Beetle pollination of peacock moraeas (Iridaceae) in South Africa

The peacock moraeas are South African geophytes withIris-like flowers that are characterized by broad outer tepals with iridescent spots. Three of the seven species were studied and found to be exclusively visited and pollinated by beetles. The primary pollinators were hopliine (Scarabaeidae) beetles in the generaPeritrichia, Monochelus, Anisonyx andAnisochelus. These beetles visit flowers for nectar, pollen and mating.Peritrichia rufotibialis was the primary pollinator ofMoraea villosa, P. abdominalis was the primary pollinator ofM. tulbaghensis, whileMonochelus sp. nov. was the primary pollinator of the third species,M. neopavonia. High levels of fruit set attest to the effectiveness of these beetles as pollinators. Average fruit set ranged from 83.5 ± 34.8% (M. villosa) to 97.2 ± 11.6% (M. tulbaghensis). Average seed set in the various populations was more variable, but was also quite high, ranging from 77.3 ± 40.8 (M. villosa) to 342.2 ± 115.0 (M. neopavonia) seeds per capsule. Although hopliine beetles are abundant and diverse in southern Africa and commonly visit flowers, this is the first well-documented report of pollination by these beetles.     

HasIxianthes (Scrophulariaceae) lost its special bee?

Ixianthes retzioides is a rare, moderately sized, shrub with large yellow flowers that are lined with glandular trichomes. Preliminary analysis indicates that the trichomes secrete a complex mixture of lipids. Floral morphology suggests thatIxianthes should be pollinated by a large oil-collecting bee, yet only one possible candidate,Rediviva gig, is known and it occurs outside of the distributional range ofIxianthes. Despite observations ofIxianthes at different sites in different years, no oil-collecting bee has been observed visiting the flowers. This together with low fruit and seed set at the main study site suggests thatIxianthes has lost, at least locally, its specialized pollinator. Plants are self-compatible and set a limited amount of seed as a result of visits by pollen-collecting bees.     

Effects of population size and pollinator limitation on seed-set of Primula sieboldii populations in a fragmented landscape

Female reproductive success, fruit and seed-set and factors affecting them were measured for 3 years (1995–1997) for 24 populations of Primula sieboldii E. Morren, a threatened Japanese plant species in a highly fragmented landscape in southern Hokkaido. The smaller populations (genets <=; 3) set almost no seeds consistently during the study years. In larger populations (genets >= 7), proportions of failed flowers without damage from fungi or herbivores were relatively high, but fruit set per flower varied among the populations in 1995. Correlations of fruit (r = 0.589, P = 0.011) and seed- set (r = 0.688, P < 0.01) with population pollinator availability were highly significant. In 1995, pollination failure would be the most plausible reason for the variation in reproductive success among the populations. However, pollinator availability was relatively high in all populations in 1996 and 1997. In these years, higher proportions of fruit failed because of antagonistic biological interactions. In the populations with poor pollinator availability in 1995, mean fruit and seed-sets were greater in the long-styled morph than in the short-styled morph. Such between-morph differences are thought to be caused by partial self-compatibility of the long-styled morph and the results of hand pollination experiments support this hypothesis.     

Plant interactions for pollinator visits: a test of the magnet species effect

Summary From 1985–1987, patterns of fruit and seed set were studied in a population of mayapple (Podophyllum peltatum), a clonal, self-incompatible herb found in deciduous woods in eastern North America. Mayapple flowers do not produce nectar, but depend on infrequent visits by nectar-seeking queen bumble bees for pollination. In all years female reproductive success in mayapple colonies was influenced by colony size (number of flowers), by the distance to neighbouring colonies and by proximity to lousewort plants (Pedicularis canadensis), a prolific nectar producer heavily visited by bumble bees. In all years fruit and seed set were greater in mayapple colonies <25 m from lousewort flowers than in matched colonies which were >50 m from lousewort. In 1985 and 1987 the frequency of queen bumble bee visits to flowers in colonies close to lousewort was about four times greater than to distant colonies. In 1986 I removed about 80% of lousewort flowers to test whether the enhanced fruit and seed set in mayapples close to lousewort was pollinator mediated. Mayapple colonies close to flowerless lousewort patches did not differ in fruit or seed set from matched colonies >50 m from lousewort. In contrast, mayapples close to flowering lousewort patches had greater fruit and seed set compared with distant colonies. Over all years, a larger proportion of mayapples close to flowering lousewort patches had enhanced fruit and seed set compared with colonies close to louseworts without flowers. Though rarely documented, this type of facilitative interaction between plants that are highly attractive to pollinators (magnet species), and co-flowering species that are rarely visited by pollinators, may be widespread in plant communities.     

多叶斑叶兰繁育系统与传粉生物学研究

为探讨多叶斑叶兰(Goodyera folisa)的繁育系统与传粉生物学特征,对其开花物候、花粉活力与柱头可授性、人工授粉、花的挥发性成分以及昆虫传粉行为进行了研究。结果表明,多叶斑叶兰的唇瓣黄色,萼片白色或白色带红褐色;单花花期为(9.4±0.8)d;花粉在开花后第1天具备活力,柱头在开花后第2天具备可授性,花粉活力和柱头可授性都在开花后第5天达到峰值。去雄套袋和不去雄套袋都不能结实,人工自花授粉、同株异花授粉和异株异花授粉的结实率分别为93.3%、95.0%和96.7%,自然结实率为43.3%。花朵的主要挥发性成分为1-辛烯-3-醇、3-辛醇和N,N-二甲基甲酰胺。多叶斑叶兰传粉者为中华蜜蜂(Apis cerana)。多叶斑叶兰具有自交亲和能力,但在自然界不具有主动自交现象,必须依赖中华蜜蜂传粉,花色及花香气味为吸引传粉者的主要因素。     

Pollen and resource limitations to lifetime seed production in a wild population of the endangered plant Disanthus cercidifolius var. longipes H. T. Chang (Hamamelidaceae)

Disanthus cercidifolius Maxim. var. longipes H. T. Chang, a plant species that only occurs in a few counties in Hunan, Jiangxi and Zhejiang Provinces and with a relatively small number of individuals, is ranked as a second Class endangered species for conservation in China. We have studied the effect of pollen and resources available to female reproduction, and the reproductive mechanism of “excess flowers with low fruit set” in Disanthus cercidifolius Maxim. var. longipes H. T. Chang was discussed. Results are as follows Pollen from different sources has significant effects on fruit set and seed set of Disanthus cercidifolius Maxim. var. longipes H. T. Chang. The pollen source rather than pollen numbers significantly affected reproduction of this species. In wild populations, producing one fruit needs about 54.8 flowers, and one satiation seed needs about 6.60 flowers or 83.19 ovules. After fertilizing, which was propitious to flower development, the abortion rate of flower buds was decreasing, but the flowering rate was increasing. The fruit set and seed set was also significantly increasing, while abortion rate of fruit was significantly decreasing. With the increasing percentages of cutting leaves, the fruit set decreased, but the abortion rate of fruit shows no significant differentiation among treatments. After cutting branches that were puny, broken and insectin-fested branches, the flower number seemed to be decreasing, but the fruit set and seed set all increased significantly. After removing some flowers, the fruit set was calculated with respect to the number of flowers remaining after the treatment increased with increasing of percentages of flower removal, whereas fruit set calculated with respect to the initial number of flowers remained constant, and the mean weights of per fruit and per seed all decreased significantly. Sufficient spatial or temporal heterogeneities in nutrient levels might allow limitation of seed set by resources and pollen in a natural population, while supplying resources may indirectly affect pollination by increasing attraction of the flowers to pollinators. There were very low fruit and seed sets in natural populations of Disanthus cercidifolius Maxim. var. longipes H. T. Chang. Different factors may have interacted to effect a low fruit set. A joint adoption of the “selection abortion hypothesis”, “ovary reserve hypothesis” and “male function hypothesis” seems to be the most likely explanation for the reproductive strategy of “excess flowers with few fruit sets” in Disanthus cercidifolius Maxim. var. longipes H. T. Chang. __________ Translated from Acta Ecologica Sinica, 2006, 26(2): 496–502 [译自: 生态学报]     

The reproductive biology ofJaborosa integrifolia (Solanaceae): Why its fruits are so rare?

Jaborosa integrifolia exhibits stigma-height polymorphism. There are individuals with flowers where anthers and stigma are at the same height but the rule is variable herkogamy, the most common type (75%) being that with an exerted stigma. Self- and cross-tubes did not differ in their capability to reach the ovary (t = –0.67,P < 0.53); they had a high growth rate (6.95 ± 2.28 mm h^–1). There is not autogamy but mostly self-incompatibility. Fruits from controlled cross-pollination showed the highest seed set and seed viability. The nectar sugar is characterized by a similar amount of glucose and fructose, and by the absence of sucrose. Although nectar secretion was continuous throughout the life of the flower, most nectar was secreted during the first 24 h after flower opening. Nectar production costs appear to be lower than in other species since nectar secretion is neither inhibited after a removal (i.e. a pollinator visit) nor reabsorbed as the flower ages. Sphingids visit the flowers mainly after midnight. They insert their proboscis down to the base of the corolla tube to reach the nectar. The upper limit to fruit production is set by pollinator visits. Fruits produced from open-pollinated flowers are often predated by numerous larvae (mainly lepidopteran ones). Considering that this species is mostly self-incompatible and pollination is limited, that each plant displays only a low number of flowers throughout the flowering season, and that there is a high rate of fruit predation, it is not surprising that fruits ofJ. integrifolia are so rare.     

Pollen limitation in three sympatric species ofVaccinium (Ericaceae) in the Upper Ardennes, Belgium

The reproductive success of three co-flowering species ofVaccinium (V. myrtillus, V. vitis-idaea andV. uliginosum) was studied in one heathland of the Upper Ardennes, Belgium, during three years (1988–1990). The purpose was to examine whether pollen limitation, flower position and flowering phenology may influence patterns of fruit set in these three sympatric species. I quantified fruit and seed set following supplementary hand-pollinations and compared this to natural fruit set. On the same plants, I also quantified fruit and seed set in relation to the spatial position of the flowers on the ramet and their temporal sequence of blooming. Hand-pollination had no significant effect on fruit set inV. vitis-idaea andV. uliginosum, but significantly increased seed number per fruit in 1989. InV. myrtillus both fruit and seed set were increased by supplementary pollination, but significantly in only one year. Analyses of position effects revealed that the fruits in the lower positions in the ramet did not mature preferentially and did not contain more seeds inV. uliginosum and inV. vitis-idaea. Flowering phenology also had no significant effect for these species. InV. myrtillus hand-pollinated flowers showed a seasonal decline in seed number, and control (naturally pollinated) flowers showed a seasonal increase in fruit set. Fruit and seed set appear to be pollen-limited rather than resource-limited inV. uliginosum. In the two other species, unfavorable weather (frosts) could be a more important cause of low fruit and seed set.     

Significance of intra-inflorescence variation on flowering time of a spring ephemeral, <Emphasis Type="Italic">Gagea lutea</Emphasis> (Liliaceae), under seasonal fluctuations of pollinator and light availabilities

I studied the relationship between seed-set patterns within inflorescences and temporal variations in light and pollinator availabilities for 2 years in the spring ephemeral species Gagea lutea in a deciduous forest. Timing of canopy closure and seasonal trend of pollinator frequency did not synchronize with the annual fluctuation in flowering phenology. In the early snowmelt year, seed-set success reflected the seasonal pollinator abundance from early to middle flowering periods. In the late snowmelt year, however, seed-set rates were independent of pollinator activity and decreased with canopy closing even after hand pollination. The restricted seed production by defoliation and the increase in seed-set rates at the forest edge suggested that seed production was supported by current photosynthetic carbon gain. Thus, annual fluctuations of reproductive success can explain the variation in flowering phenology within a population although seasonal light deterioration would serve as a selective force for flowering in the early season.     

Pollination system of Holcoglossum rupestre (Orchidaceae): A special and unstable system

The pollination system of Holcoglossum rupestre was studied in northwestern Yunnan. The species is self-compatible but pollinator-dependent and offers nectar as a reward to visitors. A species of beetle, Hybovalgus bioculatus Kolbe, is identified as the pollinator among all visitors observed. The beetles carried the pollinia in two different positions, i.e. on the buttock and the foreleg, and pollinated the flowers by two different mechanisms. The analysis suggests that Holcoglossum rupestre and its pollinator may be in an unstable pollination relationship. The natural fruit set is low, which is attributed to the low visiting frequency of pollinator and the effective mechanism to prevent self-pollination.     

Little evidence for negative effects of an invasive alien plant on pollinator services

Many invasive alien plants occur in large populations with abundant flowers which are highly attractive to pollinators, and thus might affect pollination of co-occurring native species. This study focuses on the invasive Heracleum mantegazzianum and distance-dependent effects on pollination of Mimulus guttatus in abandoned grassland over 2 years. First, we examined pollinator abundance in yellow traps at 0, 10, 30 and 60–200 m from H. mantegazzianum. We then placed M. guttatus plants at the same distances to monitor effects of the invasive species on pollinator visitation and seed set of neighbouring plants. Finally, we conducted a garden experiment to test if deposition of H. mantegazzianum pollen reduces seed set in M. guttatus. No distance effect was found for the number of bumblebees in traps, although the invasive species attracted a diverse assemblage of insects, and visitation of M. guttatus was enhanced close to H. mantegazzianum. This positive effect was not reflected by seed set of M. guttatus, and heterospecific pollen decreased seed set in these plants. Overall there is little evidence for negative effects of the invasive species on pollination of neighbouring plants, and flower visitation even increases close to the invaded patches. The functional role of the invader and suitable control strategies need further clarification, since removal of H. mantegazzianum may actually damage local pollinator populations. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

Pollination ofNuphar (Nymphaeaceae) in Europe: Flies and bees rather thanDonacia beetles

Nuphar (Nymphaeaceae) comprises a small holarctic group of aquatic perennials whose flowers are pollinated by flies, bees, and beetles. We studied pollination in different populations of the two European speciesN. lutea andN. pumila in Norway and in Germany. Flowers are self-compatible and protogynous, preventing automatic selfing, and insect pollination is required for seed production. Sympatric populations were studied in Vest-Agder county in Norway to determine whetherN. lutea andN. pumila have the same or different pollinators. Allopatric populations ofN. lutea in Germany and Norway were then compared to determine whether their pollinator spectra differ as would be expected in an open flower with seemingly easily accessible pollen and nectar. Results of the present as well as previous studies of the pollination ofN. lutea andN. pumila show that both species are mainly pollinated by flies, including apparentNuphar specialists, such as the scatophagidHydromyza livens and the ephydridsHydrellia andNotiphila, the last also a long-known pollinator ofN. advena in Florida. Pollinator overlap between sympatric heterospecific populations was small, while allopatric conspecific populations had similar visitor and pollinator spectra. We found no evidence of pollination byDonacia beetles as reported from some North American populations ofNuphar. Dedicated to Univ.-Prof. DrF. Ehrendorfer on the occasion of his 70th birthday     

The production of floral oils byMonttea (Scrophulariaceae) and the function of tarsal pads inCentris bees

Plant species that secrete oil as their primary floral reward are rare and sporadically found in the angiosperms. We report here thatMonttea, a genus previously unsuspected of being an oil-plant, produces lipids from trichome elaiophores on the inside of the lower (anterior) lip. The discovery of the production of oils by species of this S. American genus explains the occurrence of unusual dual-function collecting structures in ArgentineCentris (Hymenoptera: Anthophoridae) and explains the presence of oil-collecting bees in regions where oil-secreting flowers were previously thought to be absent. The behavior of these centridine pollinators onMonttea flowers parallels that of oil-collecting bees onDiascia (Scrophulariaceae) in S. Africa.     

Pollination biology and breeding systems in the terrestrial orchid Bletilla striata

The levels of fruit set in different habitats (three inland populations vs. three coastal populations) and the breeding system in the terrestrial orchid Bletilla striata were examined in two groups of populations in Haenam-gun, Jollanam-do Province in southern Korea. In total inland populations (mean fruit set=13.3%) showed about eight-fold higher fruit set per plant than the coastal populations (1.6%), probably due to mild climatic conditions in the inland populations during the flowering period. Honey bee workers Apis mellifera were observed in the inland populations, but no visitors were found in the coastal populations. However, a great increase in fruit set was achieved by hand-pollination: the percentage of fruit set through artificial self-pollination and geitonogamous pollination were close to 90%, indicating that B. striata is self-compatible. Agamospermy and spontaneous autogamy (automatic self-pollination) were not detected in this study. These indicate that pollinia vectors are essential for achieving fruit set in natural populations of B. striata. Except in one inland population examined in 2001, there was no significant correlation between the patch size per population and the average percentage of fruit set, which might reflect deception-pollination of B. striata and/or paucity of pollinators. Overall levels of fruit sets within and among patches in each population were homogeneous during a two-year period.     

Pollination and reproduction of a self-incompatible forest herb in hedgerow corridors and forest patches

Habitat-corridors are assumed to counteract the negative impacts of habitat loss and fragmentation, but their efficiency in doing so depends on the maintenance of ecological processes in corridor conditions. For plants dispersing in linear habitats, one of these critical processes is the maintenance of adequate pollen transfer to insure seed production within the corridor. This study focuses on a common, self-incompatible forest herb, Trillium grandiflorum, to assess plant–pollinator interactions and the influence of spatial processes on plant reproduction in hedgerow corridors compared to forests. First, using pollen supplementation experiments over 2 years, we quantified the extent of pollen limitation in both habitats, testing the prediction of greater limitation in small hedgerow populations than in forests. While pollen limitation of fruit and seed set was common, its magnitude did not differ between habitats. Variations among sites, however, suggested an influence of landscape context on pollination services. Second, we examined the effect of isolation on plant reproduction by monitoring fruit and seed production, as well as pollinator activity and assemblage, in small flower arrays transplanted in hedgerows at increasing distances from forest and from each other. We detected no difference in the proportion of flowers setting fruit or in pollinator activity with isolation, but we observed some differences in pollinator assemblages. Seed set, on the other hand, declined significantly with increasing isolation in the second year of the study, but not in the first year, suggesting altered pollen transfer with distance. Overall, plants in hedgerow corridors and forests benefited from similar pollination services. In this system, plant–pollinator interactions and reproduction seem to be influenced more by variations in resource distribution over years and landscapes than by local habitat conditions. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

毛翠雀花花序内的性分配和繁殖成功

两性花植物花序内不同位置的性分配和繁殖成功一般存在差异,通常认为资源竞争、结构效应和交配环境是形成这种差异的主要原因。为了研究雄性和雌性繁殖资源在花序内不同位置间的最优分配问题,该文以青藏高原高寒草甸典型高山植物毛翠雀花为材料,通过比较花序内不同位置的花部特征和种子性状,对其花序内的性分配模式和雌性繁殖成功进行研究,并通过观察传粉者运动特点以及人工去花和补授花粉实验,探讨花序内资源竞争和交配环境对繁殖资源分配的影响。结果表明:(1)不同位置间的雄蕊数、雄蕊鲜重/雌蕊鲜重、花粉数及花粉胚珠比从花序基部到上部显著增加,而雌蕊鲜重和胚珠数逐渐减少,表现出上部花偏雄的性分配;上部花的结籽率显著低于基部花和中部花,不同位置间的发育种子数/果实和发育种子重/果实随着花位置的升高而显著降低,说明基部花具有更佳的雌性繁殖成效。(2)去花处理后,剩余果实的单个种子重/果实显著增加,但发育种子数/果实没有显著增加;而给上部花人工补授异花花粉后,位置间结籽率的差异消失,说明传粉限制而非资源竞争导致了花序内位置依赖的种子生产模式。(3)毛翠雀花雄性先熟的开花特征,以及传粉者苏氏熊蜂从花序基部到上部的定向访花行为,导致了花序内交配环境的变化。综上结果表明,毛翠雀花花序内的性分配和繁殖成功差异是对交配环境适应的结果,对其在高山环境中实现雌雄适合度最优化具有重要意义。     

Genetic variation, genetic structure and effective population size in the tropical holoparasitic endophyteBdallophyton bambusarum (Rafflesiaceae)

The genetic population structure inBdallophyton bambusarum, an endoparasite, was studied in ten subpopulations from a subdeciduous tropical forest in Veracruz Mexico. The sample was analyzed using seven polymorphic loci in cellulose acetate electrophoresis. Isozyme data indicated that the subpopulations ofB. bambusarum contained high genetic variability (H_ep = 0.452 ± 0.045, S.E.). Our analysis suggests that almost each inflorescence ofB. bambusarum is an individual. The subpopulations studied were genetically similar (average Nei's genetic identity 0.941 ± 0.051 and F _st values 0.097 ± 0.026), suggesting that genetic differentiation among subpopulations was small. Direct estimates of effective population size was derived from observations of three fluorescent dyes, and from the genetic neighborhood area derived from these data. The neighborhood area, multiplied by the total density of individuals, gave an N_e = 124.84 plants, and when corrected to consider the proportion of males and females gave an N_e = 118.59 individuals. An indirect estimate of Nm was obtained from the F _st values (mean Nm=2.037), giving an indirect estimate of the effective population size N_b = 12.8 individuals. Both values are relatively high when compared to other plant studies. The gene flow and/or effective populations size of the studied subpopulations ofB. bambusarum are believed to be large enough to prevent differentiation among subpopulations due to genetic drift.     

Pollination effectiveness and pollinator importance in a population of Heterotheca subaxillaris (Asteraceae)

 Assessing the relative contributions to seed set for each of a plant species’ floral visitors provides an indication of the relative influence of these visitors on the plant’s reproductive success. This study examined pollinator activity and seed set in a population of Heterotheca subaxillaris, a species that exhibits a floret dimorphism (heads bearing disk and ray florets), and that is visited by both generalist foragers and specialist bees. Visits by nine bee genera and one genus of skipper were recorded in the study population. During the period of study, these insects varied in their relative abundance, in their foraging activity on a head, and in their pollination effectiveness. The pattern of pollination effectiveness shown by the different pollinators was similar for both floret types, although seed set was higher overall for ray florets. Pollinator importance, calculated as the product of pollination effectiveness and relative abundance, was dictated by a pollinator’s relative abundance. The single specialist bee species observed in the study population proved to have neither higher pollination effectiveness nor higher pollinator abundance than other pollinators. This would suggest that H. subaxillaris is not under strong selective pressure to co-specialize with its specialist visitor. Received: 21 December 1995 / Accepted: 20 June 1996     

An invasive dandelion unilaterally reduces the reproduction of a native congener through competition for pollination

The impact of invasive alien species on native species is of increasing global concern. Invasive plants have various negative effects on natives through competition; however, relatively little is known about competition for pollination. The relationship between Japanese native dandelions (Taraxacum spp.) and invasive congeners may be a typical case of such an interaction. For example, native dandelions are being replaced by invasive congeners, especially in urban and suburban areas of Japan. To explain this phenomenon, we hypothesized that when natives are mixed with attractive invasives, natives may suffer from reduced seed set because invasives deprive natives of pollinators or because pollinators frequently move between species, resulting in interspecific pollen transfer. To test this hypothesis, we studied the effect of the invasive dandelion T. officinale on the pollination and seed set of the native T. japonicum using artificial arrays of monospecific and mixed-species plots as well as natural populations. Taraxacum officinale attracted more pollinator visits, perhaps because it produced more nectar than T. japonicum. The number of pollinator visits to T. japonicum was reduced when the congeners were grown together, and pollinators moved frequently between the two species. The proportion of seed set for T. japonicum was reduced in the presence of T. officinale in both artificial arrays and natural populations. These results support our hypothesis that interspecific competition for pollination plays an important role in the recent replacement of native dandelions by invasive congeners in Japan. Because invasive dandelions are apomicts, negative effects are incurred only by sexual natives. Thus, this system can be recognized as a rare case of interspecific interaction through pollination.     

Reproductive ecology of tropical forest trees in logged and fragmented habitats in Thailand and Costa Rica

Invertebrates mediate several important ecological processes, including pollination and seed predation, and events that affect invertebrate diversity or behaviour can potentially disrupt forest regeneration processes. This study investigates the impact of logging in Thailand and forest fragmentation in Costa Rica on the pollination and seed production of two self-incompatible forest trees. Logging in a dry deciduous dipterocarp forest in Thailand resulted in reduced densities of the common dipterocarp treeShorea siamensis and variably isolated individual trees. The number of flower visits to S. siamensis by pollinating Trigona bees was not affected by logging disturbance. However, pollinators did spend longer periods of time foraging in the canopies of isolated trees which were more prevalent in logged areas where tree density had been reduced. Consequently, at the logged site few cross-pollinations were effected and fruit set of S. siamensis was considerably lower than at nearby unlogged sites where distances between flowering conspecifics were smaller. Reduced fruit set has long-term implications for the recovery of S. siamensis populations in disturbed areas, and local population genetic structure is likely to be affected as reduced outcrossing rates among trees in disturbed regions results in relatively inbred seed. In Costa Rica forest fragmentation has restricted the once widespread tree Anacardium excelsum to forest patches located in an agriculturally-dominated landscape. As with S. siamensis, the abundance of pollinators, also Trigona bees, in the canopies of A. excelsum was largely unaffected by fragment size. Nevertheless, pollination success and seed production was positively correlated with fragment size. We propose that small bees rarely move between forest fragments and gene exchange through pollination occurs predominantly among trees within fragments and, together with likely low genetic variability in small fragments, that this contributes to the observed reduced fertilisation and seed set of A. excelsum. Thus increased tree isolation tree through selective logging or habitat fragmentation by forest clearance can result in reduced seed set due to changes in the foraging patterns of poorly mobile pollinators. Even if population sizes of the pollinators are maintained following environmental perturbation, this study shows that disturbance may disrupt pollination processes through changes in pollinator foraging behaviour. More attention needs to be focussed on changes in the behaviour of species involved in key ecological interactions following disturbance events in tropical forests.