Dove Habitat Association Based on Remotely Sensed Land Cover Types in South Texas

Abstract: We tested whether presence of white-winged doves (Zenaida asiatica) and mourning doves (Z. macroura) in South Texas, USA, was associated with any of the land cover types recorded in the 2001 National Land Classification Database. We used the point-transect method to conduct presence—absence surveys for both species at 236 points encompassing 744 observations. Within predefined land cover types surrounding each survey point, we used Geographic Information Systems to determine the proportions of each land cover type present. We used randomization tests to compare proportions of land cover types present at points with and without doves. We used Program DISTANCE to estimate dove densities at survey points and to test whether certain land cover types were associated with greater dove densities. Our results indicated white-winged dove occurrence in South Texas was positively associated with urban land and cropland, whereas mourning dove occurrence was positively associated with cropland. For land cover types found to be associated with dove presence, estimated density for each dove species increased as the proportion of associated land cover type increased. These results can assist wildlife biologists in the development of a more efficient and targeted protocol for censusing doves. In addition, our methods can be applied to other species across several geographic areas and landscape scales.     

A Hierarchical Model for Estimating Change in American Woodcock Populations

Abstract The Singing-Ground Survey (SGS) is a primary source of information on population change for American woodcock (Scolopax minor). We analyzed the SGS using a hierarchical log-linear model and compared the estimates of change and annual indices of abundance to a route regression analysis of SGS data. We also grouped SGS routes into Bird Conservation Regions (BCRs) and estimated population change and annual indices using BCRs within states and provinces as strata. Based on the hierarchical model–based estimates, we concluded that woodcock populations were declining in North America between 1968 and 2006 (trend = −0.9%/yr, 95% credible interval: −1.2, −0.5). Singing-Ground Survey results are generally similar between analytical approaches, but the hierarchical model has several important advantages over the route regression. Hierarchical models better accommodate changes in survey efficiency over time and space by treating strata, years, and observers as random effects in the context of a log-linear model, providing trend estimates that are derived directly from the annual indices. We also conducted a hierarchical model analysis of woodcock data from the Christmas Bird Count and the North American Breeding Bird Survey. All surveys showed general consistency in patterns of population change, but the SGS had the shortest credible intervals. We suggest that population management and conservation planning for woodcock involving interpretation of the SGS use estimates provided by the hierarchical model.     

Assessing Landbird Monitoring Programs and Demographic Causes of Population Trends

Abstract: Population trend data from the North American Breeding Bird Survey (BBS) have been used to identify conservation priorities and justify major conservation initiatives. Yet the BBS has been criticized for potential habitat bias and reliance on abundance indices to estimate trends. We compared 1992–2003 BBS trend estimates to trend estimates derived from bird-banding data collected as part of the Monitoring Avian Productivity and Survivorship (MAPS) program for 36 wood warbler species. Similarity in trends between the 2 monitoring programs at the survey-wide and program-wide scales suggested that each program can provide accurate trend information. The MAPS program, however, was designed primarily to complement (rather than duplicate) count-based efforts, such as the BBS, by providing estimates or indices of demographic rates. Demographic data from MAPS can be used to lend insight into proximate (demographic) causes of population trends and inform management. We illustrate this with analyses of 1992–2003 MAPS data for yellow warbler (Dendroica petechia). We used reverse-time capture-recapture models to evaluate importance of new recruits (including immigrating adults and young from the previous year) relative to surviving adults in explaining variation in trend among BBS physiographic strata. We included the number of young per adult captured (an index of productivity) as a covariate in models to assess effects of productivity on trends. Survival was the key demographic driver of recent population trends. Comparison of MAPS productivity indices and adult apparent survival rate estimates to BBS trend estimates largely confirmed this inference. We suggest that increased MAPS coverage, better coordination between MAPS and the BBS, and continued development of analytical methods that link the 2 programs will enhance the value of these monitoring efforts to land managers and conservation planners working at a variety of spatial scales.     

Invasive birds in a novel landscape: habitat associations and effects on established species

Studying the relationships among introduced species, their preferred habitats, and native species can be important for predicting the effects of invasions on native populations. Examining the colonization of North America by the Eurasian collared‐dove Streptopelia decaocto, we quantified the habitat characteristics of sites most likely to be occupied by this invasive bird species in the early stages of the invasion. Further, we studied the relationship between collared‐dove abundance and the abundance of other dove species in the study area, anticipating a negative effect on established species following the introduction of a potential competitor. Linking satellite‐derived landcover data with winter bird community data gathered from 444 study sites in Florida, USA from 1999 to 2008, we found that collared‐doves were more likely to occur in landscapes that had been highly‐modified by human activity than in forested landscapes. Collared‐dove abundance increased as the proportion of the landscape characterized as low‐intensity development and medium/high‐intensity development increased. The probability of collared‐doves occurring at a site was also related to the spatial proximity of other sites reporting doves (positive spatial autocorrelation). Contrary to our expectations, the site‐level abundance of four other dove species all increased with collared‐dove abundance throughout the sampling period. Interactions between collared‐doves and native species should be further studied in different environments as this invasive bird rapidly colonizes North America.     

Co-Occurrence and Occupancy of Mourning Doves and Eurasian Collared-Doves

Understanding how land cover and potential competition with invasive species shape patterns of occupancy, extirpation, and colonization of native species across a landscape can help target management for declining native populations. Mourning dove (Zenaida macroura) populations have declined throughout the United States from 1965–2015. The expansion of the Eurasian collared-dove (Streptopelia decaocto), an introduced species with similar food preferences, may further threaten mourning dove populations. We analyzed data from 2009–2016 from a large-scale monitoring program in the Western Great Plains of the United States in a 2-species occupancy model to assess the effects of collared-doves on mourning dove distributions, while accounting for imperfect detection and variation in land cover across the landscape. Mourning dove occupancy was stable or increasing across our study area, and despite overlap in resource use and co-occurrence between mourning doves and Eurasian collared-doves, we found no evidence that collared-doves are extirpating mourning doves from preferred habitat during the breeding season. © 2020 The Wildlife Society.     

The effect of habitat on the range expansion of a native and an introduced bird species

Aim Range expansion across a heterogeneous landscape may depend on the habitat selected and used by the expanding species. If habitat selection influences range expansion then localities colonized by a species should contain a greater proportion of favoured habitat (and less non‐habitat) than other nearby localities not colonized. White‐winged doves (Zenaida asiatica) and Eurasian collared doves (Streptopelia decaocto) are two bird species that provide an excellent opportunity to test this hypothesis, because the geographic ranges of both species have been expanding in North America for more than two decades. Location Continental USA. Methods We used distribution data from the North American Breeding Bird Survey to test whether the landscapes occupied by each species contained a greater proportion of favoured habitat (urban land, grassland/pasture, shrub land and cropland) and a lower proportion of non‐habitat (forest land) than landscapes where doves were not found. We tested each species separately in each of three broad expansion areas, namely East, Central and West. We also compared rates of spatial spread between expansion areas and between the two species. Results As predicted, both species tended to occupy landscapes with greater proportions of urban land, shrub land and cropland but with less forest land compared with landscapes without doves, in all three expansion areas. Contrary to prediction, occupied landscapes tended to have slightly less grassland/pasture than unoccupied landscapes. Rates of spread differed between the two species and among expansion areas. Main conclusions Range expansion and the extent to which a species fills or saturates its range are influenced by the habitat ecology of the expanding species. Species colonize localities based on the availability of suitable habitat. However, the role of habitat in a species’ range expansion does depend somewhat on the greater geographical setting. Over large regional and geographical scales, range expansion (rate of spread and saturation) may proceed unevenly, suggesting that range expansion is a very dynamic and context‐specific process.     

Experimental infection of Eurasian collared‐dove (Streptopelia decaocto) with West Nile virus

The Eurasian collared‐dove (Streptopelia decaocto) has recently experienced a population explosion in North America. It is frequently infected with West Nile virus (WNV). To test the hypothesis that the Eurasian collared‐dove is competent to transmit WNV, we experimentally infected two cohorts of doves with two different strains of WNV, CO08, and NY99, respectively. Both virus strains induced a low‐level viremia, capable of infecting a small fraction of vector mosquitoes. We suggest that the Eurasian collared‐dove plays a relatively insignificant role as an amplifying host for WNV, but it may be important where it is locally abundant.     

Representation of Landcover Along Breeding Bird Survey Routes in the Northern Plains

ABSTRACT The North American Breeding Bird Survey (BBS) is used extensively to make inferences about populations of many North American bird species and is increasingly being used for avian conservation planning. How well BBS routes represent the landscape is poorly known, even though accuracy of representation could significantly affect inferences made from BBS data. We used digital landcover data to examine how well landcover within 400-m buffers around BBS routes represented the surrounding landscape (the route neighborhood) for 52 routes in the Prairie Pothole Region of North Dakota and South Dakota. Differences in composition between landcover along BBS routes and the route neighborhood were not statistically significant for upland cover classes. The area of temporary and seasonal wetland basins was accurately represented by BBS routes in our study area, but the area of semipermanent and permanent wetland basins was significantly underrepresented along BBS routes. Number of wetland basins and upland patches was higher along routes. Area of urban, forest, and hay landcover classes was higher along routes, although differences were not statistically significant. Amount of bias in landcover representation was negatively correlated with the proportion of each landcover type in the study area, but bias was not correlated with area of the route neighborhoods. Differences between landcover along BBS routes and the route neighborhood were primarily attributable to increased anthropogenic activity along roads and siting of roads away from relatively large, deep water bodies. Our results suggest that inferences made from BBS data in our study region are likely biased for species that are associated with deeper-water habitats or are strongly influenced by landscape fragmentation. Inferences made from BBS data for species associated with uplands or shallow wetlands are less likely to be biased because of differences in landcover composition.     

Changes in the diversity structure of avian assemblages in North America

Aim To determine the major patterns of change in avian diversity structure over space and time at a local resolution and continental extent in non‐urban areas in North America. Location The contiguous United States and southern Canada. Methods We used 1673 North American Breeding Bird Survey (BBS) routes containing 547 terrestrial and aquatic species to estimate four diversity components: species richness, individual abundance, taxonomic distinctness and species evenness. We implemented three levels of analysis to examine changes in diversity structure on a yearly basis from 1968 to 2003: (1) a canonical analysis of discriminance, (2) a univariate analysis across BBS routes, and (3) a univariate analysis at individual BBS routes. We estimated changes in similarity in species composition over time between 470,730 BBS route pairs. We also estimated the level of human activity at BBS routes using three spatial anthropogenic databases. Results BBS routes were located in non‐urban areas in association with low nighttime light activity and moderately low human population densities. The analysis of diversity structure indicated the presence of two independent patterns: (1) a temporally consistent pattern defined by increasing species richness (12% increase from 1968 to 2003) associated with limited gains in taxonomic distinctness, and (2) an association between species abundance and evenness related to variability in abundance associated with the most abundant species. The similarity analysis indicated that BBS routes located closer to the Atlantic and the Pacific coasts of the United States experienced the strongest patterns of homogenization of species composition. Main conclusions Our results suggest that avian diversity structure has changed at the local scale in non‐urban areas in North America. However, there was no evidence for a consistent continent‐wide pattern. Instead, the evidence pointed to the presence of regional factors influencing diversity patterns. This study provides a foundation for more detailed investigations of the spatiotemporal and taxonomic details of these general patterns.     

Experimental transfer of Trichomonas gallinae (Rivolta, 1878) from white-winged doves to mourning doves

Isolates of Trichomonas gallinae (Rivolta, 1878) from white-winged doves, Zenaida asiatica (L.), were transferred experimentally to young mourning doves, Zenaida macroura (L.). Twenty-three of 25 mourning doves developed infections with isolates of T. gallinae from 25 white-winged doves. In addition, eight of eight rock doves (Columba livia Gmelin) were infected with duplicate isolates. All infected recipient birds harbored avirulent isolates except for one mourning dove which died from extensive oral lesions. However, repeated attempts using this isolate of T. gallinae to produce lesions in additional recipients were unsuccessful. Despite the findings of this study, it was suggested that future dove management strategies consider the possibility of disease outbreaks involving white-winged doves and susceptible populations of mourning doves.     

Observer aging and long‐term avian survey data quality

Long‐term wildlife monitoring involves collecting time series data, often using the same observers over multiple years. Aging‐related changes to these observers may be an important, under‐recognized source of error that can bias management decisions. In this study, we used data from two large, independent bird surveys, the Atlas of the Breeding Birds of Ontario (“OBBA”) and the North American Breeding Bird Survey (“BBS”), to test for age‐related observer effects in long‐term time series of avian presence and abundance. We then considered the effect of such aging phenomena on current population trend estimates. We found significantly fewer detections among older versus younger observers for 13 of 43 OBBA species, and declines in detection as an observer ages for 4 of 6 vocalization groups comprising 59 of 64 BBS species. Consistent with hearing loss influencing this pattern, we also found evidence for increasingly severe detection declines with increasing call frequency among nine high‐pitched bird species (OBBA); however, there were also detection declines at other frequencies, suggesting important additional effects of aging, independent of hearing loss. We lastly found subtle, significant relationships between some species' published population trend estimates and (1) their corresponding vocalization frequency (n ≥ 22 species) and (2) their estimated declines in detectability among older observers (n = 9 high‐frequency, monotone species), suggesting that observer aging can negatively bias long‐term monitoring data for some species in part through hearing loss effects. We recommend that survey designers and modelers account for observer age where possible.     

Land cover differentially affects abundance of common and rare birds

While rare species are vulnerable to global change, large declines in common species (i.e., those with large population sizes, large geographic distributions, and/or that are habitat generalists) also are of conservation concern. Understanding if and how commonness mediates species' responses to global change, including land cover change, can help guide conservation strategies. We explored avian population responses to land cover change along a gradient from common to rare species using avian data from the North American Breeding Bird Survey (BBS) and land cover data from the National Land Cover Database for the conterminous United States. Specifically, we used generalized linear mixed effects models to ask if species' commonness affected the relationship between land cover and counts, using the initial amount of and change in land cover surrounding each North American BBS route from 2001 to 2016. We quantified species' commonness as a continuous metric at the national scale using the logarithm (base 10) of each species' total count across all routes in the conterminous United States in 2001. For our focal 15-year period, we found that higher proportions of initial natural land cover favored (i.e., were correlated with higher) counts of rare but not common species. We also found that commonness mediated how change in human land cover, but not natural land cover, was associated with species' counts at the end of the study period. Increases in developed lands did not favor counts of any species. Increases in agriculture and declines in pasture favored counts of common but not rare species. Our findings show a signal of commonness in how species respond to a major dimension of global change. Evaluating how and why commonness mediates species' responses to land cover change can help managers design conservation portfolios that sustain the spectrum of common to rare species.     

Statewide and local recruitment of mourning doves in Missouri

Intensive harvests have the potential to greatly affect local mourning dove (Zenaida macroura) populations, a popular gamebird and songbird. To evaluate if recruitment was commensurate with harvest, we applied a ratio-based method to estimate local and statewide mourning dove recruitment across 7 public hunting areas in Missouri from 2005 to 2011. We estimated recruitment from preharvest adult sex ratios and harvest age ratios that incorporated various methods to address potential inherent biases (e.g., bias in the adults of unknown sex in preharvest samples, bias in unknown age wings, and local differential vulnerability; DV). Data from 356 radio-marked doves revealed a DV rate, where hatch year doves were, on average 2.7× more likely to be harvested than adult doves. Recruitment estimates for local areas were highly variable and in some cases, biologically unrealistic (e.g., >10 offspring/female), because of small preharvest sample sizes. However, data pooled statewide provided recruitment estimates of 3.1 offspring/female (±0.3 SE) or 4.1 offspring/female (±0.3 SE), assuming samples of unknown sex doves were female biased or male biased, respectively. Although statewide estimates agree with directly observed rates, the sex ratios and differential vulnerability comprising them vary considerably from what has been previously assumed. Whether preharvest sex ratios are biased from trapping methods has 2 important implications; either regional approaches have overestimated recruitment or the number of females in Missouri's population is much less than originally thought. Because each of these scenarios are important to understanding the effects of regional harvest management on Missouri's dove population, they highlight the importance of a better understanding of biases involved in estimating recruitment. © 2012 The Wildlife Society.     

Uneven Rates of Landscape Change as a Source of Bias in Roadside Wildlife Surveys

ABSTRACT Roadside survey data have been used frequently to assess species occurrence and population trends and to establish conservation priorities. However, most studies using such data assume that samples are representative of either the amount of habitat or its rate of change at larger spatial scales. We tested both of these assumptions for the Breeding Bird Survey (BBS) from 1974 to 2001 in New Brunswick, Canada. Our study focused on mature forest—a cover type that we predicted would be characterized by rapid change due to human activities and that is of high ecological importance. We also sought to determine whether land cover changes adjacent to BBS routes were related to bird population trends detected in BBS data. Within all 3 time periods examined (1970s, 1980s, and 1990s), the amount of mature forest adjacent to BBS routes was significantly lower than in surrounding 1° blocks of latitude and longitude. This could be problematic for studies that use roadside data to compare the relative abundance of species. On average, mature forest declined at a rate of-1.5% per year over the 28-year study period. We detected no significant difference in the rate of change between degree blocks and BBS routes over this time span. However, in the 1970s and 1980s, mature forest declined more rapidly in degree blocks (-2.7%/yr) than adjacent to BBS routes (-0.5/yr). We also found that the BBS trend for a mature forest-associated species, blackburnian warbler (Dendroica fusca), was correlated with the trend in mature forest along BBS routes. This, combined with slower rates of mature forest change along routes in the 1970s and 1980s, suggests that BBS data may have underestimated population declines during this period. It is important that research be conducted to test for potential biases in roadside surveys caused by uneven rates of landscape change, particularly in regions characterized by rapid habitat alteration.     

Combining Breeding Bird Survey and Christmas Bird Count Data to Evaluate Seasonal Components of Population Change in Northern Bobwhite

Abstract Annual surveys of wildlife populations provide information about annual rates of change in populations but provide no information about when such changes occur. However, by combining data from 2 annual surveys, conducted in different parts of the year, seasonal components of population change can be estimated. We describe a hierarchical model for simultaneous analysis of 2 continent-scale monitoring programs. The Christmas Bird Count is an early winter survey, whereas the North American Breeding Bird Survey is conducted in June. Combining information from these surveys permits estimation of seasonal population variance components and improves estimation of long-term population trends. The composite analysis also controls for survey-specific sampling effects. We applied the model to estimation of population change in northern bobwhites (Colinus virginianus). Over the interval 1969–2004, bobwhite populations declined, with trend estimate of −3.56% per year (95% CI = [−3.80%, −3.32%]) in the surveyed portion of their range. Our analysis of seasonal population variance components indicated that northern bobwhite populations changed more in the winter and spring portion of the year than in the summer and fall portion of the year. (JOURNAL OF WILDLIFE MANAGEMENT 72(1):44–51; 2008)     

Long‐term and widespread changes in agricultural practices influence ring‐necked pheasant abundance in California

Declines in bird populations in agricultural regions of North America and Europe have been attributed to agricultural industrialization, increases in use of agrochemical application, and increased predation related to habitat modification. Based on count data compiled from Breeding Bird Survey (BBS) from 1974 to 2012, Christmas Bird Count (CBC) collected from 1914 to 2013, and hunter data from Annual Game Take Survey (AGTS) for years 1948–2010, ring‐necked pheasants (Phasianus colchicus) in California have experienced substantial declines in agricultural environments. Using a modeling approach that integrates all three forms of survey data into a joint response abundance index, we found pheasant abundance was related to the amount of harvested and unharvested crop land, types of crops produced, amount of total pesticide applied, minimum temperature, precipitation, and numbers of avian competitors and predators. Specifically, major changes in agricultural practices over the last three decades were associated with declines in pheasant numbers and likely reflected widespread loss of habitat. For example, increases in cropland were associated with increased pheasant abundance during early years of study but this effect decreased through time, such that no association in recent years was evidenced. A post hoc analysis revealed that crops beneficial to pheasant abundance (e.g., barley) have declined substantially in recent decades and were replaced by less advantageous crops (e.g., nut trees). An additional analysis using a restricted data set (1990–2013) indicated recent negative impacts on pheasant numbers associated with land use practices were also associated with relatively high levels of pesticide application. Our results may provide valuable information for management policies aimed at reducing widespread declines in pheasant populations in California and may be applicable to other avian species within agricultural settings. Furthermore, this general analytical approach is not limited to pheasants and could be applied to other taxa for which multiple survey data sources exist.     

Social aspects of the termination of incubation behaviour in the ring dove (Streptopelia risoria)

The role of male and female ring doves (Streptopelia risoria) in the termination of incubation behaviour was studied first by observing heterosexual pairs (male and female) and homosexual pairs (female and female) (experiment 1). Male/female pairs incubate for 22 days whether squab or unhatched eggs are in the nest; female/female pairs incubate for 25 days or more. Males take the lead in terminating incubation. Another group of doves, separated from their (heterosexual) partners late in incubation, were exposed to one of four stimulus conditions: (a) an incubating dove; (b) a reproductively-inactive dove; (c) a reproductively-active dove; (d) visual isolation. Females extended the duration of their incubation when exposed to the sight of another dove, except when the stimulus dove was reproductively active. Males did not extend their incubation in response to these social cues. These social stimuli terminating incubation behaviour coordinate the withdrawal from the nest of male and female in preparation for the next breeding episode in this monogamous species.     

Landscape effects on large scale abundance patterns of turtle doves Streptopelia turtur in Portugal

The European turtle dove is both a highly valued game species and a species of conservation concern, which is declining due probably to a combination of habitat degradation and unsustainable hunting. Although declines seem to be less severe in the Mediterranean region, it remains uncertain the extent to which ongoing land use changes will negatively affect this species. This study examined this issue, by estimating the effects of landscape composition on the broad scale abundance pattern of breeding turtle doves in continental Portugal. Turtle doves were surveyed in the breeding seasons of 2002 and 2003, from 3160 point counts spaced at about 1-km intervals along 158 transects of about 20 km, evenly covering the country. The frequency of occurrence of turtle doves at each transect was used as a proxy of species abundance, and related using GAM modelling to 21 variables describing land cover and woody linear features (e.g., hedgerows and riparian galleries). Turtle doves were most abundant in north- and central-eastern Portugal, with high abundances also recorded in the regions around Lisbon and along the Guadiana valley. Abundances were positively related to forest cover, particularly by broadleaved forests and by pine stands without woody understory, to cover by permanent crops, and to the density of woody linear habitats. Results suggest that conservation of Mediterranean turtle doves requires policies and management strategies reversing the pervasive trends of forest management neglect and agricultural abandonment, while preserving hedgerows and riparian galleries in more intensive agricultural landscapes.     

Use of saguaro fruit by white-winged doves: isotopic evidence of a tight ecological association

We report the use of stable isotope and crop content analyses to quantify the use of saguaro (Carnegiea gigantea) nectar and fruit by migratory desert white-winged doves (Zenaida asiatica mearsnii). Saguaro resources had characteristically ¹³C-enriched CAM values (δ¹³C=–12.8±0.7‰ SD VPDB and –13.1±0.5‰ SD VPDB for nectar and fruit, respectively) relative to other food plants used by doves (δ¹³C_C3=–24.9±3.3‰ SD VPDB). The water contained in saguaro nectar and fruit was deuterium enriched (δD=19.6±2.0‰ SD VSMOW and 48.4±1.6‰ SD VSMOW for nectar and fruit, respectively) relative to other water sources (ranging from –41 to –19‰ VSMOW). During the fruiting season, there was a positive correlation between δ¹³C in dove liver tissues and percent of saguaro in crop contents. A two-point mixing model indicated that during the peak of saguaro fruit use, most of the carbon incorporated in dove tissues was from saguaro. Desert white-winged doves appear to be saguaro specialists. Averaged over the period when doves were resident, saguaro comprised about 60% of the total carbon incorporated into dove tissues. Tissue δ¹³C and δD of body water showed a significant positive correlation, indicating that doves were using saguaro as a source of both nutrients and water. However, at the peak of saguaro utilization, the doves’ body-water δD was more positive (by about 20‰) than saguaro fruit water. We hypothesize that this enrichment is due to fractionated evaporative water losses by doves. Using dove carbon isotope data and a two end-point mixing model we estimate that, on average, doves consume the equivalent of 128 saguaro fruits per season; each fruit contains on average 26.0±14.8 g SD of pulp (wet mass) of which 19.4 g is water. Stable isotopes have been used to produce qualitative re-constructions of animal diets. Our study shows that they can be used to provide quantitative estimates of the flow of nutrients from resources into consumers as well. Received: 30 September 1999 / Accepted: 23 March 2000     

The diet of the Henderson fruit dove Ptilinopus insularis.

Over 13 months we studied the diet of the fruit dove Ptilinopus insularis , endemic to the exceptionally remote South Pacific island of Henderson. Nineteen plant species were recorded in the diet by faecal analysis and direct observation. The fruit of Procris pedunculata (Urticaceae) was the most common food. When it was unavailable, the doves turned to the young shoots of the fern Phymatosorus scolopendria. Subject to a maximum diameter of about 18 mm, the doves ate nearly all types of available fruit. By combining the diet studies with studies of plant phenology we were able to assess which, of the fruits available, were preferred. The doves preferred the larger fruit. Given that the fruit dove today eats nearly all available fruit species, we ponder on how it co-existed with at least two other pigeon species that formerly lived on Henderson Island.