Search

Social monogamy in vertebrates often involves the joint defence of an all-purpose territory by pair members. Although both sexes presumably benefit from holding a territory, sex-related variation is often observed in territory defence at the interspecific level, particularly in birds. Most studies of territorial defence in monogamous birds, however, have been conducted on songbirds or on bird species living in temperate areas. In contrast, data on nonoscine tropical bird species remain scarce. We experimentally studied territorial defence in 20 pairs of Zenaida doves, a tropical and territorial monogamous bird species that maintains year-round pair bonds. Overall, males responded to intrusions more frequently than females. However, sex of the intruder had no effect on the reactivity of male or female residents. The use of aggressive behaviours of various intensities did not differ between males and females, irrespective of the sex of intruder. Although the percentage of male responses was correlated significantly and negatively with that of the females, the observed percentage of joint defence was significantly lower than would have been expected if pair members were reacting independently of each other. Finally, variation in reactivity of females to intruders was positively related to variation in intrusion pressure between trials, whereas no such trend was observed in males. We discuss our results in relation to the mating system of Zenaida doves and previous studies of territorial defence in birds.

We quantified vigilance during feeding in the Zenaida Dove Zenaida aurita, a tropical species with stable pair-bonds and year-round territoriality. Both males and females decreased the proportion of time spent vigilant by 30% when feeding with their partner compared with when feeding alone. This reduction was achieved through increasing the length of inter-scan duration, while scan duration remained constant. No evidence was found for coordination of vigilance between pair members. The equal investment in vigilance by male and female Zenaida Doves might be related to the mutual benefits of long-term pair-bonding.

Animals may acquire information about predatory threats via direct cues, clues from the predator, or indirect cues from predator-elicited alarm signals from conspecifics. Although direct detection of a predator is assumed to be a better indicator of risk, only three previous studies have directly contrasted the relative information content of conspecific alarm calls to predator sounds. Studied species typically reacted equally or stronger to conspecific alarm calls than to predator sounds, suggesting that an indirect cue may be a more reliable indicator of risk. We studied zenaida doves’, Zenaida aurita, responsiveness to predator sounds and conspecific alarm signals. When flushed, zenaida doves produced voluntary mechanical wing whistles, which we predicted to be alarm signals. Observing responses to wing flap playbacks with and without whistles, we found that doves increased vigilance significantly more to wing whistles than to wing flaps without whistles or control playbacks. These results indicate that conspecifics interpret wing whistles as alarm signals. We then conducted another playback experiment, which demonstrated that red-tailed hawk, Buteo jamaicensis, playbacks elicited higher levels of dove vigilance than conspecific wing whistles. Contrary to other species, zenaida doves seemingly consider predator vocalizations more informative than conspecific alarm signals. Therefore, the reliability of acoustic signals associated with predation risk may strongly influence prey responsiveness to direct and indirect cues.

In Zenaida doves, Zenaida aurita, of Barbados, previous work has shown that social learning is associated with foraging ecology in the field: in captive experiments, group-foraging birds learn more readily from a conspecific tutor, while territorial birds learn from the heterospecific they most often feed with in mixed-species aggregations, the Carib grackle, Quiscalus lugubris. This study examines foraging ecology and social learning in a dove population that experiences both territorial defence and occasional group feeding, Brandon's Beach. In part I, we document the dual foraging pattern seen in this population, which lives only a few hundred metres from the group-foraging harbour population studied in previous work. In part II, we show that doves from Brandon's Beach, consistent with their dual foraging experience, learn as readily from a conspecific as they do from a heterospecific tutor; control doves from the adjacent harbour site learn primarily from a conspecific tutor. Field sightings of banded individuals caught at the harbour and Brandon's show strong site fidelity, but sufficient movement between areas (4%) to make reproductive divergence between the two neighbouring populations highly unlikely, suggesting that social learning differences between them are non-genetic.

13 pages; International audience; Although variation in plumage coloration is known to occur both between and within sexes, its study remains limited to a few bird families. The Zenaida dove Zenaida aurita is a socially monogamous tropical columbid bird species, characterized by an overall cinnamon-brownish plumage and structural colorations on the head and neck. The species has been described as sexually dichromatic for plumage, although color differences between males and females are not obvious in the field. We investigated variation in the presumably melanin-based color of the crown, mantle, breast, and belly, in the iridescent dark-blue streaks on the head, and in the symmetric iridescent patches on the neck, over the whole spectrum visible to birds. Further, unlike most previous studies, we assessed covariation between plumage color and phenotypic traits in both males and females in relation to the putative signaling function of ornaments. Zenaida doves appeared to be slightly sexually dichromatic for the hue of pigment-based colored areas, with males being on average more reddish than females. However, this difference was not discernible when considering the avian visual system. Conversely, although the reflectance spectra of iridescent plumage did not significantly differ between sexes in brightness, chroma or spectral position of the peaks, color discrimination analyses showed that individuals should be able to perceive between- or within-sex differences in the color of the iridescent patch. In addition, several color parameters of brown and iridescent feathers were significantly related to territorial status, body condition, wing chord, and, albeit weakly, to individual multilocus heterozygosity. Overall, our results thus suggest that plumage color might be a reliable signal of quality in individuals of both sexes in this species. Further studies are needed to test the potential implication of plumage coloration in mate choice and mating patterns in the Zenaida dove.

Aggregating with heterospecifics may be particularly beneficial for a species that is able to exploit the antipredator behaviour of another. Territorial zenaida doves vigorously exclude conspecific intruders from their territory, but forage with, and acquire novel foraging techniques from, carib grackles. Given that doves associate with no other conspecific than their mate and they have no vocal alarm signals of their own, they might benefit from attending to the conspicuous alarm calls of carib grackles. In the present study, we found that zenaida doves suppressed foraging both in response to a model predator and in response to the sound of grackle alarm vocalizations. Although doves' responses to the predator model also involved moving away from the immediate vicinity, their responses to grackle alarm vocalizations consisted of remaining alert and tail flicking. Together, these results strongly suggest that doves attend to the antipredator behaviour of carib grackles. These findings extend earlier work demonstrating that doves obtain foraging benefits from their association with grackles, to show that they may also obtain predator avoidance benefits.

9 pages; International audience; The relative influence of genetic and phenotypic quality on pairing status and mating patterns in socially monogamous species remains poorly documented. We studied social status and pairing patterns in relation to genetic similarity and multilocus heterozygosity (MLH) estimates from 11 microsatellite markers, and both tarsus length and wing chord (as a measure of competitive ability in territorial defence) in a socially monogamous tropical bird species where individuals defend territories year-round, alone or in pairs, the Zenaida dove, Zenaida aurita. Tarsus length and wing chord did not differ between unpaired territorial birds and paired ones in either sex, whereas paired females, but not paired males, tended to be more heterozygous than unpaired ones. Among 84 pairs, we found no evidence for assortative mating for tarsus length, wing chord, MLH or genetic similarity. However, within pairs, male wing chord was positively related to female MLH and female tarsus length was positively related to male MLH, with no evidence for local effects, suggesting assortative mating by individual quality. Although the observed pattern of mating in Zenaida doves may be the product of mutual mate choice, further assessment of this hypothesis requires direct investigation of both mating preference in each sex and lifetime reproductive success in relation to body size and MLH.

The spatial and temporal clumping of food influence an animal’s aggressiveness during competition. No studies, however, have investigated the effects of the temporal predictability of food and few studies have tested for interactions between the effects of two components of resource distribution on the rates of competitive aggression. We simultaneously manipulated the temporal predictability and the spatial clumping of food to test whether aggression increases as food becomes more predictable in time and more clumped in space. We tested these predictions using wild Zenaida doves (Zenaida aurita) in Barbados because previous work showed marked differences in social behavior between two populations, apparently related to differences in the distribution of food in space and time. There was a significant interaction between the effects of the temporal predictability and spatial clumping of food. As predicted, the rate of aggression increased as the temporal predictability of food increased, but only significantly in the spatially clumped condition. Similarly, as predicted, aggression increased as the spatial clumping of food increased, but only significantly in the temporally predictable condition. In addition, the per capita rate of aggression peaked at intermediate competitor densities in the spatially clumped condition. Differences in rates of aggression observed during experimental manipulations and between the two populations during baseline observations were generally consistent with predictions of resource defense theory. Key words: aggression, economic defendability, feeding competition, spatial clumping, temporal predictability, Zenaida aurita. [Behav Ecol 12:490–495 (2001)]

Learning diAerences predicted from ecological variables can be confounded with diAerences in wariness of novel stimuli (neophobia). Previous work on feral pigeons (Columba livia), as well as on group-feeding and territorial zenaida doves (Zenaida aurita), reported individual and social learning diAerences predicted from socialforagingmode.Inthepresentstudy,weshowthatspeedoflearningaforaging task covaries with neophobia and latency to feed from a familiar dish in the three types of columbids. Pigeons were much faster than either territorial or groupfeeding zenaida doves on all tests conducted in captivity, but showed unexpectedly strongneophobiainsomeurbanflocksduringfieldtests.Humanproximitystrongly aAected performance in group-feeding doves both in thefield and in captivity. They were slightly faster at learning than their territorial conspecifics in cage tests. In multiple regressions, species identity, but not social foraging mode, significantly predicted individual variation in learning, as did individual variation in neophobia. Wariness of novel stimuli and species diAerences associated with artificial selection appear to be more important than foraging mode and wariness of humans in accounting for learning diAerences between these columbids.

We observed Grey Kingbirds (Tyrannus dominicensis) from late February to May 2007 stealing food items from the bills of Carib Grackles (Quiscalus lugubris). This behavior occurred at two baited walk-in bird traps on the grounds of Bellairs Research Institute of McGill University in St James, Barbados. Grey Kingbirds were not seen entering traps, but were regularly observed in tree branches near traps, often chasing Carib Grackles and Zenaida Doves (Zenaida aurita) as they exited the traps with food. We describe six instances of kleptoparasitism by Grey Kingbirds from Carib Grackles. To our knowledge, this is the first report of kleptoparasitism for this species.

Abstract In Barbados, Carib grackles (Quiscalus lugubris) forage in opportunistic aggregations that include territorial Zenaida doves (Zenaida aurita) and flocks of conspecifics. In searching for and handling food, grackles use complex beak movements and modulate priority of access with a mixture of postural signals and intra- and interspecific scramble (unaggressive) competition. We show in two experiments that wild-caught grackles learn as readily from a Zenaida dove as they do from a conspecific tutor, whether tutors use similar or different food-finding techniques. Grackles also imitate the technical variant that the hetero- and conspecific tutors were shaped to demonstrate: those who observed a Zenaida dove used the dove's closed beak pecking technique, while those who observed a conspecific used the grackle's open beak pulling, probing or prying. Our findings suggest that imitation, like other forms of social learning, is strongly influenced by a species' foraging ecology. In particular, the ability to imitate novel motor skills should be favored in opportunistic species which exhibit scramble competition and which use complex searching and handling techniques to forage on embedded foods.

Group-living is often thought to be associated with social learning. Comparative tests are useful in evaluating the link between ecological variables and learning specializations, but controls are required to guard against possible confounding variables. In this paper, we test the association between group-living and social learning by comparing two opportunistic, urbanized, Columbids in a set of experiments involving shaping and individual learning controls. In part 1, we provide quantitative field data on foraging group size for the two species, the feral pigeon (Columba livia) studied in central Montréal and the Zenaida dove (Zenaida aurita) studied in coastal Barbados. The data confirm anecdotal reports that contrast the gregarious social organization of C. livia with the territorial organization of coastal Z. aurita. In part 2, we test pigeons and Zenaida doves on two food-finding tasks and show (1) that feral pigeons are better than Zenaida doves at solving all variants of the tasks presented, whether individually or socially learned, and (2) that once general species differences are taken into account, territorial Zenaida doves are not less efficient than gregarious feral pigeons at using food-finding information provided by a conspecific tutor. The results do not support the association between group-living and social learning and show that procedures like shaping and control variants of the task may help interpret interspecific differences in learning.

Abstract Adaptive views of learning predict that natural selection should lead to differences in specialized learning abilities between animals that face different ecological pressures. Group-living is thought to favour social learning, but previous comparative work suggests that differences between gregarious feral pigeons (Columba livia) and territorial Zenaida doves (Zenaida aurita) exceed the specialized effect on social tasks predicted by the adaptive hypothesis. In this paper, we show that group-foraging Zenaida doves from Barbados learn an individual shaping task more quickly than territorial Zenaida doves from a site 9 km away. These results suggest that the scramble competition associated with group-foraging favours several types of leaming, both social and non-social, and that its effects are more wide-ranging than previously thought. Since genetic isolation between Zenaida dove populations is highly unlikely, the results also suggest that differences in foraging ecology may lead to different learned responses to local reward contingencies as well as natural selection for different genotypes affecting learning. In some cases, the standard comparative prediction of ecologically-correlated learning differences may therefore not distinguish between adaptive specialization and general process theories.

10 pages; International audience; Understanding how fitness is related to genetic variation is of crucial importance in both evolutionary ecology and conservation biology. We report a study of heterozygosity-fitness correlations in a wild, noninbred population of Zenaida Doves, Zenaida aurita, based on a sample comprising 489 individuals (382 adults and 107 juveniles) typed at 13 microsatellite loci, resulting in a data set comprising 5793 genotypes. In both adults and juveniles, and irrespective of sex, no evidence was found for an effect of either multilocus or single-locus heterozygosity on traits potentially related to fitness such as foraging tactic, competitive ability, and fluctuating asymmetry. In contrast, a significant negative correlation between body condition and multilocus heterozygosity, indicative of outbreeding depression, was found in juveniles, whereas no such trend was observed in adults. However, the frequency distribution of heterozygosity did not differ between the two age classes, suggesting compensatory growth by heterozygous juveniles. We discuss our results in relation to some practical limitations associated with studies of heterozygosity-fitness correlations, and suggest that tropical bird species with allopatric divergence between island populations may provide a good biological model for the detection of outbreeding depression.

9 pages; International audience; Discriminant function analysis (DFA) based on morphological measurements is a quick, inexpensive, and efficient method for sex determination in field studies on cryptically monomorphic bird species. However, behind the apparent standardization and relative simplicity of DFA lie subtle differences and pitfalls that have been neglected in some studies. Most of these concerns directly affect assessment of the discriminant performance, a parameter of crucial importance in practice because it provides a measure of the quality of an equation that may be used in later field studies. Using results from 141 published studies and simulations based on a large data set collected on adult Zenaida Doves (Zenaida aurita), we assessed the effects of sexual dimorphism, sample size, and validation methods on discrimination rates. We compared the three most common methods used to estimate the proportion of correctly classified males and females by DFA: resubstitution, jackknife, or sample splitting. Results from simulations indicate that these procedures may lead to opposite conclusions, especially when the sample size is small. In particular, the resubstitution techniques appear to be overoptimistic, and we therefore recommend that DFA accuracy be estimated by the jackknife cross-validation procedure. In addition, we show that most previous studies failed to present DFA accuracy with 95% confidence intervals, which hampers comparisons among studies. Finally, our results suggest that large sample sizes should be preferred over repeated measurements of the same individuals, because random measurement error is likely to have only a weak effect on the accuracy of the discriminant rate.

11 pages; International audience; The term “resource polymorphism” refers to the existence of alternative phenotypes in relation to resource use, as a result of disruptive selection. Evidence for resource polymorphism is widespread in fish but remains scarce in birds. Although Zenaida Doves (Zenaida aurita) usually defend year-round territories, doves on Barbados can also be observed foraging at seed-storage sites in large flocks with little, if any, inter-individual aggression. On the basis of morphological variation, it has been suggested (Sol et al. 2005) that this represents a case of resource polymorphism, primarily driven by competition for territories. Using new data, we revisited the evidence for resource polymorphism in Zenaida Doves on Barbados. In particular, we added replicates in time and space for territorial and flock-foraging birds and used molecular markers to assign sex to adults and juveniles. In addition, we used microsatellite markers to assess potential genetic differentiation between flock-feeding and territorial doves. Our results confirm previous observations that territorial adults were larger than flock-feeding ones, whereas the reverse was observed in juveniles. Contrary to previous observations, we found a significant excess of females among flock-feeding adults, whereas the sex ratio was balanced in territorial adults and in juveniles. In addition, we observed no significant difference in body condition and no genetic differentiation between territorial and flock-feeding individuals. Overall, our data question the existence of resource polymorphism in Zenaida Doves in Barbados. We suggest alternative, more parsimonious explanations, based on age- and sex-related differences in the relative benefits of holding a territory.

7 pages; International audience; Animals' ability to adjust their behaviour when environmental conditions change can increase their likelihood of survival. Although such behavioural flexibility is regularly observed in the field, it has proven difficult to systematically quantify and predict inter-individual differences in free-living animals. We presented 24 Zenaida doves (Zenaida aurita) on 12 territories with two learning tests in their natural habitat in Barbados. The dove pairs showed high site fidelity and territoriality, allowing us to test individuals repeatedly while accounting for the effects of territorial chases and pair bonds on our learning measures. We used a foraging apparatus that enabled Zenaida doves to access seed, yet excluded other species, and measured doves' performance on colour discrimination and reversal learning tests. We found that (1) doves on all 12 territories passed the two tests; (2) mates within a pair were consistently solvers or scroungers; (3) sex, body condition and territorial chases did not consistently affect learning rates; (4) tameness was a significant negative predictor of learning to feed from the foraging apparatus and (5) scrounging within pairs seemed to facilitate learning. Our study presents a method to quantify intraspecific differences in behavioural flexibility in the field and relate these to individuals' physical and social traits.

This study compared the direction of social learning in 2 populations of Barbados Zenaida doves (Zenaida aurita). One population (St. James) is territorial; it competes aggressively with conspecifics but scramble competes with heterospecifics. The other population (Deep Water Harbour) forages in large homospecific flocks. Field observations were conducted to quantify intraspecific and interspecific patterns of foraging association and aggression. Wild-caught doves from both areas were then tested on novel foraging tasks demonstrated by either a conspecific or a heterospecific tutor. In all experiments, St. James doves learned more readily from the heterospecific tutor (Carib grackle -Quiscalus lugubris-), whereas Deep Water Harbour doves learned more readily from the conspecific tutor. The type of competitive feeding interaction in the field (i.e., scramble vs. interference) appears to better predict the pattern of social learning in an experiment than does species identity.

I collected line transect data and capture-recapture data to estimate basic demographic parameters of Zenaida doves (Zenaida aurita) in farmlands and second-growth forests in eastcentral Puerto Rico during 1990-92. Zenaida doves ar widely hunted, and data on demographic parameters are needed for population management and monitoring. Mean survival rate and mean capture rate per sampling period of hatching-year (RY; o i = 0.73, p i = 0.07) and after-hatching-year (AHY; o i = 0.75, p i = 0.06) doves did not differ (Ps > 0.10). The standard Jolly-Seber model with time-specific survival and capture rates fit the capture-recapture data (Model A in JOLLY; P > 0. 10). There was no difference between the survival and capture rates of newly versus previously captured HY and AHY birds (Ps > 0.10). Survival and capture rates varied by sampling period (Ps < 0.01 ). For AHY and HY birds combined, mean survival rate per sampling period was 0.84, and mean capture rate was 0.05. Mean population size was 1,451, and mean recruitment rate (births + immigrants was 591/sampling period. Density of Zenaida doves was 0.97-1.78/ha during 1991-92. Density of nests was 2.44/ ha during January-December 1991. 1.69 during April-June 1991, and 1.44 during April-June 1992. Line transect surveys of Zenaida doves and nests were positively correlated (P < 0.01). The number of Zenaida doves and nests was positively correlated with capture-recapture estimates of population size (P

We surveyed several offshore islands around Culebra for nesting Zenaida Doves and seabirds to determine if the doves preferred islands with nesting seabirds, and whether or not the doves selected the same types of sites on all islands. Doves nested on eight of the islands with seabirds, and did not nest on four islands that were devoid of nesting seabirds. Nest-site characteristics differed among the islands, although there were no inter-island differences with respect to nest size, slope at the nest, and percent of the nest visible from above and from 1 m. Thus slope and nest visibility seem to be significant habitat cues for nesting Zenaida Doves.

We examined nest site selection of the zenaida dove (Zenaida aurita) on two islands on the Culebra National Wildlife Refuge, Puerto Rico, to determine the adaptations of the doves for nesting within seabird colonies. Whereas bridled (Sterna anaethetus) and roseate (Sterna dougallii) terns used particular parts of islands, zenaida doves nested in all available habitats. Nonetheless, most dove nests were within the parts of the islands where terns nested. On both islands, doves nested on flatter places near taller rocks, on less visible places farther from clearings, and closer to tern nests than were the random points. On Cayo del Agua doves nested mainly in rock crevices, whereas on Cayo Raton they nested mainly on the ground under vegetation. The doves nested closer to the terns than expected, suggesting that they derive antipredator advantages from nesting near the terns. SELECTING A BREEDING SITE involves choosing a general habitat, such as marsh, field, or forest, then a specific territory within that habitat, and finally selecting a nest site (Burger 1985). Appropriate habitat selection affects fitness as measured by reproductive success (Gibo et al. 1976). Choice of a territory and the nest site may be influenced by the proximity of conspecifics, by physiognomic aspects of the habitat, or by a bird's previous nest site (Partridge 1978, Titus & Mosher 1981). Individuals may have a strong attraction to conspecifics as in colonies of seabirds, or may exclude conspecifics from large territories resulting in the spacing out of pairs with many meters between them. Heterospecifics may also influence nest site choices, as some birds avoid other species because of the risk of predation (Greig-Smith 1982) or aggression (Murray 1971). Some species regularly nest in association with heterospecifics that provide early warning or predator defense (Koskimies 1957, Fuchs 1977, Nuechterlein 1981, Eriksson & Gotmark 1982, Burger 1984). Physiognomic aspects of the habitat affecting nest site selection include structural stability (McCrimmon 1978, Coon et al. 1981), foliage density (Furrer 1980), slope (Volkman & Trivelpiece 1981), elevation for ground nests (Burger & Lesser 1978, Burger 1980), availability of places to take flight (Duffy 1984), vegetation height (Morse 1985, Cody 1985), nest hole characteristics (Alatalo et al. 1985), orientation relative to the sun (Finch 1983, Reynolds & Knapton 1984), location within the colony (center vs edge, Coulson 1968), proximity to water (Jackson & Burchfield 1975, Brennan et al. 1987), and perch site characteristics (Gochfeld 1979, Jones 1985). The physiognomic features used to select nest sites affect breeding activities such as territorial defense, mate acquisition, predator detection and/or avoidance, inclement weather amelioration, and food acquisition (for some species). Competition for suitable nest sites is presumably stronger within a species, but interspecific competition occurs for some species, particularly colonially nesting species (Burger 1981). Size plays a major role in interspecific interactions, the larger species usually winning direct aggressive interactions (Schoener 1974), or through interference competition (Case & Gilpin 1974). Many colonies of seabirds and of Ciconiiformes are mixed species colonies, and competition for nest sites in such colonies has been examined (McCrimmon 1978, Burger 1979, Volkman & Trivelpiece 1981). In addition, other species, such as passerines or shorebirds, may nest in these colonies, and in some cases this may represent a strong preference (Koskimies 1957). Yet how these species select nest sites within a colony of waterbirds is seldom examined. In this paper we examine habitat and nest site I Received 6 August 1987, revision accepted 9 March 1988. 244 BIOTROPICA 21(3): 244-249 1989 This content downloaded from 207.46.13.148 on Sun, 11 Sep 2016 04:09:10 UTC All use subject to http://about.jstor.org/terms selection in zenaida doves (Zenaida aurita) nesting with roseate (Sterna dougallii) and bridled (S. anaethetus) terns on two islands in the Culebra National Wildlife Refuge, Puerto Rico. We were interested in determining if these three species use the same general habitat, if the doves selectively nest near the terns (to derive possible antipredator advantages), and whether the dove nest site choices differ from the available habitat. STUDY AREA AND METHODS The zenaida dove occurs throughout the West Indies where it normally breeds on bushes and trees in coastal areas (Gosse 1847, Biaggi 1970, Raffaele 1983, J. Wiley, pers. comm.) although with extensive clearing, the doves now range throughout the islands (J. Wiley, pers. comm.). On the islands of the Culebra Archipelago, however, most nests are on the ground. Robertson (1962) reported that zenaida doves withdraw to offshore islets to breed in aggregations in the Virgin Islands. They do, however, also nest on the main islands (Nellis et al. 1984). We studied zenaida doves nesting on two small islands, Cayo Raton and Cayo del Agua, on the Culebra National Wildlife Refuge, off the east coast of Puerto Rico during June 1987. The doves nest on most islands around Culebra, but we chose to study them on islands with bridled and roseate terns that had a diversity of habitats from rocky slopes and cliffs to flat, heavily-vegetated places. Cayo Raton is dome shaped with 20 m high cliffs on the west, and steep vegetated slopes on the other sides (2.5 km from the Culebra mainland). It is covered on top with low, dense, bushy vegetation (Conocarpus erectus, Cyperus planifolius) that is 80-140 cm high. Cayo del Agua, 0.5 km from Raton, has low cliffs with the flat top surface having low, dense vegetation (2 km from Culebra). Further descriptions of the islands are given by Kepler and Kepler (1978). We timed our visit to coincide with tern and dove nesting activities so the locations of all nests could be determined. On these islands vegetation changes only slightly during the nesting season (S. Furniss, pers. comm.), thus the vegetation in early June was similar to when the doves selected their nest sites 2-3 weeks before. We spent five days on Cayo Raton and one day on Cayo del Agua. We designated four general habitat types according to vegetation cover (Table 1). Dense vegetation habitats contained dense vines or bushes, whereas habitats without vegetation were either smooth rock boulders or rocky slopes. We used five transects (10 m wide) to determine the percent of each island composed of each vegetation type. Dove nests were located by walking transects that were 1-2 m apart over the entire island. At this distance probably all doves were flushed from their nests. In rocky areas, we searched all crevices for eggs or chicks. We recorded the following data for all dove nests located: distance of observer from the nest when the bird flushed, clearance above the nest (distance free from vegetation or rocks), substrate (dirt, rock, vegetation), slope (00) of the nest and of zone 1 m around the nest, distance to clearing (where dove could easily take flight) and colony edge, dominant vegetation species, height of nearest rock and vegetation, height of vertical rock (within 50 cm of the nest), percent vegetation cover within 0. 5, 1 and 5 m of the nest, percent rock cover within 1 m, percent grass cover within 1 m, distance to the nearest vegetation, visibility of nest from above, and visibility of nest from 1 and 2 m, nest contents, nearest neighbor distance, and species of nearest neighbor. Percent cover is an estimate of the percent of the ground covered by vegetation, whereas visibility refers to the percent of the eggs or chicks that can be seen from above, from 1 m and 2 m from the nest. Thus, there might be only 40 percent vegetation cover within 1 m of the nest, but it might completely obscure the nest (0% visibility). We recorded the same characteristics for random points on both islands. Random points were selected by using a table of random numbers to generate the distance along the transect, the direction from the transect, and the distance away from the transect of each point. We used Kruskal-Wallis x2 tests to analyze differences between nests and random points, and contingency table x2 tests to analyze differences in occurrence among habitats. We present means ? standard errors in the text.

Etude menee sur cinq ans (sur Ste Croix et St Thomas) des variations saisonnieres de la population de: Zenaida aurita, Herpestes auropunctatus, Columba squamosa, C. leucocephala