habitat, and a significant avoidance of acacia woodland. Habitat preference ratios largely reflected prey availability in each habitat. Outer areas, as well as core areas of home ranges, were often used by two or more prides. Overlaps showed no correlation to relatedness among prides or habitat type. Thus, whereas home-range sizes and overlaps were determined by factors that could not be revealed by demographic factors or analyses of habitat composition or genetic structure, lion space use within each home range seemed driven mostly by prey availability, which mainly varies with habitat type.     

Shoaling fish can size-assort by chemical cues alone

Animals that form groups are typically assorted by phenotype. For example, fish shoals are notably composed of closely size-matched individuals, yet the sensory mechanisms that promote this behaviour have not been fully determined. Here, we show that two freshwater shoaling fish species, three-spined stickleback and banded killifish, have a greater preference for the chemical cues of conspecifics that are the same size as themselves than for those of larger or smaller conspecifics. We suggest that this ability to determine their own size relative to conspecifics may be based on chemical self-referencing. This provides a novel insight to the mechanisms underlying a widespread phenomenon in social behaviour, and provides further evidence of the crucial role played by chemical cues in structuring the interactions of fishes.     

Aboveground and belowground effects of single-tree removals in New Zealand rain forest

There has been considerable recent interest in how human-induced species loss affects community and ecosystem properties. These effects are particularly apparent when a commercially valuable species is harvested from an ecosystem, such as occurs through single-tree harvesting or selective logging of desired timber species in natural forests. In New Zealand mixed-species rain forests, single-tree harvesting of the emergent gymnosperm Dacrydium cupressinum, or rimu, has been widespread. This harvesting has been contentious in part because of possible ecological impacts of Dacrydium removal on the remainder of the forest, but many of these effects remain unexplored. We identified an area where an unintended 40-year "removal experiment" had been set up that involved selective extraction of individual Dacrydium trees. We measured aboveground and belowground variables at set distances from both individual live trees and stumps of trees harvested 40 years ago. Live trees had effects both above and below ground by affecting diversity and cover of several components of the vegetation (usually negatively), promoting soil C sequestration, enhancing ratios of soil C:P and N:P, and affecting community structure of soil microflora. These effects extended to 8 m from the tree base and were likely caused by poor-quality litter and humus produced by the trees. Measurements for the stumps revealed strong legacy effects of prior presence of trees on some properties (e.g., cover by understory herbs and ferns, soil C sequestration, soil C:P and N:P ratios), but not others (e.g., soil fungal biomass, soil N concentration). These results suggest that the legacy of prior presence of Dacrydium may remain for several decades or centuries, and certainly well over 40 years. They also demonstrate that, while large Dacrydium individuals (and their removal) may have important effects in their immediate proximity, within a forest, these effects should only be important in localized patches containing high densities of large trees. Finally, this study emphasizes that deliberate extraction of a particular tree species from a forest can exert influences both above and below ground if the removed species has a different functional role than that of the other plant species present.     

Trichinella spiralis infection in voles alters female odor preference but not partner preference

Females may choose mates based on secondary sex traits that reflect disease resistance. Accordingly, females should be able to distinguish between unparasitized and parasitized males, and should prefer to mate with unparasitized individuals. Mate and odor preferences for uninfected males or males infected with the nematode, Trichinella spiralis, were examined among prairie voles (Microtus ochrogaster) and meadow voles (M. pennsylvanicus). In a 15-min odor preference test, only female meadow voles distinguished between bedding from parasitized and unparasitized conspecific males, and preferred to spend time with bedding from unparasitized males. Although T. spiralis infection influenced odor preference in female meadow voles, there was no effect of infection status on mate preference among either species. Testosterone and corticosterone concentrations were not different between parasitized and unparasitized males. However, among prairie voles, males that spent an increased amount of time with females during the mate preference test had elevated testosterone concentrations. Taken together, these data suggest that (1) female meadow voles can discriminate between unparasitized and parasitized males, (2) the effects of infection on steroid hormone concentrations may be masked by the effects of social interactions, and (3) parasites may represent a selective constraint on partner preference in voles; however, the life cycle of parasites may influence female preference and should be considered in studies of female preference. Received: 23 April 1998 / Accepted after revision: 25 October 1998     

Reduction of the association preference for conspecifics in cave-dwelling Atlantic mollies, Poecilia mexicana

Cave animals are widely recognised as model organisms to study regressive evolutionary processes like the reduction of eyes. In this paper, we report on the regressive evolution of species discrimination in the cave molly, Poecilia mexicana, which, unlike other cave fishes, still has functional eyes. This allowed us to examine the response to both visual and non-visual cues involved in species discrimination. When surface-dwelling females were given a chance to associate with either a conspecific or a swordtail (Xiphophorus hellerii) female, they strongly preferred the conspecific female both when multiple cues and when solely visual cues were available to the female. No association preference was observed when only non-visual cues were provided. In contrast, cave-dwelling females showed no preference under all testing conditions, suggesting that species recognition mechanisms have been reduced. We discuss the role of species discrimination in relation to habitat differences.     

Social discrimination tuning in ants: template formation and chemical similarity

To investigate the role of template plasticity in shaping nest-mate recognition processes in ants, we constructed experimental mixed-species groups of Manica rubida with either Myrmica rubra, Tetramorium bicarinatum or Formica selysi. Selecting Ma. rubida as the focal species, we observed the behaviour within mixed-species groups and the transfer rates of cuticular hydrocarbons (CHC) onto the focal ants, and we also tested the aggression of the focal species reared either alone or in association with each of the three different species. We show that Ma. rubida workers were always amicable towards their mixed group members, as towards members of the respective parental colonies, irrespective of the associated species. They did, however, express different levels of aggression towards single-species groups of the other species tested, depending on the species with which they were reared. The study suggests that similarity in CHC profiles in two species leads to a narrow template in mixed groups, while dissimilarity is followed by lower levels of aggression (a broader template), at least against species with similar CHC compound compositions (i.e. both a broader template in the focal ants and familiarity with the compound groups of the tested individuals operate together). This refutes the hypothesis that ants reared in mixed-species groups are systematically more tolerant. It also demonstrates that heterospecific information is not treated equally during development. We suggest that post-imaginal learning, template reforming and decision making are more precisely tuned when the two species' chemical complexes are similar.     

Egg cannibalism by sticklebacks: spite or selfishness?

Summary Although it is generally agreed that humans can be spiteful, there are few if any, unambiguous examples of spite by non-human animals. Data are presented suggesting that female threespine sticklebacks show spiteful behaviour. In the field, they seek out conspecific eggs to attack while largely ignoring those of a closely-related sympatric species, the blackspotted stickleback. This occurs despite the fact that the latter's nests are more abundant and less well protected. In the laboratory, female threespine sticklebacks attack the eggs of conspecifics more than those of blackspotted sticklebacks, those of sympatric conspecific females more than those of allopatric females, and older eggs more than younger ones. Because there was no evidence of greater energetic or nutritional advantages from eating conspecific rather than heterospecific eggs, or older eggs rather than younger ones, threespine sticklebacks may be spiteful. Alternative proximate and evolutionary hypotheses to explain this discriminant egg-eating are discussed.     

Lack of innate preference for morph and species identity in mate-searching <Emphasis Type="Italic">Enallagma</Emphasis> damselflies

Insect mate recognition is often viewed as stereotypic, innate, and species-specific. However, male damselflies can learn to identify female-specific color morphs as potential mates. A suite of male mimicry hypotheses assume that heteromorphic females, which differ from males in color pattern, are more easily recognized as “female” and thus lack the inherent, anti-harassment advantage that the more male-like signal provides for andromorphs. Using two measures of male preference, we investigated whether naïve males have a preexisting sensory bias for a given morph color in Enallagma civile, a species that appeared to exhibit extreme plasticity in morph expression across generations within a breeding season. E. civile males raised in the absence of females exhibited no preference for either morph, whereas males raised with one female type exhibited a learned sensory bias for that morph. Male Enallagma also lacked a bias toward conspecific females over a congeneric sister species. In a naturally naïve population of Enallagma ebrium, males reacted sexually to both morphs of Enallagma hageni as often as they did to conspecific females, whose thoracic spectra were nearly identical with those of E. hageni. Moreover, despite the similar thoracic spectra of males and andromorphs, both of which reflected UV, males rarely reacted sexually to other males. Our results falsified implicit assumptions of male mimicry hypotheses, supported learned mate recognition, and suggested a scenario for speciation via sexual conflict.     

Spatial and behavioral interactions between a native and introduced salamander species

Behavioral interactions with native species may influence the invasiveness of introduced species. The salamanders Plethodon glutinosus and P. jordani in the eastern United States share many life history traits and demonstrate complex interspecific interactions that range geographically from competitive exclusion to sympatry. P. jordani was introduced to Mountain Lake Biological Station, Virginia, USA, between the years 1935 and 1945. We tested whether competition for space may influence the invasion of P. jordani into native P. glutinosus habitat by utilizing data from natural distributions, a field experiment, and controlled laboratory experiments. No environmental variables differed where P. glutinosus and P. jordani were collected in the field at the site of P. jordani introduction. In the field experiment, P. glutinosus was more fully exposed during foraging bouts in cages shared with heterospecifics as opposed to ones shared with conspecific salamanders. Condition (mass relative to body length) of salamanders at the end of the 3 months did not differ between conspecific and heterospecific treatments. In the laboratory, P. glutinosus most often attained the single burrow in the arena, but residency status had no effect. Species cohabited the burrow 50% of the time. Pair-wise encounters in the laboratory indicated that both species spend less than 20% of the time in aggressive behaviors as juveniles. Adults showed no behavior interpreted by us as aggression during pair-wise encounters. Received: 19 December 1999 / Accepted: 18 March 2000     

Factors affecting vigilance within wild mixed-species troops of saddleback (<Emphasis Type="Italic">Saguinus fuscicollis</Emphasis>) and moustached tamarins (<Emphasis Type="Italic">S. mystax</Emphasis>)

We examine vigilance within a mixed-species troop of saddleback (Saguinus fuscicollis) and moustached (S. mystax) tamarins over a complete year. Saddleback tamarins were consistently more vigilant than moustached tamarins. This may be linked to their preference for lower strata. In accordance with previous studies of other primates, vigilant tamarins of both species were significantly further away from their nearest neighbours, and were also at lower heights in the forest than non-vigilant individuals. There was no observed sex difference in the amount of time spent vigilant. In terms of modes of scanning, the saddleback tamarins looked up significantly more frequently than the moustached tamarins, whereas there was no difference between the species in the frequency of side sweeps. There were no differences between the sexes in the frequencies of either type of vigilant behaviour. The proportion of time spent vigilant was higher than average immediately prior to entering a sleeping site for saddleback tamarins, but not for moustached tamarins. Both species were more vigilant immediately after exiting a sleeping site than at other times of the day. There was significant variation in the amount of time devoted to vigilance over the course of the year. These findings are discussed with respect to the social structure, ecology and main predator threats facing these species.Communicated by D. Watts     

Avian Community Response to Lowland Tropical Rainforest Isolation: 40 Years of Change at La Selva Biological Station, Costa Rica

Abstract:  Since 1960, most of the forest surrounding the La Selva Biological Station, an intensively studied tropical research facility in Costa Rica, has been converted to agricultural uses. We used quantitative censuses and analysis of previously published categorical abundances to assess changes in the bird community, and we evaluated potential causes of species-specific changes by assessing their association with habitat, diet, participation in mixed-species flocks, and nest type. Approximately the same percentage of species increased as decreased in abundance from 1960 to 1999 (10–20% of all species, depending on method of assessment). Diet was the single most important trait associated with declining species. At least 50% of the species that declined have insectivorous diets. Use of forest habitat and participation in mixed-species flocks were also significant factors associated with declines, but nest type was unrelated to change in abundance. The species that increased in abundance tended to occur in open habitats and have omnivorous diets. These results reinforce the importance of several population risk factors associated with tropical understory insectivory and mixed-species flocking: patchy spatial distribution, low population density, large home range, and dietary specialization. La Selva's protected area (1611 ha), despite a forested connection on one boundary with a higher elevation national park, is apparently too small to maintain at least one major guild (understory insectivores). This first quantitative assessment of bird community change at La Selva highlights the need to intensify study of the mechanisms and consequences of biological diversity change in tropical forest fragments.