Detecting the effects of environmental change above the species level with beetles in a fragmented tropical rainforest landscape

Abstract.  1. Invertebrate data identified to coarse groupings according to taxonomy or ecology are easier to obtain than species-level data. However, it is unclear if such data are suitable for detecting environmental change.
2. We compared the performance of four types of data (species, family, trophic group, and body size) to quantify the responses of beetle assemblages with three types of environmental change on the Atherton Tableland in north-eastern Queensland, Australia. Clearing of rainforest creates two levels of environmental change: first the extreme, forest loss, and second, the more subtle change associated with forest fragmentation. A strong rainfall gradient across the study landscape also influences the biota, imposing a third type of environmental change, independent of rainforest modification.
3. Almost 20 000 beetles were sampled from six replicate sites in each of pasture, small rainforest remnants, and both the edges and interiors of large rainforest remnants.
4. All four types of data showed significant multivariate differences in assemblage composition between pasture and rainforest sites. Species-level data in multivariate analyses showed an effect of fragmentation on small remnants and also showed variation which corresponded with the spatial aridity gradient across sites. Both of these patterns were only weakly evident or were non-existent at the level of family, trophic group, or body size class. No level of taxonomic resolution gave consistently stronger results when univariate tests were applied to individual component taxa.
5. We conclude that data above the species level may not have the sensitivity required to detect more subtle forms of environmental change.     

Importance and intensity of competition along a fertility gradient and across species

Questions: 1. Can the importance and the intensity of competition vary independently along a nutrient gradient? 2. Are these variations species dependent? Location: Sub‐alpine pastures of the northern French Alps. Methods: Competition intensity measures how much competition decreases the performances of an organism. Competition importance measures how much competition contributes to affect performance, among other processes (such as environmental stress or disturbance). Competition intensity and importance were measured on three co‐occurring species: Festuca rubra, a perennial grass, and two forbs of contrasting basal area, Chaerophyllum hirsutum and Alchemilla xantho‐chlora. A neighbour removal experiment was performed on Festuca rubra in three sub‐alpine grassland communities differing in fertility and on Chaerophyllum hirsutum and Alchemilla xanthochlora in the two more fertile of these communities. The importance of competition was quantified using an index proposed by Brooker et al. (2005). Results: Competition intensity and importance showed different patterns of variation along the fertility gradient for Festuca rubra: competition importance decreased with decreasing fertility whereas competition intensity did not change. The largest forb was the least affected by competition. Our results suggest that the importance of competition for all three species depended on their individual tolerance to low nutrient availability. Conclusions: 1. The distinction between the importance and the intensity of competition is helpful to explain conflicting results obtained on the variations of competition indices along productivity gradients. 2. The choice of a phytometer can affect the conclusions drawn from empirical studies.     

The importance of importance

Failure to distinguish between 'importance' and 'intensity' of competition has hindered our ability to resolve key questions about the role interactions may play in plant communities. Here we examine how appropriate application of metrics of importance and intensity is integral to investigating key theories in plant community ecology and how ignoring this distinction has lead to confusion and possibly spurious conclusions. We re-explore the relationship between competition intensity and importance for individuals across gradients, and apply our review of concepts to published data to help clarify the debate. We demonstrate that competition importance and intensity need not be correlated and show how explicit application of the intensity and importance of competition may reconcile apparently incompatible paradigms.     

Density-dependent effects on the reproductive fitness of the New Zealand beech scale insect (Ultracoelostoma assimile) across multiple spatial scales

Abstract.  1. Intraspecific competition can be an important factor influencing the individual fitness of organisms. This study reports density-dependent effects on the fecundity of the beech scale insect ( Ultracoelostoma assimile , Hemiptera: Margarodidae) on naturally occurring, canopy-dominant red beech ( Nothofagus fusca , Fagaceae) trees in New Zealand. For the first time an increasing intensity of intraspecific competition at increasing spatial scales within individual host trees is demonstrated.
2. Beech scale insect 'tests' containing adult female scale insects and eggs were collected from the trunks of 10 red beech trees [17.7–48.5 cm in diameter at breast height (1.4 m), d.b.h.] with varying densities of scale insect infestation. The relationship between individual female fecundity and scale insect density at three spatial scales: local (within 5 cm), lower trunk (below 2 m above ground level), and whole tree, was tested.
3. Beech scale insect fecundity was density dependent, with total female egg load decreasing with increasing scale insect density. The strength of the density-dependent effect increased with increasing spatial scale, suggesting that scale insects are creating a tree-wide drain on the quality of phloem sap, rather than depleting nutrients from localised, high-density areas within trees. These results indicate that at high densities, the New Zealand beech scale insect can have a negative effect on the nutritional quality of the phloem of red beech, thus negatively affecting conspecifics elsewhere on individual host trees. The increasing intensity of the effect with increasing spatial scale within individual trees emphasises the importance of measuring density-dependent effects at the appropriate spatial scale.     

Effects of patch size and density on flower visitation and seed set of wild plants: a pan-European approach

1.  Habitat fragmentation can affect pollinator and plant population structure in terms of species composition, abundance, area covered and density of flowering plants. This, in turn, may affect pollinator visitation frequency, pollen deposition, seed set and plant fitness.
2.  A reduction in the quantity of flower visits can be coupled with a reduction in the quality of pollination service and hence the plants' overall reproductive success and long-term survival. Understanding the relationship between plant population size and/or isolation and pollination limitation is of fundamental importance for plant conservation.
3.  We examined flower visitation and seed set of 10 different plant species from five European countries to investigate the general effects of plant populations size and density, both within (patch level) and between populations (population level), on seed set and pollination limitation.
4.  We found evidence that the effects of area and density of flowering plant assemblages were generally more pronounced at the patch level than at the population level. We also found that patch and population level together influenced flower visitation and seed set, and the latter increased with increasing patch area and density, but this effect was only apparent in small populations.
5.   Synthesis. By using an extensive pan-European data set on flower visitation and seed set we have identified a general pattern in the interplay between the attractiveness of flowering plant patches for pollinators and density dependence of flower visitation, and also a strong plant species-specific response to habitat fragmentation effects. This can guide efforts to conserve plant–pollinator interactions, ecosystem functioning and plant fitness in fragmented habitats.     

Competition between tree seedlings and herbaceous vegetation: support for a theory of resource supply and demand

1 We measured competition intensity (CI) between herbaceous vegetation and tree seedlings ( Quercus macrocarpa and Q. ellipsoidalis ) along an experimental moisture–light gradient. Contrasting theories were tested by comparing variation in competition intensity to changes in neighbour biomass and resource supply and demand.
2 CI based on survival was inversely correlated with net soil water supply (gross supply minus demand by herbaceous vegetation). CI was not positively correlated with either gross resource supply or neighbour biomass, contrary to predictions of Grime's triangular model for plant strategies.
3 Many of the inconsistencies and conflicting results that have characterized the recent literature on plant competition could be eliminated if changes in competition intensity along a resource gradient are compared with changes in net resource supply rather than changes in productivity or neighbour biomass.
4 Tree seedling success in savannas and grasslands may be strongly influenced by the intensity of competition from herbaceous vegetation. Factors that reduce soil water content are likely to increase competition intensity (and reduce seedling success) in these environments, while factors that increase soil water content will favour seedling success through decreased competition for water with herbaceous vegetation.     

Intensity and Importance of Competition for a Grass (Festuca rubra) and a Legume (Trifolium pratense) Vary with Environmental Changes

How plant competition varies across environmental gradients has been a long debate among ecologists. We conducted a growth chamber experiment to determine the intensity and importance of competition for plants grown in changed environmental conditions. Festuca rubra and Trifolium pratense were grown in monoculturs and in two- and/or three-species mixtures under three environmental treatments. The measured competitive variations in terms of growth (height and biomass) were species-dependent. Competition intensity for Festuca increased with decreased productivity, whilst competition importance displayed a humpback response. However, significant response was detected in neither competition intensity nor importance for Trifolium. Intensity and importance of competition followed different response patterns, suggesting that they may not be correlated along an environmental gradient. The biological and physiological variables of plants play an important role to determine the interspecific competition associated with competition intensity and importance. However, the competitive feature can be modified by multiple environmental changes which may increase or hinder how competitive a plant is.     

Competitive interactions and the effect of herbivory on Bt-Brassica napus, Brassica rapa and Lolium perenne

1.  The probability of a transgenic crop establishing a feral population outside cultivated areas and possibly outcompeting naturally occurring species needs to be assessed to make an ecological risk assessment of the transgenic crop.
2.  The interaction between herbivory and competition is thought to determine the ecological success of insect-resistant plants, and this interaction was investigated in a competition experiment with transgenic insect-resistant Bt- Brassica napus , Brassica rapa , Lolium perenne , and herbivory from the large white butterfly Pieris brassicae .
3.  As expected, herbivory had a negative effect on the biomass of B. rapa at high plant densities. The competitive ability of L. perenne , when growing with B. rapa , increased significantly with the level of herbivory on B. rapa .
4.  To predict the effect of herbivory in a natural ecosystem, plant competition between the two annual Brassica species was analysed in a population ecological model. It was concluded that it is probable that transgenic Bt- B. napus plants may invade a natural habitat if herbivory is sufficiently high and the habitat is suitable for B. napus .
5.   Synthesis and applications . The results indicate that it is important to study the interaction between herbivory and competition when assessing the ecological risk of insect-resistant genetically modified crops. Furthermore, combining ecological data from manipulated experiments with population ecological modelling is a fruitful approach when conducting environmental risk assessments.     

Collaboration in scientific research: a critical need for freshwater ecology

1. Collaboration means actively working together to achieve things which could not be done alone. This article attempts to provide an overall, unified, guiding framework for collaboration in freshwater ecology by discussing aspects of collaboration at individual and organizational levels, and addressing international linkages.
2. The essential elements of collaboration are communication and trust, and effective project management. Barriers to effective collaboration include competition, organizational cultures and organizational instabilities.
3. The success of collaboration can be measured by tangible benefits such as increased numbers of peer-reviewed publications, the production of working models and a number of intangible benefits.
4. Interactions between individuals lie at the heart of an effective collaboration; organizational arrangements should facilitate this interaction. Some governments are encouraging collaboration to increase cost efficiency and allocate accountability. This trend should continue on an international level.
5. Collaboration is a key to future research in freshwater ecology.     

Competition components along productivity gradients – revisiting a classic dispute in ecology

Competition is ubiquitous in plant communities with various effects on plant fitness and community structure. A long-standing debate about different approaches to explain competition is the controversy between David Tilman and Philip Grime. Grime stated that the importance of competition relative to the impact of the environment increases along a productivity gradient, while Tilman argued that the intensity of competition is independent of productivity. To revisit this controversy, we assumed that the effects of plant–plant interactions are additive and applied the new competition indices by Díaz-Sierra et al. (2017) in a field experiment along a productivity gradient in S-Germany, using the rare arable plant Arnoseris minima as a study species. The ‘target technique' was applied, to separate the effects of root and shoot competition. The study plants were exposed to five competition treatments with three replicates in 18 sites, respectively. We investigated the expectation that root competition is more intense in unproductive sites than shoot competition. Additionally, we predicted survival to be less affected by competition than growth-related plant parameters. Using the biomass of individuals without competition as a proxy for site productivity there was a positive relationship with competition importance but no relationship with competition intensity when plants experienced full competition. Survival of the target plants was unaffected by competition. Root competition was the main mechanism determining the performance of the target plants, whereas the effect of shoot competition was relatively low albeit increasing with productivity. We conclude that when considering plant–plant interactions additive both Grime's and Tilman's theories can be supported.     

Strategic ejaculation in the egg parasitoid Trichogramma turkestanica (Hymenoptera: Trichogrammatidae)

Abstract.  1. Several parameters influence sperm allocation by males, including their size and sperm stock, intra-specific variability, quality of females', as well as the risk and intensity of sperm competition.
2. Models predict that males should invest the maximum ejaculate size when sperm competition intensity is low. As sperm competition intensity increases, males should decrease the number of sperm transferred during mating.
3. This decrease in sperm transfer to females occurs because the benefits gained by males with each extra unit of expenditure on sperm decrease. When sperm supply is not unlimited, males could expect a better return by keeping some or all sperm for mating under lower competition intensity.
4. In this study, the ejaculate size of males that were kept in groups of one, five or 10 males prior to mating, has been investigated in the haplodiploid egg parasitoid Trichogramma turkestanica Meyer (Hymenoptera: Trichogrammatidae).
5. As predicted by theory, the number of sperm transferred decreased significantly with an increase in the number of rivals.
6. This is the first study showing strategic sperm allocation depending on sperm competition intensity in a parasitoid.     

Alteration of the gastrointestinal microbiota of mice by edible blue-green algae

Aims:  To characterize the effect of edible blue-green algae (cyanobacteria) on the gastrointestinal microbiota of mice.
Methods and Results:  C57BL/6J mice were fed a diet supplemented with 0% or 5% dried Nostoc commune , Spirulina platensis or Afanizominon flos-aquae (w/w) for 4 weeks. Molecular fingerprinting of the colonic microbiota using denaturing gradient gel electrophoresis revealed that administration of N. commune induced major alterations in colonic microbiota composition, while administration of S. platensis or A. flos-aquae had a more subtle impact. Community profile analysis revealed that administration of N. commune did not reduce microbial diversity indices of the colonic microbiota. Despite its pronounced effects on the bacterial composition in the colon, total bacterial numbers in the gut of mice fed N. commune were not reduced as assessed by quantitative real-time PCR and bacteriological culture.
Conclusions:  The results presented here show that administration of blue-green algae, and especially N. commune , alters colonic microbiota composition in mice with limited effects on total bacterial numbers or microbial diversity.
Significance and Impact of the Study:  Blue-green algae are consumed in many countries as a source of nutrients and to promote health, and they are intensively studied for their pharmaceutical value. Given the importance of the gut microbiota for many host functions, the effects of blue-green algae on gut microbial ecology revealed during this study should be considered when using them as food supplements or when studying their pharmaceutical properties.     

Metacommunity patterns of highly diverse stream midges: gradients,chequerboards, and nestedness,or is there only randomness?

Abstract.  1. Several non-random patterns in the distribution of species have been observed, including Clementsian gradients, Gleasonian gradients, nestedness, chequerboards, and evenly spaced gradients. Few studies have examined these patterns simultaneously, although they have often been studied in isolation and contrasted with random distribution of species across sites.
2. This study examined whether assemblages of chironomid midges exhibit any of the idealised distribution patterns as opposed to random distribution of species across sites within the metacommunity context in a boreal drainage system. Analyses were based on stream surveys conducted during three consecutive years. Analytical approaches included ordinations, cluster analysis, null models, and associated randomisation methods.
3. Midge assemblages did not conform to Clementsian gradients, which was evidenced by the absence of clearly definable assemblage types with numerous species exclusive to each assemblage type. Rather, there were signs of continuous Gleasonian variability of assemblage composition, as well as significant nested subset patterns of species distribution.
4. Midge assemblages showed only weak relationships with any of the measured environmental variables, and even these weak environmental relationships varied among years.
5. Midge assemblages did not appear to be structured by competition. This finding was somewhat problematic, however, because the two indices measuring co-occurrence provided rather different signs of distribution patterns. This was probably a consequence of how they actually measure co-occurrence.
6. Although midge assemblages did not show a perfect match with any of the idealised distribution patterns, they nevertheless showed a resemblance to the empirical patterns found previously for several plant and animal groups.     

On the different nature of top-down and bottom-up effects in pelagic food webs

SUMMARY 1. Each individual planktonic plant or animal is exposed to the hazards of starvation and risk of predation, and each planktonic population is under the control of resource limitation from the bottom up (growth and reproduction) and by predation from the top down (mortality). While the bottom-up and top-down impacts are traditionally conceived as compatible with each other, field population-density data on two coexisting Daphnia species suggest that the nature of the two impacts is different. Rates of change, such as the rate of individual body growth, rate of reproduction, and each species' population growth rate, are controlled from the bottom up. State variables, such as biomass, individual body size and population density, are controlled from the top down and are fixed at a specific level regardless of the rate at which they are produced.
2. According to the theory of functional responses, carnivorous and herbivorous predators react to prey density rather than to the rate at which prey are produced or reproduced. The predator's feeding rate (and thus the magnitude of its effect on prey density) should hence be regarded as a functional response to increasing resource concentration.
3. The disparity between the bottom-up and top-down effects is also apparent in individual decision making, where a choice must be made between accepting the hazards of hunger and the risks of predation (lost calories versus loss of life).
4. As long as top-down forces are effective, the disparity with bottom-up effects seems evident. In the absence of predation, however, all efforts of an individual become subordinate to the competition for resources. Biomass becomes limited from the bottom up as soon as the density of a superior competitor has increased to the carrying capacity of a given habitat. Such a shift in the importance of bottom-up control can be seen in zooplankton in habitats from which fish have been excluded.     

Effect of light and predator abundance on the habitat choice of plant-attached zooplankton

1. The diurnal variations in the habitat choice of the periodically plant-attached cladoceran Sida crystallina together with light environment and predator abundance were studied.
2. The density of S. crystallina attached to floating leaves of Nuphar lutea increased between 18:00 and 20:00 hours, when light intensity underneath the leaves was temporarily increased, and decreased again when light intensity declined. A proportion of S. crystallina remained in the swimming mode underneath the leaves even during daylight, indicating that the water column sheltered by the leaves is safer than the open water.
3. In the water adjacent to the leaves, the density of S. crystallina increased steeply in the dark. The increase was not accompanied by a decrease in S. crystallina attached to plant leaves, indicating that the nocturnal increment in the open water density of S. crystallina was due to migration from daytime refuges other than floating leaves.
4. Sida crystallina was most intensively consumed by perch ( Perca fluviatilis ). Predation threat by fish had weaker effects on the density of S. crystallina attached to plant leaves than on cladocerans in the adjacent water. Cladocerans underneath floating plant leaves, whether attached or not, are probably less vulnerable to fish predation than those outside the leaf cover.
5. The results suggested that light intensity is the proximate factor regulating the attachment of zooplankton to the lower surfaces of floating macrophyte leaves. Light intensity has a positive effect on the density of S. crystallina attached to the floating leaves and a negative effect on density in the water. Predation threat by fish has a strong effect on the migration of zooplankters into the open water habitat.     

Interaction of specialist root and shoot herbivores of Alliaria petiolata and their impact on plant performance and reproduction

Abstract.  1. This study explored interactions of two spatially and temporally separated weevils and their impact on Alliaria petiolata (garlic mustard) survival, growth, and reproduction at different herbivore densities.
2. The root-mining weevil Ceutorhynchus scrobicollis attacks A. petiolata rosettes from October to April, and larvae complete development in May. The shoot-mining weevil Ceutorhynchus alliariae attacks bolting plants in April/May with larvae completing development in June–July. Priority effects were expected, with early attack of C. scrobicollis affecting the later attacking C. alliariae , mediated through changes in plant growth or chemistry.
3. Attack by C. scrobicollis significantly increased plant mortality and changed plant architecture, while C. alliariae only significantly reduced plant height. Attack by C. scrobicollis also increased nitrogen content of stems.
4. Root feeding by C. scrobicollis affected the feeding niche of C. alliariae , but increased stem nitrogen content did not result in increased stem miner survival. While reduced height and stem diameters as a result of C. scrobicollis attack reduced C. alliariae attack at the stem level, attack at plant level and recruitment was unaffected.
5. Weevil density had no effect on plant performance, most likely due to strong intraspecific competition, and there were no synergistic effects between the two herbivores.
6. Overall, attack by C. scrobicollis was more detrimental to A. petiolata growth, seed output, and survival than attack by C. alliariae . Consequently, C. scrobicollis has been prioritised as a potential biocontrol agent for control of A. petiolata in North America.     

Positive interactions can increase size inequality in plant populations

1.  Large variation in the size of individuals is a ubiquitous feature of natural plant populations. While the role of competition in generating this variation has been studied extensively, the potential effects of positive interactions among plants, which are common in high-stress environments, have not been investigated.
2.  Using an individual-based 'zone-of-influence' model, we investigate the effects of competition, abiotic stress and facilitation on size inequality in plant monocultures. In the model, stress reduces the growth rate of plants, and facilitation ameliorates the effects of stress. Both facilitation and competition occur in overlapping zones of influence. We tested some of the model's predictions with a field experiment using the clonal grass Elymus nutans in an alpine meadow.
3.  Facilitation increased the size inequality of model populations when there was no density-dependent mortality. This effect decreased with density as competition overwhelmed facilitation. The lowest size inequality was found at intermediate densities both with the model and in the field.
4.  When density-dependent mortality was included in the model, stress delayed its onset and reduced its rate by reducing growth rates, so the number of survivors at any point in time was higher under harsh than under more benign conditions. Facilitation increased size inequality during self-thinning.
5.   Synthesis . Our results demonstrate that facilitation interacts with abiotic stress and competition to influence the degree of size inequality in plant populations. Facilitation increased size inequality at low to intermediate densities and during self-thinning.     

Host-marking behaviour as a quantitative signal of competition in the walnut fly Rhagoletis juglandis

Abstract.  1. Walnut-infesting flies in the Rhagoletis suavis species group actively re-use hosts for oviposition despite engaging in a genus-typical host-marking behaviour which, in other Rhagoletis groups, deters oviposition. In a study of the walnut fly, R. juglandis (Cresson), alternative hypotheses for the putative marking behaviour were evaluated.
2. The oviposition site attraction hypothesis proposes that the host mark guides females to oviposition sites on occupied fruit. The competition intensity signal hypothesis proposes that the host mark is an indicator of the level of competition to be incurred if fruit are re-used.
3. In a field cage, females were presented simultaneously with fruit previously exposed to 25 females that were also allowed to oviposit and engage in the putative marking behaviour, and control fruit on which females were allowed only to oviposit. The occurrence of host marking reduced a female's propensity to oviposit from 46% to just over 10%, consistent with the competition intensity signal hypothesis only.
4. In a laboratory assay, the duration of host marking was correlated positively with the size of a female's clutch. This result, also consistent with the competition intensity signal hypothesis, suggests that the amount of marking pheromone on a fruit is a reliable indicator of the number of eggs already deposited within.
5. In a second field-cage experiment, females were allowed to mark on fruit for 0, 10, 20, or 30 min and fruit were presented to test females. Whether or not females alighted on a particular host was not affected by the duration of marking; however, the frequency of both ovipositor probing and egg deposition decreased with increasing duration of marking. Consistent with the competition intensity signal hypothesis, this result suggests that the host mark permits females to assess the level of competition that a clutch will incur within re-used fruit.     

Bottom-up, top-down, and within-trophic level pressures on a cactus-feeding insect

Abstract.  1. The relative importance of host-plants and predators in the population dynamics of herbivorous insects, and the frequency and intensity of inter-specific competition among herbivores, have both been intensively studied and debated. The joint effects of bottom-up, top-down, and within-trophic level interactions, however, have rarely been integrated in a single system.
2. I studied the dynamics of the cactus bug ( Narnia pallidicornis ), a specialist feeder on tree cholla cactus ( Opuntia imbricata ), in response to variable host-plant quality, spider predation, and interactions with cactus-feeding beetles ( Moneilema appressum ). Previous work suggests that cactus reproductive effort (the proportion of meristems allocated to reproduction) is an important component of host-plant quality for Narnia. I conducted a 2-year field experiment to test the hypotheses that Narnia abundance is positively related to host-plant reproductive effort, and that interactions with predators and putative competitors alter the shape of this relationship.
3. I found strong support for the first prediction (positive Narnia– plant quality relationship) in both years, but neither predator removal nor beetle exclusion had detectable effects on this relationship in either year. I conclude that the dynamics of this insect herbivore are driven predominantly from the bottom-up, and that available data from this work and from previous studies are too variable to permit broad generalisations for the combined effects of host-plants, predation, and competition on herbivore dynamics.     

Positive effects of shrubs on plant species diversity do not change along a gradient in grazing pressure in an arid shrubland

Facilitative or positive interactions among species are driven mainly by the environmental amelioration or protection from grazing provided by nurse plants. Some studies have suggested that protection from grazing is inconsequential in water-limited environments because of low herbivore densities and their grazing effects. Others, however, argue that herbivores have a major effect on semi-arid plant communities, and that protection from grazing is a significant factor driving positive plant–plant interactions in such environments. We identified a gradient in grazing pressure in a semi-arid shrubland in south-eastern Australia along which we compared soil condition, incident radiation and plant composition beneath two nurse shrub species with open (shrub-free) interspaces. Our aim was to assess the degree of microclimatic amelioration provided by both shrubs, and changes in the interactions (intensity, importance and frequency) between both nurse shrubs and their understorey species, and their effects on species richness at the community level. Both the relative interaction intensity (RII) and interaction importance (I_imp) indices of plant–plant interactions were generally positive and independent of grazing pressure. Soil beneath both nurse plants had significantly greater indices of nutrient cycling and infiltration, and contained more C and N than soil in the open. Almost twice as many species occurred under the canopies of both shrubs (44 species) than in the open (23 species), and the composition of species differed significantly among microsites. Fifty-four percent of all perennial plant species occurred exclusively under shrubs. Our results suggest that environmental amelioration is a stronger driver of the facilitatory effect of shrubs on their understorey species than protection from grazing. Our conclusions are based on the fact that the substantial effect of plant–plant interactions on plant species richness was largely independent of grazing pressure. Irrespective of the underlying mechanism for this effect, our study illustrates the ecological role of shrubs as refugia for understorey plants in semi-arid environments and cautions against management practices aimed at reducing shrub populations.