Search

Functional traits can determine pairwise species interactions, such as those between plants and pollinators. However, the effects of biogeography and evolutionary history on trait-matching and trait-mediated resource specialization remain poorly understood.We compiled a database of 93 mutualistic hummingbird-plant networks (including 181 hummingbird and 1,256 plant species), complemented by morphological measures of hummingbird bill and floral corolla length. We divided the hummingbirds into their principal clades and used knowledge on hummingbird biogeography to divide the networks into four biogeographical regions: Lowland South America, Andes, North & Central America, and the Caribbean islands. We then tested: (a) whether hummingbird clades and biogeographical regions differ in hummingbird bill length, corolla length of visited flowers and resource specialization, and (b) whether hummingbirds' bill length correlates with the corolla length of their food plants and with their level of resource specialization.Hummingbird clades dominated by long-billed species generally visited longer flowers and were the most exclusive in their resource use. Bill and corolla length and the degree of resource specialization were similar across mainland regions, but the Caribbean islands had shorter flowers and hummingbirds with more generalized interaction niches. Bill and corolla length correlated in all regions and most clades, that is, trait-matching was a recurrent phenomenon in hummingbird-plant associations. In contrast, bill length did not generally mediate resource specialization, as bill length was only weakly correlated with resource specialization within one hummingbird clade (Brilliants) and in the regions of Lowland South America and the Andes in which plants and hummingbirds have a long co-evolutionary history. Supplementary analyses including bill curvature confirmed that bill morphology (length and curvature) does not in general predict resource specialization.These results demonstrate how biogeographical and evolutionary histories can modulate the effects of functional traits on species interactions, and that traits better predict functional groups of interaction partners (i.e. trait-matching) than resource specialization. These findings reveal that functional traits have great potential, but also key limitations, as a tool for developing more mechanistic approaches in community ecology.A free Plain Language Summary can be found within the Supporting Information of this article.

The physiological demands of flight exert strong selection pressure on avian morphology and so it is to be expected that the evolutionary loss of flight capacity would involve profound changes in traits. Here, we investigate morphological consequences of flightlessness in a bird family where the condition has evolved repeatedly. The Rallidae include more than 130 recognized species of which over 30 are flightless. Morphological and molecular phylogenetic data were used here to compare species with and without the ability to fly in order to determine major phenotypic effects of the transition from flighted to flightless. We find statistical support for similar morphological response among unrelated flightless lineages, characterized by a shift in energy allocation from the forelimbs to the hindlimbs. Indeed, flightless birds exhibit smaller sterna and wings than flighted taxa in the same family along with wider pelves and more robust femora. Phylogenetic signal tests demonstrate that those differences are independent of phylogeny and instead demonstrate convergent morphological adaptation associated with a walking ecology. We found too that morphological variation was greater among flightless rails than flighted ones, suggesting that relaxation of physiological demands during the transition to flightlessness frees morphological traits to evolve in response to more varied ecological opportunities., This study investigated the consequences of the evolution of flightlessness in birds from the Rallidae family. The results show similar morphological responses within the flightless species with a shift in energy allocation from the forelimbs to the hindlimbs. By testing the phylogenetic signal, this research proves that the differences between flightless and flighted birds are not an effect of the phylogeny but reflect a morphological convergent adaptation to the loss of flight.

Invasive rats have colonized most of the islands of the world, resulting in strong negative impacts on native biodiversity and on ecosystem functions. As prolific omnivores, invasive rats can cause local extirpation of a wide range of native species, with cascading consequences that can reshape communities and ecosystems. Eradication of rats on islands is now becoming a widespread approach to restore ecosystems, and many native island species show strong numerical responses to rat eradication. However, the effect of rat eradication on other consumers can extend beyond direct numerical effects, to changes in behavior, dietary composition, and other ecological parameters. These behavioral and trophic effects may have strong cascading impacts on the ecology of restored ecosystems, but they have rarely been examined. In this study, we explore how rat eradication has affected the trophic ecology of native land crab communities. Using stable isotope analysis of rats and crabs, we demonstrate that the diet or trophic position of most crabs changed subsequent to rat eradication. Combined with the numerical recovery of two carnivorous land crab species (Geograpsus spp.), this led to a dramatic widening of the crab trophic niche following rat eradication. Given the established importance of land crabs in structuring island communities, particularly plants, this suggests an unappreciated mechanism by which rat eradication may alter island ecology. This study also demonstrates the potential for stable isotope analysis as a complementary monitoring tool to traditional techniques, with the potential to provide more nuanced assessments of the community- and ecosystem-wide effects of restoration.

Seabird islands house large colonies of seabirds that spend most of their time at sea but return to land to breed. Seabirds deposit large amounts of nutrients of marine origin on islands; many species also disturb soil and vegetation during the building and maintenance of nests. Physical and chemical disturbance by seabirds alters the chemical composition of soils and vegetation and modifies plant and animal communities. In many cases, the additional nutrients brought in by seabirds subsidise native animal populations, including endemic species frequently found on remote islands. Introduction of predators such as rats, cats and mustelids has resulted in population decreases or local extinction of many seabird species, and this leads to large changes in nutrient cycling and plant and animal populations. Recent efforts at eradication of introduced seabird predators have been successful, but re-establishment of seabirds and seabird island communities will likely take additional active restoration. Key Concepts: Seabirds deposit nutrients of marine origin in terrestrial environments. Seabirds impose physical disturbance on soils and vegetation in the process of building and maintaining nests. Seabird islands often house endemic and native plants and animals that are no longer found on the mainland. The presence of seabirds selects for plants that are tolerant of high-disturbance and high-nutrient conditions. Nutrient subsidies from guano (i.e. nitrogen and phosphorus) can support increased primary and secondary productivity. Introduced mammalian predators have eradicated seabirds and other species on many seabird islands. The reduction of seabird populations by introduced predators alters plant and animal communities and nutrient cycling. Eradication of seabird predators is a necessary first step towards restoration of seabird islands. Restoration of seabird islands often requires active management of seabirds, other animals and plant populations. Protection of seabird islands will require international cooperation on land and at sea. Keywords: allochthonous inputs; ecosystem engineering; invasive predators; island endemism; island restoration; nutrient subsidies; physical disturbance; predator eradication; superpredator

Aim: We undertook the largest comparative study to date of the form of the island species–area relationship (ISAR) using 207 habitat island datasets and 601 true island datasets. We also undertook analyses of (a) the factors influencing z‐ and c‐values of the power (log–log) model and (b) how z and c vary between different island types. Location:Global. Methods: We used an information theoretic approach to compare the fit of 20 ISAR models to 207 habitat island datasets. Model performance was ranked according to pre‐set criteria, including metrics of generality and efficiency. We also fitted the power (log–log) model to each dataset and analysed variation in parameter estimates and model fits as a function of key dataset characteristics using linear models and constrained analysis of principal coordinates. Results: The power (nonlinear) model provided the best fit to the most datasets, and was the highest ranked model overall. In general, the more complex models performed badly. Average z‐values were significantly lower for habitat island datasets than for true islands, and were higher for mountaintop and urban habitat islands than for other habitat island types. Average c‐values were significantly lower for oceanic islands, and significantly higher for inland water‐body islands, than for habitat islands. Values of z and c were related to dataset characteristics including the ratio of the largest to smallest island and the maximum and minimum richness values in a dataset. Main conclusions: Our multimodel comparisons demonstrated the nonlinear implementation of the power model to be the best overall model and thus to be a sensible choice for general use. As the z‐value of the log–log power model varied in relation to ecological and geographical properties of the study systems, caution should be employed when using canonical values for applied purposes.

AimPatterns that relate species richness with fragment area (the species–area relationship, SAR) and with isolation (the species–isolation relationship, SIR) are well documented. However, those that relate species density – the number of species within a standardized area – with fragment area (D-SAR) or isolation (D-SIR) have not been sufficiently explored, despite the potential for such an analysis to disentangle the underlying mechanisms of SARs and SIRs. Previous spatial theory predicts that a significant D-SAR or D-SIR is unlikely to emerge in taxa with high dispersal limitation, such as plants. Furthermore, a recent model predicts that the detection and the significance of D-SARs or D-SIRs may decrease with grain size. We combined a literature review with grain size-dependent sampling in a fragmented landscape to evaluate the prevalence and grain size-dependent nature of D-SARs and D-SIRs in plants.LocationWorldwide (review) and a semi-arid agro-ecosystem in Israel (case study).MethodsWe combined an extensive literature review of 31 D-SAR studies of plants in fragmented landscapes with an empirical study in which we analysed grain size-dependent D-SARs and D-SIRs using a grain size-dependent hierarchical sampling of species density and species richness in a fragmented, semi-arid agro-ecosystem.ResultsWe found that significantly increasing D-SARs are rare in plant studies. Furthermore, we found that the detection of a significant D-SAR is often possible only after the data have been stratified by species, habitat or landscape characteristics. The results from our case study indicated that the significance and the slopes of both D-SARs and D-SIRs increase as grain size decreases.Main conclusionsThese results call for a careful consideration of scale while analysing and interpreting the responses of species richness and species density to fragmentation. Our results suggest that grain size-dependent analyses of D-SARs and D-SIRs may help to disentangle the mechanisms that generate SARs and SIRs and may enable early detection of the effects of fragmentation on plant biodiversity.

Everglades tree islands are patches of woody vegetation ranging in size from 0.1 to 70 ha embedded in a marsh matrix. Tree islands are an essential component of the tree island–ridge–slough topographic continuum and the trees provide a mechanism for the accumulation of phosphorus in the landscape, enhancing landscape biocomplexity. This article reviews the literature of tree island ecology and describes the predicted effects of five ecosystem restoration scenarios on tree islands. Elevations of the highest points on 404 islands across the Everglades were used to assess the effects of the scenarios at a landscape level, while the plant communities of nine islands were modeled to assess local effects. Evaluation of the restoration scenarios were based on three critical components needed to maintain and restore tree islands: hydrology that allows the survival of woody species, seasonally appropriate water flow to move nutrients off the heads of islands, and oligotrophy. All restoration options generally improved water conditions for tree growth, although flooding of islands in some areas was predicted to increase. Scenarios with the greatest amount of decompartmentalization and largest capacity to remove phosphorus from water entering the Everglades provided the best chance of delivering vigorous pulses of sheet flow in directions parallel to historic flows and maintaining oligotrophy. Tree islands are “bellwethers” of restoration actions because their maintenance on the landscape requires multiple ecosystem functions.

Aim We estimated the patterns of seed deposition provided by the eyed lizard, Timon lepidus, and evaluated whether these patterns can be generalized across plant species with different traits (fruit and seed size) and spatial distributions. Location Monteagudo Island, Atlantic Islands National Park (north-western Spain). Methods We radio-tracked seven lizards for 14 days and estimated their home ranges using fixed kernels. We also geo-referenced all fruit-bearing individuals of four plant species dispersed by eyed lizards in the study area (Corema album, Osyris alba, Rubus ulmifolius and Tamus communis), measured the passage time of their seeds through the lizard gut, and estimated seed predation in four habitats (bare sand, grassland, shrub and gorse). Seed dispersal kernels were estimated using a combination of these data and were combined with seed predation probability maps to incorporate post-dispersal seed fate (‘seed survival kernels’). Results Median seed gut-passage times were around 52–98 h, with maximum values up to 250 h. Lizards achieved maximum displacement in their home ranges within 24–48 h. Seed predation was high (80–100% of seeds in 2 months), particularly under Corema shrub and gorse. Seed dispersal kernels showed a common pattern, with two areas of preferential seed deposition, but the importance of these varied among plant species. Interspecific differences among dispersal kernels were strongly reduced by post-dispersal seed predation; hence, seed survival kernels of the different plant species showed high auto- and pairwise-correlations at small distances (

Zeeba Khan benefited from a JAE-PRE fellowship from the Spanish National Research Council (CSIC). This work was financed by the Spanish Ministry of Science and Education project (CGL2007-61165/BOS) and is also framed within project CGL2010-18759/BOS, supported by FEDER funds from the European Union.

Aim Large-bodied vertebrates often have a dramatic role in ecosystem function through herbivory, trampling, seed dispersal and nutrient cycling. The iconic Galapagos tortoises (Chelonoidis nigra) are the largest extant terrestrial ectotherms, yet their ecology is poorly known. Large body size should confer a generalist diet, benign digestive processes and long-distance ranging ability, rendering giant tortoises adept seed dispersers. We sought to determine the extent of seed dispersal by Galapagos tortoises and their impact on seed germination for selected species, and to assess potential impacts of tortoise dispersal on the vegetation dynamics of the Galapagos. Location Galapagos, Ecuador. Methods To determine the number of seeds dispersed we identified and counted intact seeds from 120 fresh dung piles in both agricultural and national park land. To estimate the distance over which tortoises move seeds we used estimated digesta retention times from captive tortoises as a proxy for retention times of wild tortoises and tortoise movement data obtained from GPS telemetry. We conducted germination trials for five plant species to determine whether tortoise processing influenced germination success. Results In our dung sample, we found intact seeds from >45 plant species, of which 11 were from introduced species. Tortoises defecated, on average, 464 (SE 95) seeds and 2.8 (SE 0.2) species per dung pile. Seed numbers were dominated by introduced species, particularly in agricultural land. Tortoises frequently moved seeds over long distances; during mean digesta retention times (12days) tortoises moved an average of 394m (SE 34) and a maximum of 4355m over the longest recorded retention time (28days). We did not find evidence that tortoise ingestion or the presence of dung influenced seed germination success. Main conclusions Galapagos tortoises are prodigious seed dispersers, regularly moving large quantities of seeds over long distances. This may confer important advantages to tortoise-dispersed species, including transport of seeds away from the parent plants into sites favourable for germination. More extensive research is needed to quantify germination success, recruitment to adulthood and demography of plants under natural conditions, with and without tortoise dispersal, to determine the seed dispersal effectiveness of Galapagos tortoises.

Aim To compare the ability of island biogeography theory, niche theory and species–energy theory to explain patterns of species richness and density for breeding bird communities across islands wi ...

Faecal DNA isolation identifies two Columba species P. M rrero et al. Short communic Molecular identification of two threatened pigeon species (Columbidae) using faecal samples PATRICIA MARRERO, 1 * VICENTE M. CABRERA, 2 DAVID P. PADILLA 1 & MANUEL NOGALES 1 1 Island Ecology and Evolution Research Group, IPNA-CSIC, La Laguna E-38206, Tenerife, Canary Islands, Spain 2 Department of Genetics, University of La Laguna, La Laguna E-38271, Tenerife, Canary Islands, Spain

Aim The most obvious, although not exclusive, explanation for the increase of species richness with increasing sample area (the species–area relationship) is that species richness is ultimately linked to area-based increases in habitat heterogeneity. The aim of this paper is to examine the relative importance of area and habitat heterogeneity in determining species richness in nature reserves. Specifically, the work tests the hypothesis that species–area relationships are not positive if habitat heterogeneity does not increase with area. Location Sixteen nature reserves (area range 89–11,030 ha) in central Hungary. Methods Four-year faunistic inventories were conducted in the reserves involving c. 70 fieldworkers and 65 taxonomists. CORINE 50,000 land-cover maps were used for calculating the heterogeneity of the reserve landscape (number of habitat types, number of habitat patches and total length of edges). Results Large reserves were less heterogeneous than small reserves, probably because large reserves were established in large blocks of unproductive land whereas small reserves tended to be in more fertile land. In total, 3975 arthropod species were included in the analysis. The slope of the species–area relationship was positive only for Neuroptera and Trichoptera. There was no significant relationship in the other nine taxa examined (Collembola, Acari, Orthoptera, Thysanoptera, Coleoptera, Araneae, Diplopoda, Chilopoda, Diptera). The density (number of species ha−1) of all species, however, showed a positive correlation with heterogeneity. Main conclusions The general lack of fit of species–area relationships in this study is inconsistent with most previous published studies. Importantly, and unlike many other studies, habitat heterogeneity was not correlated with reserve area in the studied system. In the absence of this source of covariation, stronger relationships were identified that suggested a fundamental link between species richness and habitat heterogeneity. The results indicate that habitat heterogeneity rather than area per se is the most important predictor of species richness in the studied system.

Aim To identify the factors determining the relationship between habitat breadth and range size in the Mediterranean alien flora. Strong relationships have previously been uncovered between the two variables, which raises the possibility that habitat breadth could be used as a powerful predictor of invasiveness. However, previous investigations have yet to explain the phenomenon adequately, and greater understanding is required to establish its importance. Location Ninety-one islands in the Mediterranean basin. Methods I examined a data base of Mediterranean naturalized plant species created for the EU project EPIDEMIE, and containing information on island distributions, habitat occupancy and other descriptive parameters compiled from published floras. As a null hypothesis, I assumed that the relationship between habitat breadth and range size was a spatial scale artefact: if species colonize habitats at random, with the likelihood determined by a weighted probability in each habitat, then perceived habitat breadth will increase with the number of populations examined. I modelled this effect, and also considered circumstantial evidence (via indirect indicator variables) for six hypotheses to determine whether both habitat breadth and range size may be jointly driven by a major third factor (introduction frequency, residence time, dispersal ability, generation time, dependence on mutualisms, release from herbivory). The explanatory power of the hypotheses was compared using correlative statistics. Results When habitat breadth was modelled according to the null hypothesis, the form of its relationship with range size was very similar to that from the observed data, and the predicted range size estimates explained almost half the observed variance. The unexplained component may have been due to the same mechanism operating at finer spatial scales. From the direction of the residual effect, increases in habitat breadth were probably driven by range size rather than vice versa. None of the alternative hypotheses were found to be significant predictors of habitat breadth and range size in invasive species. Main conclusions Although a number of minor explanations cannot be ruled out (e.g. that widespread species have greater regional genetic diversity enabling them to adapt to local conditions), it seems likely that the habitat breadth effect is largely artefactual and does not have particularly great value as a predictor of invasiveness.

Aim To investigate the chronology, causes and consequences of the extinction of the autochthonous Pleistocene small mammals of Mallorca. Location Mallorca (Gymnesic Islands, Balearics, Western Mediterranean). Methods We have obtained the first direct 14C ages from the bone collagen of selected samples of two extinct endemic small mammals from Mallorca: the Balearic dormouse, Eliomys morpheus (Rodentia: Myoxidae) and the Balearic shrew, Asoriculus hidalgoi (Soricomorpha: Soricidae). We also present evidence for the absence of both endemics from the earliest Mallorcan archaeological sites and for the introduction of the garden dormouse, Eliomys quercinus, and the wood mouse, Apodemus sylvaticus. Combined information from direct dating of bone collagen of E. quercinus and A. sylvaticus and from cultural associations provides an adequate framework to establish the chronology of the faunal change and to compare it with the chronological information available on climatic change and the first arrival of humans on the islands. Results The chronological record includes the latest evidence available for the survival of endemic species and the earliest introduction of small mammals into Mallorca. We present ‘uncertainty periods for extinction’ (UPEs) of both endemic mammals based on the chronology of their last occurrence and on the inferred timing of their extinction (restricted UPEs). Main conclusions Possible causes for the extinction of autochthonous small mammals on Mallorca are discussed. Once we have discarded climatic causes, predation by invasive species, competition with newcomers and habitat deterioration, the introduction of diseases emerges as the most reasonable explanation for these extinctions. Based on the identification of changes in keystone species in Mallorcan ecosystems, we propose a tentative schedule of key ecological changes that have taken place over the past 5 millennia.

As evidence mounts that sea levels are rising, it becomes increasingly important to understand the role of ocean water within terrestrial ecosystem dynamics. Coastal sand dunes are ecosystems that occur on the interface of land and sea. They are classic ecotones characterized by zonal distribution of vegetation in response to strong gradients of environmental factors from the ocean to the inland. Despite the proximity of the dune ecosystem to the ocean, it is generally assumed that all vegetation utilizes only freshwater and that water sources do not change across the ecotone. Evidence of ocean water uptake by vegetation would redefine the traditional interpretation of plant-water relations in the dune ecosystem and offer new ideas for assessing maritime influences on function and spatial distribution of plants across the dune. The purpose of this study was to identify sources of water (ocean, ground, and rain) taken up by vegetation using isotopic analysis of stem water and to evaluate water uptake patterns at the community level based on the distribution and assemblage of species. Three coastal dune systems located in southern Florida, USA, and the Bahamian bank/platform system were investigated. Plant distributions across the dune were zonal for 61-94% of the 18 most abundant species at each site. Species with their highest frequency on the fore dune (nearest the ocean) indicate ocean water uptake as evidenced by delta 18O values of stem water. In contrast, species most frequent in the back dune show no evidence of ocean water uptake. Analysis of species not grouped by frequency, but instead sampled along a transect from the ocean toward the inland, indicates that individuals from the vegetation assemblage closest to the ocean had a mixed water-harvesting strategy characterized by plants that may utilize ocean, ground-, and/or rainwater. In contrast, the inland vegetation relies mostly on rainwater. Our results show evidence supporting ocean water use by dune vegetation and demonstrate an exciting relationship between seawater and ecotonal shifts in plant function of a terrestrial ecosystem.

Recent works have found the presence of native congeners to have a small effect on the naturalization rates of introduced plants, some suggesting a negative interaction (as proposed by Charles Darwin in The Origin of Species), and others a positive association. We assessed this question for a new biogeographic region, and discuss some of the problems associated with data base analyses of this type. Location Islands of the Mediterranean basin. Presence or absence of congeners was assessed for all naturalized alien plants species at regional, local and habitat scales. Using general linear models, we attempted to explain the abundance of the species (as measured by the number of islands where recorded) from their congeneric status, and assessed whether the patterns could be alternatively accounted for by a range of biological, geographical and anthropogenic factors. A simulation model was also used to investigate the impact of a simple bias on a comparable but hypothetical data set. Data base analyses addressing Darwin's hypothesis are prone to bias from a number of sources. Interaction between invaders and congenerics may be overestimated, as they often do not co-occur in the same habitats. Furthermore, intercorrelations between naturalization success and associated factors such as introduction frequency, which are also not independent from relatedness with the native flora, may generate an apparent influence of congenerics without implying a biological interaction. We detected no true influence from related natives on the successful establishment of alien species of the Mediterranean. Rarely-introduced species tended to fare better in the presence of congeners, but it appears that this effect was generated because species introduced accidentally into highly invasible agricultural and ruderal habitats have many relatives in the region, due to common evolutionary origins. Relatedness to the native flora has no more than a marginal influence on the invasion success of alien plants in the Mediterranean, although apparent trends can easily be generated through artefacts of the data base

Summary 1 Species attributes have been used to explain invasion patterns assuming the prevalence of biological mechanisms, although this approach often suffers several methodological and conceptual limitations, such as local idiosyncrasies, differences among habitats, phylogenetic constraints and insufficient sample size. 2 The relative importance of 15 species traits for explaining the abundance over 350 naturalized alien plant species was assessed across five Mediterranean islands (Corsica, Crete, Majorca, Malta and Sardinia). A comparative analysis accounting for phylogeny was used to examine variation in semi-quantitative estimates of species abundance in comparable habitats across the five island floras. 3 Species were divided into those with affinity for semi-natural, agricultural and ruderal habitats. Both vegetative and reproductive attributes were evaluated for individual islands and averaged across all islands. 4 Vegetative propagation, large leaf size, summer flowering, long flowering period and dispersal by wind or vertebrates were positively associated with average alien abundance across all five islands. Fewer significant trends were found in island-specific patterns. 5 The relative importance of a few reproductive traits is reflected in over-representation of Caryophyllales, Asterales and Poales (late flowering, large seed size and anemochory). Although significant covariation in traits was found there was no evidence for welldefined invasive syndromes. 6 Succulence was important in ruderal habitats, long flowering period in agricultural habitats and vertebrate seed dispersal in semi-natural habitats, suggesting that empty niches, avoidance of competitors and exploitation of mutualists, respectively, are important. 7 The study highlights the importance of estimating invasion success across a wide region, but analyses of specific invasion stages are also needed. Reproductive traits, which may be more relevant for long-distance colonization, and vegetative traits, which determine local dominance and persistence, were, nevertheless, both related to abundance within islands.

Aims To test a new concept for island ecology and metapopulation studies: if population persistence is a result of colonization and extinction processes, one would expect remarkable variability in the age of neighbouring populations – there would be both old and young (recently established) populations. Location The lake Mamry archipelago, Poland. Methods Forty trap transects with a total of 114 Barber traps were operating from June to September in 1997 and in 1998. The traps (0.5-L plastic beakers, mouth diameter 120 mm, with a 20 × 20 cm wooden roof) were placed by applying a stratified sampling design into 13 habitat types on the 16 islands and two mainland sites. The frequency of macropterous individuals in wing-dimorphic populations of three carabid species (Pterostichus melanarius, P. anthracinus and Carabus granulatus) was used as an indicator of population ‘age’. Results The frequency of macropterous individuals in the populations varied from 0% to 100% for P. melanarius (18.5% on average), from 0% to 91% for P. anthracinus (12.6%) and from 0% to 29% for C. granulatus (8.9%). Populations hosted more long-winged individuals (and were therefore interpreted as being younger) on smaller islands, compared with those inhabiting large islands and the mainland sites. The results also revealed that the viability of the populations of the autumn-breeding P. melanarius might be more affected by population size than that of the two studied spring breeders. Island connectivity did not have a significant effect on the frequency of macropterous individuals in the studied populations. Main conclusions The inverse relationship between the proportion of macropterous individuals and island size contrasts with the accepted theory of dispersal, which assumes that there is selection against dispersal on small islands. A regression analysis for population age and habitat characteristics reveals the extinction probability of a given population. I suggest that investigations based on variability in population age can help in studies of colonization–extinction processes that would otherwise face logistic and methodological obstacles.

Aim To determine the geographical variation in serotiny in a common New Zealand shrub as a contemporary indicator of past fire regimes. The distribution of serotiny could then be used to explore factors limiting the spread of fire-dependent vegetation. Location South Island, New Zealand. Methods Serotiny was assessed as the proportion of closed capsules on a shoot of standard stem diameter for 5–35 plants in 45 widely scattered populations of Leptospermum scoparium. Site characteristics, including locality, altitude, rainfall, habitat type and minimum burnable area were recorded at sampling sites. Results Serotiny was distributed bimodally within and among populations with capsules either mostly closed or mostly open. There was considerable geographical variation in capsule behaviour, most of which we attribute to variation in fire history. In wetlands and other sites unsuitable for forest growth, populations were all serotinous above a minimum area of 30 km2 and nonserotinous below this threshold. In grassy habitats in the drier eastern areas, most populations were serotinous. The nonserotinous exceptions occurred in areas thought to have been cleared of forests by Polynesian settlers before the arrival of Europeans or in areas with numerous barriers to fire in the form of large rivers, floodplains, glaciers and barren mountain tops. Conclusions We suggest that serotiny in L. scoparium is a reflection of a long history of fire in the South Island. As such, it provides a contemporary signature of a past fire regime. Landscape barriers to the spread of fire were major obstacles limiting the spread of serotiny and associated fire-dependent vegetation. Rivers, lakes, glaciers, and sparsely vegetated floodplains prevented the spread of fire in high rainfall regions more than the lack of dry weather. People, by igniting fires in small open areas seldom struck by lightning, could radically increase the importance of fire on islands.

Aim We assess the importance of three relevant and readily obtainable lifehistory traits (dispersal syndrome, stem height and growth form) and biogeographical origin (European vs. non-European) on the local and regional abundance of over 400 exotic plant species across eight Mediterranean islands. Location The Mediterranean islands of Lesbos, Rhodes, Crete, Malta, Corsica, Sardinia, Majorca and Minorca. Methods We adopt two abundance criteria for each exotic species: the proportion of islands in which the species occurs (regional abundance), and a qualitative estimate of species abundance within each of five islands (local abundance). Subsequently, we assess the relationship between local and regional abundance, as well as the role of key life-history traits on both regional and local abundance. These analyses were undertaken separately for the European exotics and the non-European exotics. Results Only 10.9% of the species occur on more than four islands, and only four species are present on all eight islands. Both local and regional abundances were higher for the non-European than the European species. Local and regional abundances were positively correlated, particularly for exotics with non-European origins. Wind-dispersed species tended to have higher regional abundance than species dispersed by other means but this trend only occurred for local abundance on two islands - Corsica and Majorca. Neither a species' growth form nor its stem height explained trends in regional or local abundance. Conclusions Although wind-dispersed exotics are more widespread in the Mediterranean, plant life-history traits appear to play a lesser role in invasion success than area of biogeographical origin. In general, exotic species of non-European origin were more abundant at both local and regional scales. Invasion patterns should be interpreted at both local and regional scales, but the stochastic nature of biological invasions may limit deterministic interpretations of invasion patterns, especially if islands are studied in isolation.

Aim We evaluated how an elevation gradient affects: (1) the availability of food required by a specialist seed-eater, Loxioides bailleui Oustalet (Drepanidinae), or palila, and hence the distribution of this endangered Hawaiian bird, and (2) the distribution of alien threats to Loxioides populations, their primary foods, and their dry-forest habitat, and hence strategies for their conservation. Location We worked throughout the subalpine forest that encircles Mauna Kea Volcano, Hawai‘i Island, but we focused our studies mainly on the western slope between 2000 and 3000 m elevation, where the gradient of elevation was greatest and palila were most abundant. Methods We determined phenology and productivity patterns of the endemic dry-forest tree species, Sophora chrysophylla (Salisb.) Seem. (Fabaceae), or m amane, which provides Loxioides with most of their food, and another common endemic tree, Myoporum sandwicense A. Gray (Myoporaceae), or naio, which provides some resources, along a 786-m elevation gradient at monthly intervals for 10 years (Sophora only). We also determined the availability each month of moth larvae (Lepidoptera) for that were important in the diet of nestling and adult palila. In addition, we documented the incidence of parasitism on moth larvae by several wasp (Hymenoptera) and fly (Diptera) species, and we determined the distribution of predatory wasps and ants (Hymenoptera), which potentially threaten insect prey of birds. Percentage cover of alien grass species that pose fire threats in palila habitat and other weeds were assessed during one survey. Small mammal abundance and distribution were determined by trapping during three (rodent) or five (carnivore) surveys. Results Sophora flower and seed (pod) availability varied predictably along the elevation gradient, with about 4 months separating peaks in reproduction at high and low elevations. This, together with highly variable production of flowers and pods within elevation strata, resulted in Sophora resources being available to Loxioides throughout the year on the western slope of Mauna Kea. Sophora produced flowers and pods more seasonally where gradients of elevation were short; thus, resources were available less consistently. In contrast, Myoporum produced flowers and fruits with little variation with respect to season or elevation. The availability of important insect prey of Loxioides was also related to elevation, in part because threats to Lepidoptera larvae from parasitic wasps were generally less at higher elevations. Threats to insect prey from predatory ants was also less at higher elevations but the abundance of predatory wasps was not related to elevation. Several weeds that pose the most serious threats to Loxioides habitat were more abundant at mid and low elevations, and alien grass cover was somewhat greater at mid elevation, thereby increasing fire risks in the

Allochthonous marine input is a key component of the dynamics of islands and terrestrial coastal ecosystems on islands in the Gulf of California, Mexico, where an unproductive desert juxtaposes a highly productive ocean. In this area, seabirds are a major conduit bringing marine productivity to land. Seabirds leave guano on roosting and nesting islands, and carrion on nesting islands. We analyze how seabirds affect the abundance of a dominant consumer group, tene- brionid beetles, on 25 islands in the Gulf of California over a three-year period. Tenebrionid densities vary by three orders of magnitude among islands. Beetles are most abundant in areas influenced by seabirds: they are approximately five times more dense on nesting and roosting islands than on other islands and on mainland sites, and approximately six times more dense inside vs. outside colonies. Path analyses show that seabirds significantly affect beetles by two distinct pathways. On roosting islands, effects are mainly indirect: guano, a fertilizing nutrient, significantly enhances plant productivity; beetles eat plant detritus. On nesting islands, significant effects are primarily direct: beetles eat seabird carrion; the indirect pathway (guano and plants) contributes little on these islands. By providing energy and nutrients to fuel a diverse array of consumer populations, seabirds are central to the dynamics of these island ecosystems. Scavengers and avian parasites directly increase by eating seabird products. Likewise, populations of consumers that eat detritus, plant tissues, and seeds are enhanced indirectly via the fertilizing effects of guano on plants. Increases in these primary consumers indirectly facilitate high densities of many predators. Thus, consumer populations on these islands are deeply subsidized by substantial input from seabirds. Because consumers cannot influence the renewal rate of their allochthonous resources, the dynamics of these consumers and their food webs are largely donor controlled.

Summary Aim Our knowledge of landbirds on tropical Pacific islands is often comprised of brief, one-time surveys. We know little of species’ habitat preferences, and this information is critical for understanding the resource requirements or population status of native species and the impacts of human activity on island birds. LocationTutuila Island, American Samoa. Methods The spatial patterns in forest birds were investigated. This island harbours large tracts of native forest, a variety of disturbed and non-native forests, and some relatively healthy bird populations. The distribution of birds was correlated with forest types using habitat data collected at fifty-seven sites and avian census data collected monthly for 1–4 years at these sites. ResultsDifferences in location and in vegetation structure and composition between native and non-native habitats are important influences on the distribution of birds on Tutuila. Among native species, for example, the purple-capped fruit-dove (Ptilinopus porphyraceus Temminck) is dependent upon native habitat, the Samoan starling (Aplonis atrifusca Peale) occurs in all habitats, and the cardinal honeyeater (Myzomela cardinalis Gmelin) is more abundant in low elevation, non-native habitat. ConclusionsThis research reinforces the importance of quantitative assessment of habitat relationships in the study and conservation of Pacific birds.

This paper reports on a multivariate analysis of a large data set on the occurrence of 900 vascular plant species in over 1500 localities in the Southwest Finnish archipelago, which was collected by the Finnish botanist Ole Eklund. Four major types of islands are distinguished as to island size and degree of human impact. Subdivisions are based mainly on contrasts between the presence or absence of limestone in the bedrock, and of wetlands. The major floristic gradients could be related to the following factors: (1) decreasing island size and degree of human impact and increasing maritime influence; (2) increasing occurrence of limestone; (3) a trend from open habitats with high insolation to forested habitats; and (4) increasing amount of wetlands and rock pools. For the flor- istic characterization of island types a system of socio-ecological species groups was developed based on Ellenberg's indicator figures for the flora of Central Europe. The old controversy, as to the explanation of the distribution of plant species in the archipelago, between Palmgren (‘dispersal is the limiting factor’) and Eklund (‘environmental differentiation is decisive’) is discussed and related to the scale of observation. Our results suggest that environmental differentiation is of prime importance, but on the local scale the species composition of islands may be partly governed by dispersal processes.

Since the biologically extirpating eruption of Krakatau (Sunda Strait) in 1883, Rakata (Krakatau's remnant) and two closely adjacent islands, Sertung and Panjang, have been colonized by over 200 species of vascular plants. They now carry species-poor mixed tropical forest, including some twenty-three species of Ficus. Data on the sequence of colonization over the last century by twenty- four Ficus species, twenty-three species of volant frugivores, and by agaonid fig-wasps, presumably from the large islands of Java and Sumatra, each some 44 km distant, are summarized. The potential of the volant frugivores as dispersers of fig seeds is assessed, the pollination problems involved in the colonization of islands by figs are reviewed and patterns of colonization by fig species and by their bird and bat dispersers are identified and discussed. In 1930 a new island, Anak Krakatau, emerged from Krakatau's submerged caldera. This active volcano suffered a self-sterilizing eruption in 1952/1953 and has been colonized, under considerable constraint from its own volcanic activity, probably largely from the (selected) species pool present on Rakata, Sertung and Panjang, 2-4 km away. Its vegetation is at an earlier successional stage (grassland and Casuarina woodland) than that of the three older islands, and in 1992 the Casuarina woodland was in an early stage of transition to mixed forest. The colonization of Anak Krakatau by Ficus species, agaonid wasps and volant frugivores over a critical decade (1982-92) is reviewed, including preliminary assessments of the effects of pollinator limitation on four pioneer fig species and indications of a possible effect of the presence of avian raptors, particularly the peregrine falcon, on fig colonization and forest diversification.

Scale, the scale dependency of patterns and processes, and the ways that organisms scale their responses to these patterns and processes are central to island and landscape ecology. Here, we take a database of studies in island ecology and investigate how studies have changed over a 40-year period with respect to spatial scale and organisms studied. We demonstrate that there have been changes in the spatial scale of islands studied and that there is taxonomic bias in favour of vertebrates in island ecological studies when compared to scientific publications as a whole. We discuss how such taxonomic bias may have arisen and discuss the implications for ecology and biogeography.

Dwarfed and deformed plants of tree species occur as "tree islands" in a matrix of tundra vegetation near the upper limit of tree growth in the Southern Rocky Mountains. Many of these islands are asymmetrical, having a surface layer of dead branches to windward and some live branches to leeward. I made repeated observations of these islands in all seasons from 1956 to 1967 along a 1.75-km belt -20 m wide on Niwot Ridge in the Front Range west of Denver, Colorado. The location of islands was mapped on a transparent overlay on an aerial photograph, annual growth of stem tips was measured, and an abundance of notes and photographs was taken. Some tree islands of Engelmann spruce (Piteea engelmnannii) and subalpine fir (Abies lasiocarpa) are conspicuous because they grow in microsites in which no tree seedlings have been found, and which general observations suggest would be outside their environmental amplitude. These islands often have live branches connected to windward with dead branches, some of which have roots. These observations led to the hypothesis that some tree islands move along the ground climbing out of more moderate microsites in which a seed germinated and moving into microsites for which they are not well adapted. In this movement process, wind-exposed organs are killed, while leeward stems grow to leeward often taking root and forming an upright stem system in the process of "stem tip layering." Islands with interconnected live and dead stems to windward, illustrating movements of 5 m, are common; isolated dead stumps and wood fragments in a line extending to windward indicate that some islands may have moved at least 15 m. The origin of tree islands, support for the movement hypothesis, and other features of tree island ecology are discussed.

To examine effects of pollinator diversity on pollination biology of plants, we studied early successional, hummingbird-pollinated plants for 13 mo on Trinidad and Tobago. A large, land- bridge island just off the Venezuelan coast, Trinidad has 16 hummingbird species. Smaller and more isolated, Tobago has a subset of five of these. Study sites in the Arima Valley on Trinidad had 17 species of plants adapted for hummingbird pollination; 12 of these, plus one other, existed on com- parable Tobago sites. From hypotheses of island ecology and plant-pollinator coevolution, we developed four predic- tions; results supported the first three: (1) Overall, among the eight plant species observed at least 4 d (up to 28 d) on each island, Tobago flowers received significantly fewer visits than nonspecific Trinidad flowers from hummingbirds and the Bananaquit Coereba flaveola, the only other avian visitor. A two-way ANOVA showed that the treatment (island) effect significantly exceeded even the tremendous month-to-month, plant-to-plant variation in visits received within each plant population. (2) Pollen loads carried by hummingbirds and Bananaquits on Tobago tended to be mixed, whereas birds on Trinidad often carried single-species loads. (3) In two of three plant species studied inten- sively, Tobago flowers secreted greater volumes of nectar than Trinidad flowers. The third species showed a slight but significant increase as well. An indirect measure of nectar standing crop, the time a hummingbird spent probing a flower, suggested that Tobago plants as a group secreted higher nectar volumes than Trinidad conspecifics. We found no significant differences in sugar concentration, how- ever. (4) Contrary to predictions that flowering seasons should overlap less on Tobago than on Trinidad, there was no discernible difference in flowering seasons between Trinidad and Tobago conspecifics. Evidently, pollination service on Tobago is currently more erratic than on Trinidad, and has been sufficiently erratic in the past to select for increased nectar production but not so erratic as to override other, stabilizing, selective forces on flowering times.

The biotas of islands, long a favorite province of biogeographers and evolutionary biologists, have received much attention within the last few years. For example, Carlquist (1965, 1966a, 1966b, 1966c, 1967a, 1967b) has considered many aspects of insular biology including thel dispersal mechanisms which' transport disseminules to islands, the effect of island ecology on establishment, and the evolutionary phenomena that are evident once species have become established. Among the subjects which interested Carlquist as well as earlier students is the relationship between biotic diversity and both island size and distance from the source region. For most island systems one can detect a typical area-diversity pattern, with larger islands supporting the richest biota. Also, evident is an inverse relationship between species number and distance from the source. Islands that are far removed have fewer species than similar-sized islands that are more proximate. Thus the biota of an island reflects the biotic diversity of the source region, the size of the island, and the distance from the source. More recently the problem of insular biogeography has been dealt with differently. MacArthur and Wilson (1963, 1967) have suggested that the relationships between diversity, size, and distance can be approached more mathematically. Th'ey have proposed that the biota of an island contains an equilibrium number of species, with the equilibrium number controlled by a balance between immigration rate and extinction rate. The slop,e of the immigration curve is distance-dependent, implying that the immigration rate for far islands is low. The extinction curve is area dependent, with a steeper slope (hence higher rate) on small islands. Both rates are controlled by the number of species present on the island. These relationships are illustrated in Figure 1. From this figure it can be seen that as the biota of the island approaches the equilibrium number, the immigration rate drops and the extinction rate rises.. The point of intersection of the curves indicates the "equilibrium number" of species or that fraction of the "species pool" represented at equilibrium. On large islands the slope of the immigration curve will be steeper, hence the immigration curve will intersect the extinction curve further to the right and the equilibrium number will increase. As distance from source increases (and immigration rate consequently decreases), species number decreases because the immigration curve will intersect the extinction curve further to the left. The contributions of both Carlquist's and MacArthur and Wilson's work are significant. The phenomena of dispersal, immigration, colonization, extinction, and speciation which attend the insular syndrome have been more clearly defined, new approaches have been suggested, and much discussion has been engendered. We have found it interesting to examine certain aspects of the equilibrium model in the light of Carlquist's data on dispersal and the ecological factors which affect establishment (1965, 1967a). Some inconsistencies are evident. In addition, it appears to us that the application of the model can provide erroneous conclusions in certain instances. Thus in the present paper we would like to deal with some of these problems, specifically: (1) the atypical situation presented by exceptionally small is