Sequential conflicting selection due to multispecific interactions triggers evolutionary trade-offs in a monocarpic herb

Trade-offs are crucial in understanding phenotypic evolution of organisms. A main source of trade-offs is conflicting selection, a phenomenon very likely in complex multispecific scenarios in which many potential selective agents coexist. The main goal of this study is to investigate the selective trade-offs arising due to conflicting selection on female-fitness components in Erysimum mediohispanicum. I quantified the selection exerted on 10 plant traits by a mutualistic (pollinators) and antagonistic (gall-makers, predispersal and postdispersal seed predators, mammalian herbivores) multispecific assemblage acting sequentially throughout eight selective episodes of the plant, from floral bud to juvenile production. Variation in lifetime female fitness (quantified as number of juveniles) was related mostly to variation in number of flowers, fruit initiation, and seedling establishment. The direction of selection changed among different selective episode for many traits. Most importantly, conflicting selection was frequent in the study system, with half of the phenotypic traits experiencing opposing selection in different selective episodes. Selection at individual life-cycle stages diverged remarkably from selection based on total fitness. Consequently, the evolution of many traits is determined by the relative importance of each episode of selection, with conflicting selection inevitably yielding evolutionary compromises.     

Effects of herbivore identity on plant fecundity

Herbivores are pervasive, yet their effects on plant fecundity are often variable. One potential source of variation in herbivore impacts results from differing feeding modes of herbivores. We examined the relative importance of inflorescence-feeding insects versus large ungulates in affecting the fecundity of Balsamorhiza sagittata (Asteraceae), a dominant native perennial forb in western Montana, USA. We quantified these effects across four sites that varied in elevation to determine how consistent herbivore impacts were across these divergent sites. Insect herbivores were present in flower heads at all sites but they significantly depressed plant fecundity at two of the four sites. At the two sites where herbivore suppression had significant effects, this treatment let to 1.6 and 3.4-fold increases in seed production, respectively. In contrast, across all sites ungulates had minimal impacts on balsamroot flower and seed production. Seed addition experiments revealed that at some sites and in some years B. sagittata is seed-limited, as there was a positive relationship between seed input and seedling recruitment. Thus, reductions in seed production from heavy insect herbivory may limit balsamroot recruitment in some locations. Overall, results show that inconspicuous insects have stronger effects on balsamroot fecundity than do mammalian herbivores, but the magnitude of negative impacts at both the individual and population-level vary by site.     

Effects of two types of herbivores on the population dynamics of a perennial herb

The joint effects of multiple herbivores on their shared host plant have received increasing interest recently. The influence of herbivores on population dynamics of their host plants, especially the relative roles of different types of damage, is, however, still poorly understood. Here, we present a modelling approach, including both deterministic and stochastic matrix modelling, to be used in estimating fitness effects of multiple herbivores on perennial plants. We examined the effects and relative roles of two specialist herbivores, a pre-dispersal seed predator, Euphranta connexa, and a leaf-feeding moth, Abrostola asclepiadis, on the population dynamics and long-term fitness of their shared host plant, a long-lived perennial herb Vincetoxicum hirundinaria (Asclepiadaceae). We collected demographic data during 3 years and combined these data with the effects of natural levels of herbivory measured from the same individuals. We found that both seed predation and leaf herbivory reduced population growth of V. hirundinaria, but only very high damage levels changed the growth trend of the vigorously growing study populations from positive to negative. Demographic modelling indicated that seed predation had a greater impact on plant population growth than leaf herbivory. The effect of leaf herbivory was weaker and diminished with increasing level of seed predation. Evaluation of individual fitness components, however, suggested that leaf herbivory contributed more strongly to host plant fitness than seed predation. Our results emphasize that understanding the effects of a particular herbivore on plant population dynamics requires also knowledge on other herbivores present in the system, because the effect of a particular type of herbivory on plant population dynamics is likely to vary according to the intensity of other types of herbivory. Furthermore, evaluating herbivore impact from using individual fitness components does not necessarily reflect the long-term effects on total plant fitness.     

Seed predators limit plant recruitment in Neotropical savannas

Despite the well‐documented impacts of consumers on seed abundance the link between seed predation and plant population dynamics remains poorly understood because experimental studies linking patterns of predation with seedling establishment are rare. We used experimental manipulations with six woody plant species to elucidate the effects of seed predator type, habitat, and plant species identity on rates of seed predation and seedling recruitment in the Neotropical savannas known as the Cerrado. We found that seed predation rates are consistently high across a diversity of local habitat types, with important inter‐habitat variation in seed predation for three of the six species used in our experiments. We also found that seed predation has a clear demographic signal – experimentally excluding predators resulted in higher rates of seedling establishment over the course of two seasons. Because the intensity of seed predation varied between species and habitats, it may play a role in structuring local patterns of plant abundance and community composition. Finally, our results lend support to the recent hypothesis that herbivores have major and underappreciated impacts in Neotropical savannas, and that top–down factors can influence the demography of plants in this extensive and biodiversity‐rich biome in previously unexplored ways.     

How does herbivory affect individuals and populations of the perennial herb Paeonia officinalis?

In this study we quantified variability in foliage herbivory and pre-dispersal seed predation and its effects on plant performance and demography in populations of a rare and protected perennial herb, Paeonia officinalis. An individual-based survey was performed during four years in four populations, which contained plants in both open habitat and woodland. We detected marked spatial and temporal variation among and within populations in foliage herbivory (by insects) and pre-dispersal seed predation (by insects, rodents and Roe Deer). Foliage herbivory decreased with plant demographic stages in open habitats, from seedlings to reproductive individuals, but no significant trend was detected in woodland habitats. This may be due to different demographic origin of larger vegetative plants in this habitat. Depending on demographic stage, herbivory was higher in open habitats or not significantly different between habitats. This suggests differences in herbivore abundance in different habitats within sites. Pre-dispersal seed predation remained weak and did not depend on habitat. We did not detect any consequence of foliage herbivory on seedling mortality and individual growth in our study. Our results illustrate the need to investigate plant-herbivore interactions over several years in distinct populations in order to more accurately evaluate herbivore impact on plant population dynamics.     

Population dynamics of the shrub Acacia suaveolens (Sm.) Willd.: Seed production and predispersal seed predation

Seed production and predispersal seed predation in the shrub Acacia suaveolens were examined over 3 consecutive years in eight populations in south-eastern Australia. Seed-crop sizes varied both between and within populations of different ages. Seed production was maximal in the first one to four flowering seasons after establishment, and then declined with plant age. The size of the annual seed-crop was also influenced by rainfall for that year. Predispersal seed predation varies between populations over fruiting seasons with the initial large seed-crops resulting in predator satiation. Within one fruiting season, no significant variation in the extent of predispersal seed predation was found in any of three populations studied. There were two major forms of predispersal seed loss: toss of whole fruits to Melanterius corosus (Coleoptera: Curculionidae) and external insect seed grazers, and loss of individual seeds within fruits to M. corosus. Exclusion experiments showed that seeds lost to these predispersal seed predators would otherwise have been matured by the parent plant. The effects of predispersal seed predation cannot be directly related to seedling recruitment. Indirectly, such predation may influence the dispersion of seeds in the soil profile and hence, subsequent recruitment.     

Temporal variation in the female components of reproductive success over the extended flowering season of a Mediterranean perennial herb

We examined the reproductive success of the perennial herb Lobularia maritima during its extended flowering and fruiting season. The within- and between-year variability of the female components of reproductive success (from flower, fruit and seed production to seed survival, seed germination and seedling establishment) were analysed during four flowering seasons. All the components of reproductive success studied showed a significant within- and between-year variation. September was the period of the year with the maximum values of flower and fruit production, and the highest germination and establishment rates. Nevertheless, seed losses due to both predispersal and postdispersal seed predation during this period were also the highest, seriously reducing seed output in this period. On the other hand, in those periods in which seed production was low, i.e. January and May, the percentage of seeds lost to seed predators was the lowest. Reproductive success in each period of the flowering season was estimated using a simple demographic model, in which the information concerning all the components already calculated was integrated. The two variables used to estimate reproductive success in each period, i.e. the number of new individuals produced per plant and the probability of a seed becoming an adult plant, showed relatively small differences over the year. These results suggest a counter-balance of the different components of reproductive success in this species, with favourable and unfavourable periods for the different components being compensated during its extended flowering season.     

Propagule size and predispersal damage by insects affect establishment and early growth of mangrove seedlings

Variation in rates of seedling recruitment, growth, and survival can strongly influence the rate and course of forest regeneration following disturbance. Using a combination of field sampling and shadehouse experiments, we investigated the influence of propagule size and predispersal insect damage on the establishment and early growth of the three common mangrove species on the Caribbean coast of Panama: Avicennia germinans, Laguncularia racemosa, and Rhizophora mangle. In our field samples, all three species exhibited considerable intraspecific variation in mature propagule size, and suffered moderate to high levels of predispersal attack by larval insects. Rates of insect attack were largely independent of propagule size both within and among trees. Our experimental studies using undamaged mature propagules showed that, for all three species, seedlings established at high rates regardless of propagule size. However, propagule size did have a marked effect on early seedling growth: seedlings that developed from larger propagules grew more rapidly. Predispersal insect infestations that had destroyed or removed a substantial amount of tissue, particularly if that tissue was meristematic or conductive, reduced the establishment of propagules of all three species. The effect of sublethal tissue damage or loss on the subsequent growth of established seedlings varied among the three mangrove species. For Avicennia, the growth response was graded: for a propagule of a given size, the more tissue lost, the slower the growth of the seedling. For Laguncularia, the response to insect attack appeared to be all-or-none. If the boring insect penetrated the outer spongy seed coat and reached the developing embryo, it usually caused sufficient damage to prevent a seedling from developing. On the other hand, if the insect damaged but did not penetrate the seed coat, a completely healthy seedling developed and its growth rate was indistinguishable from a seedling developing from an undamaged propagule of the same size. Similar to Avicennia, if an infestation did not completely girdle a Rhizophora seedling, it survived, but grew at a reduced rate. In summary, our experiments demonstrated that natural levels of variation in propagule size and predispersal damage by insects translate into significant differences in seedling performance in terms of establishment and/or early growth. Such differences are sufficiently large that they could influence the intensity and outcome of competitive interactions during forest regeneration.     

Phenological synchrony affects interaction strength of an exotic weevil with Platte thistle,a native host plant

Phenological synchrony of a consumer population with its resource populations is expected to affect interaction intensity. We quantified phenological variation and synchrony of populations of an invasive Eurasian flower head weevil, Rhinocyllus conicus, that consumes florets, ovules, and seeds of developing flower heads of a native North American thistle, Cirsium canescens, in Sand Hills prairie in Nebraska, USA. Variation in timing of adult activity among weevil populations was larger than variation in timing of flower head development among C. canescens populations, and it drove the observed variation in the phenological synchrony between weevil and host plant populations. Furthermore, the degree of phenological synchrony between populations was significant in explaining variation in weevil egg load on the newly acquired host plant. Because population growth of C. canescens is limited by predispersal seed losses to floral herbivores, variation in the synchrony of herbivore and plant flowering will affect the density of the plant population. These results provide strong quantitative support for the hypothesis that the synchrony of insect activity with plant resources can determine the magnitude of impact of floral herbivores on their host plant populations.     

Spatiotemporal variation in predispersal seed predation intensity

The effect of predispersal seed predation by Bruchus atomarius (Bruchidae, Coleoptera) on individual performance and population dynamics of the perennial forest herb, Lathyrus vernus (Leguminosae), was investigated in 11 permanent plots over 4 years. Seed predation and parameters describing intra-specific neighbour distance, plant size, inflorescence size, flowering phenology and current and previous herbivore damage were measured on all plants. In addition, demographic information from all plots was analysed using transition matrix population models in order to estimate the influence of seed predation on population growth rates. Predispersal seed predation rates differed significantly among years. Plot averages ranged from 0 to 83.7%. However, most of the variation occurred among individuals. Within individuals there was no consistency in predation rates among years. Exposure to herbivory, plant size and flowering phenology did not affect predation rates but individuals with larger inflorescences suffered from significantly higher predation. Seed predation in L. vernus was not influenced by neighbour distances of individual plants but it was positively correlated with the average density of seeds within plots, suggesting that seed predation is density dependent at the patch level. The reduction in population growth rate due to seed predation ranged from 0 to 7.6%. The sensitivity of population growth rate to reductions in seed production varied considerably among years and plots. This variation was mainly due to differences in the reproductive value of seeds and seedlings. The intensity of seed predation over the range found was not correlated with changes in population growth rate. The results of this study suggest that the influence of external factors, like seed predation, on population growth rate largely depends on the demographic transition rates in the investigated population.     

Effects of intraspecific and interspecific density on the demography of a perennial herb, Sanicula europaea

We examined the effects of intraspecific and interspecific competition on demographic processes in the perennial herb Sanicula europaea by manipulating the density of neighbouring plants. We followed the response in terms of survival, growth and reproduction and in terms of seedling recruitment. The demographic data from all phases of the life cycle enabled us to assess also the overall effects of treatments on population growth rate (λ) by transition matrix models. We also decomposed the differences in λ between control and treatments, using life table response experiments (LTRE). To study the effects of competition on recruitment in more detail and to evaluate the role of seed availability, we sowed seeds at different densities with or without vegetation removal.
Vegetative growth and flowering frequency of established individuals was not significantly affected by removal treatments, which suggest no, or a delayed response to released competition. Neighbour removal had no effect on seedling emergence but enhanced recruitment through a higher seedling survival. Conspecific and simultaneous conspecific and heterospecific removal of plants led to an increase in population growth rate (λ), whereas heterospecific removal alone led to a decrease. Emergence of seedlings and fate of vegetative established individuals contributed most to differences in λ between the control and the different treatments. Seed addition enhanced seedling emergence but, as seedling and juvenile survival were density dependent, densities of established individuals appear not to be seed limited.
In S. europaea removal treatments had different effects on established individuals and recruitment. This suggests that studies quantifying the effects of competition over the entire life cycle and performed in a natural environment are necessary to assess the importance of competition in perennial plant populations.     

Influence of herbivores on a perennial plant: variation with life history stage and herbivore species

Herbivores have diverse impacts on their host plants, potentially altering survival, growth, fecundity, and other aspects of plant performance. Especially for longer-lived plant species, the effects of a single herbivore species can vary markedly throughout the life of the host plant. In addition, the effects of herbivory during any given life history stage of a host plant may also vary considerably with different types of herbivores. To investigate the effects of herbivory by black-tailed deer (Odocoileus hemionus columbianus) and snails (Helminthoglypta arrosa and Helix aspersa) on a nitrogen-fixing shrub, Lupinus chamissonis, we established three exclosure experiments in a sand dune system on the coast of northern California. These experiments documented that deer browsing significantly reduced the volume and growth rate of lupines in the seedling and juvenile life stages. Since plant volume was strongly correlated with aboveground dry biomass for lupines, such herbivore-induced reductions in volume should translate into losses of aboveground biomass. Deer browsing also significantly altered the likelihood of attack by and density of a leaf-galling cecidomyid fly (Dasineura lupinorum), suggesting that a vertebrate herbivore indirectly affected an invertebrate herbivore in this system. Although deer did not significantly affect the survival of lupine seedlings and juveniles, individuals protected from deer had consistently greater survival in the two separate experiments. Our results revealed that snails did not have a significant effect on the survival or growth of juvenile plants, despite being common on and around lupines. An exclosure experiment revealed that herbivory by deer significantly reduced the shoot lengths of mature shrubs, but led only to a minimal reduction in growth rates. In addition, we found that browsed shrubs had significantly greater inflorescence production, but also produced individual seeds with significantly reduced mass. Collectively, these data indicate that deer and snails have widely differing effects on their shared host plant; browsing by deer indirectly affects insect herbivores, and the impacts of deer change markedly with the life history stage of their host plant.     

Plant damage from and defenses against ‘cryptic’ herbivory: A guild perspective

Abstract

Despite considerable interest in the factors affecting trophic cascades in terrestrial systems, there has been relatively little attention paid to the importance of the herbivore-plant link in explaining why some systems “cascade” (have strong top-down effects on plant survival and population growth) and others “trickle” (have top-down effects on plant damage, but little effect on plant fitness). This is despite the fact that herbivore guild identity has long been recognized as a major force affecting herbivore-plant interactions. We address the potential importance of herbivore guild identity in determining the strength of tritrophic interactions by reviewing literature concerning plant damage from and induced defenses against two “cryptic” herbivore guilds, predispersal seed predators and root/stem borers. Although both guilds are capable of strongly affecting plant fitness, the impact of root/stem borers on plants in natural systems seems far greater than that of predispersal seed predators. The large impact of root/stem borers occurs via their disruption of plant vascular systems, while a variety of factors (safe-site-limited plant populations, long-lived seed banks, temporal plant escape, etc.) each seem important in explaining the smaller effect of predispersal seed predators. While the lack of attention to herbivore guilds is understandable, given the (by necessity) single-species focus of much trophic cascade research, we suggest that predator suppression of root/stem borers and predispersal seed predators will, respectively, yield strong versus weak top-down effects on plant fitness. The potential tritrophic consequences of herbivore feeding mode highlight the importance of research on varied predator-herbivore chains that share a common basal resource.     

Choice of voles among genotypes of birch seedlings: its relationship with seedling quality and preference of insects

Selective feeding by herbivores on establishing seedlings has been suggested to affect genotype frequencies in several plant populations. The existence of genotypes susceptible to herbivores calls for an explanation in such populations. In the present study we assessed the choice of multiple herbivores, field voles (Microtus agrestis) and insects, among genotypes of silver birch (Betula pendula) representing variation occurring in a naturally regenerated stand. We examined how food choice of voles and insects is related to each other, competitive ability among the seedling genotypes and variation in soil fertility. We set up a field experiment and randomly assigned seedling populations, composed of mixed genotypes, to fully crossed insect exposure and fertilization treatments. After the first growing season we exposed a half of the seedling populations to vole herbivory. Voles selected clearly among the genotypes: they preferred the fastest growing seedlings as well as those with a low density of resin droplets on their stems. The preference of voles and insects among the genotypes was tightly correlated. We conclude that the effects of herbivory compensate those of intraspecific competition in this system and thus favor coexistence of genotypes differing in their susceptibility to herbivores.     

Thermal tolerance affects mutualist attendance in an ant–plant protection mutualism

The influence of native fauna on non-native plant population growth, size, and distribution is not well documented. Previous studies have shown that native insects associated with tall thistle (Cirsium altissimum) also feed on the leaves, stems, and flower heads of the Eurasian congener C. vulgare, thus limiting individual plant performance. In this study, we tested the effects of insect herbivores on the population growth rate of C. vulgare. We experimentally initiated invasions by adding seeds at four unoccupied grassland sites in eastern Nebraska, USA, and recorded plant establishment, survival, and reproduction. Cumulative foliage and floral herbivory reduced C. vulgare seedling density, and prevented almost any reproduction by C. vulgare in half the sites. The matrix model we constructed showed that this herbivory resulted in a reduction of the asymptotic population growth rate (λ), from an 88 % annual increase to a 54 % annual decline. These results provide strong support for the hypothesis that indigenous herbivores limit population invasion of this non-native plant species into otherwise suitable grassland habitat.     

Bigger is not always better: conflicting selective pressures on seed size in Quercus ilex

Most theoretical treatments of the evolutionary ecology of offspring size assume a simple and direct effect of investment per offspring on offspring fitness. In this paper I experimentally determine the relationship between seed mass and several main fitness components of the oak Quercus ilex, to estimate phenotypic selection acting on seed mass during the early life cycle and to discover any potential selective conflicts occurring between different stages from dispersal to establishment. I found a positive effect of acorn size on most fitness components related to seedling establishment. Large size increased germination rate and seedling survival, accelerated germination timing, and enhanced seedling growth. Nevertheless, there was also a direct negative effect of acorn size on survival to predation, because large acorns were highly preferred by the main postdispersal seed predators at the study site, wild boars and wood mice. Because of the low probability of escape from predation, the fitness of large acorns estimated on this component was significantly lower than the fitness of smaller acorns. Therefore, seed size affected fitness in two different ways, yielding opposing and conflicting selective forces. These findings suggest that the general assumption that offspring fitness is a fixed positive function of seed size needs to be reconsidered for some systems. The existence of conflicting selection might explain the occurrence of an optimal seed size in some plant species without invoking a seed number-size trade-off.     

Pollen limitation, seed predation and scape length in Primula farinosa

Floral display and reward production may affect the attractiveness of a plant to a range of interacting animals including pollinators, herbivores, and vectors of pathogenic fungi. The optimal floral phenotype should therefore depend on the relative importance of selection exerted by both mutualistic and antagonistic animals. The perennial, rosette herb Primula farinosa is polymorphic for scape length. Natural populations may include both plants with flowers displayed well above the ground (the long-scaped morph) and those with flowers positioned very close to the ground (the short-scaped morph). In this study, we conducted a field experiment to examine how the relative fitness of the two scape morphs is affected by interactions with pollinators and fruit predators in two different microhabitats (high and low vegetation). As predicted based on the difference in floral display, supplemental hand-pollination showed that fruit initiation was more strongly pollen-limited in the short-scaped than in the long-scaped morph, and that this difference was significantly larger in high than in low vegetation. Moreover, plants with a short scape experienced lower levels of fruit predation than plants with a long scape. Among open-pollinated controls, there was no significant difference in seed output between the two scape morphs. However, among plants receiving supplemental hand-pollination, short-scaped plants produced significantly more seeds than long-scaped plants. The results suggest that the positive and negative effects of a prominent floral display (increased pollination and seed predation, respectively) balance in the study population, but also that the short-scaped morph would have an advantage at higher pollination intensities. Spatial and temporal variation in pollinator activity and seed predation should result in associated variation in the relative fecundity of the two scape morphs.     

Differential Effects of Hunting on Pre-Dispersal Seed Predation and Primary and Secondary Seed Removal of Two Neotropical Tree Species

Many of the mammals undergoing drastic declines in tropical forests worldwide are important seed dispersers and seed predators, and thus changes in mammal communities due to hunting will affect plant recruitment. It has been hypothesized that larger-seeded species will suffer greater reductions in seed removal and thus greater increases in predispersal seed predation than smaller-seeded species. We compared primary and secondary seed removal and predispersal seed predation of two tree species between hunted and nonhunted sites in Central Panama. Seeds of Oenocarpus mapora (Arecaceae) are 16-times greater in size than those of Cordia bicolor (Boraginaceae). We quantified primary seed removal and predispersal seed predation using seed traps, and we assessed secondary seed removal using seed removal plots. Primary removal of C. bicolor was 43 percent lower in the hunted sites, while primary removal of O. mapora was not significantly different. Secondary removal of unprotected O. mapora seeds on the ground was 59 percent lower in hunted sites, while secondary removal of C. bicolor was not significantly different. Predispersal seed predation of O. mapora by mammals was significantly lower in hunted sites, while predispersal seed predation by insects was not significantly different in either species. In combination with other studies, our results suggest that seed size is not a reliable predictor of the impacts of hunting. Mammal defaunation differentially affects stages and modes of seed dispersal and seed predation of different plant species, suggesting that these influences are complex and related to multiple plant traits.     

Order of herbivore arrival on wild cabbage populations influences subsequent arthropod community development

In plant–arthropod associations, the first herbivores to colonise a plant may directly or indirectly affect community assembly on that particular plant. Whether the order of arrival of different arthropod species further modulates community assembly and affects plant fitness remains unclear. Using wild Brassica oleracea plants in the field, we manipulated the order of arrival of early‐season herbivores that belong to different feeding guilds, namely the aphid Brevicoryne brassicae and caterpillars of Plutella xylostella. We investigated the effect of herbivore identity and order of arrival on community assembly on two B. oleracea plant populations during two growth seasons. For this perennial plant, we evaluated whether foliar herbivory also affected herbivore communities on the flowers and if these interactions affected plant seed production. Aphid infestation caused an increase in parasitoid abundance, but caterpillars modulated these effects, depending on the order of herbivore infestation and plant population. In the second growth season, when plants flowered, the order of infestation of leaves with aphids and caterpillars more strongly affected abundance of herbivores feeding on the flowers than those feeding on leaves. Infestation with caterpillars followed by aphids caused an increase in flower‐feeding herbivores compared to the reversed order of infestation in one plant population, whereas the opposite effects were observed for the other plant population. The impact on plant seed set in the first reproductive year was limited. Our work shows that the identity and arrival order of early season herbivores may have long‐term consequences for community composition on individual plants and that these patterns may vary among plant populations. We discuss how these community processes may affect plant fitness and speculate on the implications for evolution of plant defences.     

Effects of forest fragmentation on Heliconia acuminata seedling recruitment in central Amazonia

I present the results of a 2-year experiment comparing seed predation, seed germination, and seedling survivorship patterns of the Amazonian understory herb Heliconia acuminata in forest fragments and continuous forest. These empirical results were compared with natural patterns of recruitment in permanent 5,000 m² demographic plots adjacent to experimental areas. The number of naturally occurring seedlings established in demographic plots was 1.5-6 times greater in continuous forest than it was in 1-ha or 10-ha fragments. This result mirrors the pattern of seedling establishment in experimental transects, in which seeds in fragments were 3-7 times less likely to become established than those in continuous forest. Predation of experimentally sown seeds was extremely low at all sites, and is therefore not responsible for the observed pattern. Instead, reductions in seedling abundance in forest fragments are probably due to lower levels of seed germination. Forest fragments have higher air and soil temperatures, lower relative humidity, and increased leaf-litter accumulation, all of which can alter the cues used to initiate germination. While the growth of seedlings was similar in forest fragments and continuous forest, seedling survivorship in fragments was highly variable. These results suggest that altered environmental conditions may exacerbate reductions in plant recruitment resulting from modified plant-animal interactions. Strategies aimed at reducing the intensity of abiotic edge effects should therefore be incorporated into plant conservation efforts.