Synoptic patterns of meiofaunal and macrofaunal abundances and specific composition in littoral sediments

During recent years, many investigations on small zoobenthos have been performed at the island of Sylt. As these studies were carried out sporadically over many years and as different extraction methods were used, comparisons of the results have been hampered. Therefore, in August/September 1986, 24 sites were sampled and evaluated using one quantitative method throughout. Sites range from mud to exposed sand and from the sublittoral to the supralittoral. Macrofauna and the taxa Plathelminthes, Polychaeta, and Oligochaeta are determined to species level. Macrofaunal (>0.5 mm) abundance is highest in mud and continuously decreases with increasing exposure to wave action. Meiofaunal (<0.5 mm) abundance is less variable. Nematoda dominate in mud and muddy sand, Copepoda in sheltered and exposed sand, other taxa only intermittently. Related to surface area, no correlation between macro-and meiofaunal abundance is apparent. Plathelminthes and Copepoda reach highest abundance per surface area in sand but their per volume density is higher in mud and muddy sand. Related to sediment volume instead of surface area, the meiofaunal abundance pattern is very similar to the macrofaunal pattern. The faunal composition changes gradually along the tidal gradient without general faunal boundaries. On an averange, the faunal similarity of neighbouring sites is highest in Oligochaeta and lowest in Plathelminthes. Presumably, Oligochaeta tolerate wider ranges of environmental factors. This may explain the low number of oligochaete species. On the other hand, Plathelminthes seem to adapt to relatively narrow ranges of factors and their species richness is highest. Because of macrofaunameiofauna interaction it is suggested that the meiofaunal assemblage will be least stable in mud and muddy sand, and most stable in exposed sand.     

Utility of ITS region sequence and structure for molecular identification of <Emphasis Type="Italic">Tilia</Emphasis> species from Hyrcanian forests,Iran

Nucleotide sequences from the internal transcribed spacers (ITS) of nuclear ribosomal DNA and 5.8S gene were used to infer the phylogeny of Tilia species (represented by 13 distinct populations) growing in different geographical areas of Hyrcanian forests in northern Iran. Four well-supported lineages were revealed, including that of a new species, T. hyrcana, with stellate trichomes on both sides of the leaves and petiole. T. hyrcana is a well-supported cladospecies, with the ITS sequence and secondary structure following the diagnosable phylogenetic species concept, and is also characterized by a distinct morphology. A controversial species is Tilia rubra subsp. caucasica, with three different forms—an assemblage of taxa characterized by a lack of stellate trichomes on leaves—while Tilia begonifolia is distinguished by stellate trichomes on the underside of both leaves and petiole. The fourth lineage group, T. dastyla, is characterized by the presence of trichomes on the style. A single taxon found in the west of the Hyrcanian forest region is similar to T. begonifolia, but due to the former being located in a distinct group, a reassessment of the diagnostic morphology is recommended. ITS sequence data also suggested a closer relationship between T. rubra and T. begonifolia. Compensatory base change analysis was not strong enough to separate individual species within the Tilia genus. In general, the study supports the utility of ITS sequence data and secondary structure as accessory taxonomic characteristics with which to help clarify the systematics of the Tilia genus.     

Meiobenthic gradients with special reference to Plathelminthes and Polychaeta in an estuarine salt marsh creek—a small-scale model for boreal tidal coasts?

Environmental conditions in salt marsh creeks are intermediate between the open tidal coast and estuaries. A large salt marsh creek at the island of Sylt (North Sea) was studied in order to test whether its fauna is more similar to that of the open tidal coast or to that of estuaries. Because of a sandy bar at the seaward opening, the tidal range is only 10 cm in the creek, and the water level never drops below the level of the sand bar. Zoobenthos in the sandy bottom and on the sandy shores was studied at both ends and in the middle of the creek. Polychaeta and Plathelminthes were determined to species level. On an average, 2115 metazoans were found below 10 cm² of surface area. At the seaward end of the creek, abundance and taxonomic composition are similar to that of the adjoining Wadden area. Nematoda are the dominant taxon, followed by Copepoda, Plathelminthes and Oligochaeta. Taxonomic composition is different at the landward end. Plathelminthes and Nematoda are most abundant followed by Copepoda. Both Oligochaeta and Polychaeta are scarce at these newly eroded sites. Plathelminth abundance at the landward end of the creek is exceptionally high (770–935·10 cm⁻²). Contrary to what is generally found in estuaries, the species density of Plathelminthes shows a significant increase toward the land. The species composition of Polychaeta and Plathelminthes indicates that the sites below mean high tide level of the creek correspond to the adjacent eulittoral Wadden area while the fauna of the supralittoral sites of the creek is similar to the fauna of supralittoral tidal coasts. Typical sublittoral species did not occur in the salt marsh creek. Thus, salt marsh creeks may be regarded as a small-scale model for the tidal coast. In context with the results obtained, the definition of estuaries is discussed.     

Assessment of Zn mobilization in the rhizosphere of Thlaspi caerulescens by bioassay with non-accumulator plants and soil extraction

This study used co-cultivated plants as a bioassay to test the hypothesis that the roots of the zinc-hyperaccumulating plant Thlaspi caerulescensmobilize Zn from less-available pools in the soil. Thlaspi caerulescens was grown in uncompartmentalised pots, or pots that were divided by solid or mesh barriers to limit the extent of root intermingling (rhizosphere interaction) with co-cultivated Thlaspi arvense or Festuca rubra. Thlaspi caerulescens did not increase the concentration of Zn in either indicator species, suggesting that T. caerulescens does not strongly mobilize Zn in its rhizosphere. The increase in the shoot mass of T. arvense when its roots were permitted to intermingle with those of T. caerulescens was explained by greater intensity of competition of T. arvense compared to T. caerulescens.There was no effect of co-cultivation with T. caerulescens on the shoot biomass of F. rubra. Despite the absence of increased Zn-availability to the co-cultivated species, the mass of Zn accumulated by T. caerulescens was 3-times greater than the mass of Zn depleted from the pool of extractable-Zn in the soil, measured by extraction with 1 M ammonium nitrate. The results are consistent with the hypothesis that the rapid Zn-uptake systems in the roots of T. caerulescens deplete the soluble-Zn at a rate equal to, or faster than that at which Zn is replenished to the soil solution via plant/microbially mediated mobilization or the Zn-buffering capacity of the soil.     

Infection with the fungal endophyte Epichloë festucae may alter the allelopathic potential of red fescue

Red fescue (Festuca rubra) is a perennial grass used as both forage and turfgrass. Asymptomatic plants of this species are systemically infected by the fungal endophyte Epichloë festucae, which has a beneficial effect on the infected plants. The aim of this study was to determine the effect of the endophyte Epichloë festucae on the allelopathic potential of F. rubra against four associated pasture species that are also considered as weeds in lawns, Trifolium pratense, Trifolium repens, Lotus corniculatus and Plantago lanceolata. Two experiments were designed to evaluate the allelopathic effect of extracts from the roots and leaves of endophyte‐infected (E+) and non‐infected (E?) plants on the germination and seedling growth of the four target species. Regardless of the endophyte status of the host plant, leaf extracts elicited a stronger reduction in germination and seedling growth than root extracts. Extracts from E+ plants reduced the speed of germination index of Trifolium spp. to a greater extent than those from E? plants. Radicle length of the target species was the parameter most affected by the presence of the endophyte in F. rubra. Root extracts from E+ plants had a greater inhibitory effect on the radicle growth of the target species than did root extracts from E? plants. A greater concentration in total phenolic compounds was found in the roots of E+ plants than of E?; however, this difference was not observed in the leaves. Thus, the allelopathic potential of F. rubra is altered in infected plants.     

Genetic differentiation between the ant Myrmica rubra and its microgynous social parasite

Hymenopteran inquiline species have been proposed to originate by sympatric speciation through intraspecific social parasitism. One such parasite, Myrmica microrubra, was recently synonymized with its Myrmica rubra host, because comparisons across Europe indicated insufficient genetic differentiation. Here, we use microsatellite markers to study genetic differentiation more precisely in a sample of Finnish M. rubra and its inquilines collected at two localities, supplemented with mitochondrial DNA sequences. The parasite had much lower genetic variation than the host at three of the four loci studied. Genetic differentiation between the host populations was moderate (F _ST = 0.089), whereas the parasite populations were more strongly subdivided (F _ST = 0.440). The host and parasite were highly genetically differentiated both across populations (F _ST = 0.346) and in strict sympatry (0.327, 0.364), a result that remained robust both in a haplotype network and in PCA ordination. Individual assignments of genotypes indicated that gene flow between sympatric host and inquiline populations is reduced by about an order of magnitude relative to the gene flow within the morphs. Our results suggest that the parasitic morph of M. rubra may be an incipient species, but it remains unclear to what extent the observed genetic differentiation between host and inquiline is due to possible assortative mating and selection against hybrids or to recurrent bottlenecking and genetic drift. We conclude that an explicitly functional species concept would be unambiguous in treating this inquiline as a full species, as it begets its own kind and maintains its integrity in spite of occasional interbreeding with the host.     

Assemblages of native bryophytes in secondary forests with introduced Quercus rubra

We have studied the diversity of bryophytes in planted Polish post‐agricultural forests dominated by the native Scots pine Pinus sylvestris and the alien (North American) red oak Quercus rubra. The planted sites would be suitable for a mesic coniferous forest (abbreviation: CFS) or mesic broadleaved forest (abbreviation: BFS). We analysed the structure and composition of the bryophyte assemblages in relation to forest site and substrate availability. Special attention was paid to the introduced Q. rubra as a host species for native bryophytes. A total of 54 bryophyte species (9 liverworts and 45 mosses) were found in the 90 plots ( = phytosociological relevés, 10 × 10 m in area; 45 at each forest site) studied. DCA analysis showed that the bryophyte assemblages of the P. sylvestris–Q. rubra secondary forest community differed significantly between CFS and BFS sites; the similarity of the composition of bryophyte species was 36.8%. The substrate preferences (epigeic, epixylic, epiphytic), as well as the growth form and life form of the recorded bryophytes, also differed between CFS and BFS, while the proportion of bryophytes that had a particular life strategy was very similar. The introduced Q. rubra was inhabited by 28 bryophyte species, including two liverworts. This tree hosted 64% of the CFS and 47% of BFS bryophyte flora and as a host for epiphytes the species successfully fulfilled the functional role of the native oaks (Q. robur and Q. petraea). Thus, the introduction of Q. rubra may contribute to the restoration of post‐agricultural forests and to the conservation of epiphytic bryophyte species. On the other hand, the negative impact of Q. rubra observed on the ground flora (including bryophytes) puts the benefits of Q. rubra for the conservation of native biodiversity in general in question.     

The improvement and utilization in forestry of nitrogen fixation by actinorhizal plants with special reference toAlnus in Scotland

Summary Alnus species are used widely in Britain for land reclamation, forestry and other purposes. Rapid juvenile growth of the AmericanAlnus rubra makes it an attractive species for planting on N-deficient soils, particularly those of low organic content. In small plot trials, this species is nodulated by indigenous soil frankiae as effectively asAlnus glutinosa. Over a three year period both species return similar amounts of N to the ecosystem, estimated at up to 10–12 kg N ha^–1. Several strains ofFrankia have been isolated from local (Lennox Forest)A. rubra nodules. These differ morphologically and in their growth on different culture media, both from each other and fromA. glutinosa nodule isolates. AllAlnus isolates, however, have a total cellular fatty acid composition qualitatively similar to some other Group B frankiae. Glasshouse tests in N free culture suggest thatA. rubra nodules formed after inoculation of seedlings with American spore (–) isolates are three times more effective in N fixation than those inoculated with LennoxA. rubra spore (+) nodule homogenates. By contrast, the early growth of seedlings inoculated with spore (–)Frankia strains suggests at best a 35% improvement in N fixing activity over seedlings inoculated with LennoxA. rubra nodule isolates. Nevertheless, this improvement in activity, together with the better performance of seedlings inoculated with isolates compared with those treated with crushed nodule preparations, suggest that it would be worthwhile commercially to inoculate nursery stock with a spore (–)Frankia strain.     

Myrmica rubra microhabitat selection and putative ecological impact

1. Myrmica rubra (European fire ant) has invaded northern latitude coastal areas in North America. This macroscale distribution suggests that M. rubra is moisture‐ and temperature‐limited, but the distribution of the invaded range may reflect the legacy of original introduction locations preserved by limited dispersal. 2. This study examined a two‐decade population change in M. rubra (1994–2015) and the microscale abiotic (moisture and temperature), biotic (plants), anthropogenic (pesticide) and physiological (thermal tolerance) limits on the invasion at the Tifft Nature Preserve in Buffalo, NY (U.S.A.). Changes in the abundance of native ants and other invertebrates were also examined. 3. Despite localised declines with pesticide treatments, overall M. rubra forager abundance increased 27% between 1994 and 2015. Abundance increased the most in the warmest areas (native ants were unaffected by temperature), but M. rubra colony locations were strongly linked to higher soil moisture and lower soil temperature. Myrmica rubra ants also exhibited relatively low thermal tolerances consistent with high‐latitude and high‐elevation ants. 4. Where local M. rubra populations increased the most, native ant species decreased, and where local M. rubra populations declined, native ant species increased. Some arthropod species had lower abundance with M. rubra presence, but the impacts were less striking. 5. Myrmica rubra population growth was promoted at the microhabitat scale where relatively higher temperatures prompted foraging, and relatively lower temperatures and high moisture supported nesting. These results suggest that macroscale M. rubra invaded‐range distributions in northern climates near coastal areas are replicated at the microscale where the ant prefers cooler, moister microsites.     

THE GROWTH AND MINERAL COMPOSITION OF THREE SPECIES OF SPERGULARIA AS AFFECTED BY SALINITY AND NUTRIENTS AT HIGH SALINITY

The effects of salinity on the growth of three species of Spergularia and the effects of some levels of different nutrients on the growth and mineral composition of the three species under high salinity were investigated in sand culture. In response to the effects of salinity, S. marina and S. rupicola behaved like true halophytes in terms of survival, mean dry weight and plant part weight ratios. Spergularia rubra also behaved like an ecological halophyte in terms of dry weight but in terms of survival, it behaved like a glycophyte. The poor growth exhibited by the halophytes, S. marina and S. rupicola under high salinity condition was significantly reversed with the addition of most nutrients, while the glycophyte, S. rubra showed no favorable response, rather the combination of some nutrients further reduced its growth. The increased growth in S. marina was as a result of increased root growth while in S. rupicola it was as a result of increased root and stem growth. High salinity conditions resulted in a marked reduction in calcium and potassium content and an increase in sodium and chloride content of leaves of the species. Addition of many of the nutrients to plants under high salinity condition generally resulted in increased sodium, calcium, potassium, and magnesium content of leaves, except that the addition of phosphate resulted in decreased amounts of calcium and magnesium. Results are discussed in relation to the salt tolerance of the species and how the supply of nutrients to these species in their natural habitats might help them overcome high salinity problems.     

Effects of leguminous ground cover competition on red birch and soil nutrient status in the nursery

Legumes as ground cover are regularly planted to increase nitrogen economy of crops and to improve soil. In the present study various clover species were evaluated as vegetative ground cover in nursery field production of micropropagated red birch (Betula pubescensEhrh. f. rubraUlvinen f. nova) in two 2-year experiments. The clover species and cultivars, Trifolium pratenseL. ‘Bjursele’, T. repens L. ‘Jogeva’, T. repens L. ‘Sonja’, T. hybridum L. ‘Frida’, T. incarnatum L. ‘Opolska’, T. resupinatum L. and T. subterraneum L. were compared to grass sod Festuca rubra L. ‘Ensylva’ and to a coverless ground (control). The last one was kept weed free by hand hoeing. Birch (leaves, stems, branches and roots) and soil nutrient concentrations (N, P, K, Ca, Mg and Fe) were analysed and nutrient ratios in birch determined. The annual clovers, T. incarnatumL., T. resupinatumL. and T. subterraneumL., provided about the same nutrient status in birch as did the control. Perennial clovers and grass were strong competitors with trees. High levels of P and Mg in birch leaves relative to N concentration were typical for poorly growing seedlings. Neither annual nor perennial clovers did generally improve soil nutrient status. This revised version was published online in June 2006 with corrections to the Cover Date.     

Moisture conditions and the presence of bryophytes determine fescue species abundance in a dry calcareous grassland

Festuca ovina is the abundant matrix-forming species and F. rubra a subordinate species in shallow-soil calcareous grasslands. F. pratensis is a transient species, occurring sparsely in this community. We hypothesised that the different abundances of these three species are primarily due to the differential effect of moisture conditions on their germination and early establishment, and that the effect of the pattern of rainfall intensity depends on the presence or absence of a bryophyte layer. We studied the dependence of the germination and establishment of the three fescue species on the moisture conditions both in the laboratory and in the patches of intact grassland community (microcosms). In a laboratory germination experiment, F. pratensis showed the highest, F. rubra , the intermediate and F. ovina, the lowest drought tolerance. In microcosms, the establishment of F. ovina was the highest. At the same time, the annual mortality of seedlings of F. ovina was the lowest. All three species responded positively to an increasing irrigation level. Differently from F. ovina, F. rubra showed a positive response only in plots from which the bryophyte layer had been removed, while F. pratensis responded positively to both irrigation and bryophyte removal. We conclude that moisture conditions have a differential effect on the three fescue species mainly in the seedling establishment, not in the germination phase. For the successful establishment of F. rubra and F. pratensis, the coincidence of high rainfall and local disturbance, removing bryophytes, is required. The presence or absence of bryophytes had no effect on establishment in dry years, while in rainy years the removal of bryophytes has a clear positive effect.     

Plathelminthes and Plathelminthomorpha — paraphyletic taxa

The concept of monophyly of Plathelminthes and Plathelminthomorpha (Plathelminthes and Gnathostomulida) is critically re-analysed. Based on new morphological and molecular evidence, it is concluded that both taxa are paraphyletic assemblages. The Acoelomorpha (Xenoturbellida?, Nemertodermatida and Acoela) are considered as the earliest offshoot of the bilaterian stem line, primarily lacking paired cerebral ganglia, an orthogonal nervous system and ultrafiltration nephridia. They are followed by Plathelminthes sensu stricto (i.e. the rhabditophoran turbellarians and Neodermata), Catenulida, and Gnathostomulida, the latter group is probably the sister group of the Syndermata (Rotifera and Acanthocephala). The respective characters, as well as the implication for the proposed stem species of the Bilateria are outlined and discussed.     

Effect of the breeding system on the genetic structure in threeCephalanthera spp. (Orchidaceae)

The genetic variation of seven enzymes for a total of nine loci was investigated in three species of terrestrial orchids of the genusCephalanthera:C. longifolia, C. rubra, andC. damasonium. These species are characterized by presenting different breeding types: outbreeding , outbreeding with facultative vegetative reproduction, and inbreeding, respectively. Electrophoretic evidence points to a difference in the behaviour of each of the three species which seems strictly related to the breeding type. On the basis of our resultsC. longifolia behaves as a normal outbreeder, whileC. rubra presents the influence of vegetative reproduction in some populations and not in others.C. damasonium shows a total lack of both among and within populations genetic variation, which is most probably due to the autogamic breeding type.     

Seed dispersal and realized gene flow of two forest orchids in a fragmented landscape

• Species with vast production of dust‐like windborne seeds, such as orchids, should not be limited by seed dispersal. This paradigm, however, does not fit recent studies showing that many sites suitable for orchids are unoccupied and most seeds land close to their maternal plant. To explore this issue, we studied seed dispersal and gene flow of two forest orchid species, Epipactis atrorubens and Cephalanthera rubra, growing in a fragmented landscape of forested limestone hills in southwest Bohemia, Czech Republic.
• We used a combination of seed trapping and plant genotyping methods (microsatellite DNA markers) to quantify short‐ and long‐distance dispersal, respectively. In addition, seed production of both species was estimated.
• We found that most seeds landed very close to maternal plants (95% of captured seeds were within 7.2 m) in both species, and dispersal distance was influenced by forest type in E. atrorubens. In addition, C. rubra showed clonal reproduction (20% of plants were of clonal origin) and very low fruiting success (only 1.6% of plants were fruiting) in comparison with E. atrorubens (25.7%). Gene flow was frequent up to 2 km in C. rubra and up to 125 km in E. atrorubens, and we detected a relatively high dispersal rate among regions in both species.
• Although both species occupy similar habitats and have similar seed dispersal abilities, C. rubra is notably rarer in the study area. Considerably low fruiting success in this species likely limits its gene flow to longer distances and designates it more sensitive to habitat loss and fragmentation.

     

Interspecific variations in responses of <Emphasis Type="Italic">Festuca rubra</Emphasis> and <Emphasis Type="Italic">Trifolium pratense</Emphasis> to a severe clipping under environmental changes

Understanding and predicting possible responses of grassland species to global change is of important meaning for adapting grassland management to a changed and changing environment. A laboratory clipping experiment was conducted to examine the interspecific responses in an ecological context of competition and environmental changes. Festuca rubra and Trifolium pratense, either in monoculture or two- and three-species mixtures, were grown in three environmental combinations (ambient and increased temperature, repetitive N supply, and simulated acid rain), respectively. After a growth time of three months, plants were clipped at the height of 1.0 cm above soil surface. Plant height and aboveground biomass prior clipping, and survival rate and regrowth (height and biomass) after clipping were analyzed. F. rubra and T. pratense responded differently in compensatory growth and competition intensity to environmental change and co-existing species. The differences in their physiological and ecological traits may account for species-dependent responses. The present study emphasizes that predicting the plant assemblage response in the face of global change requires in understanding the integrating effects of abiotic and biotic factors.     

Seed selection for grassland restoration: competitive effect of a dominant grass is mediated by seed source and nutrient availability

Cultivars are selected for advantageous traits, which may enable them to have a greater competitive effect (CE), particularly under high nutrient conditions. Sites with high nutrient availability may favor the development of dominant grasses rather than subordinate species, especially in nutrient‐limited systems, such as calcareous grasslands. The aim of this experiment was to determine whether seed source influences the CE of a dominant grass on a subordinate species, and whether this relationship is mediated by nutrient availability. A greenhouse experiment with three nutrient levels was established; Dianthus carthusianorum was chosen as the subordinate “phytometer” to detect variation in the CE of different Festuca rubra seed sources. The grass species was sourced from 13 cultivar and 12 commercially propagated, but not selected, wild sources. When CE was calculated from biomass, propagated wild seed sources of F. rubra had a greater CE on the subordinate species than cultivars for medium and high nutrient levels. Based on phytometer height, propagated wild seed sources of F. rubra had a greater CE under all three nutrient levels. Our findings do not support the general notion that cultivars are more competitive than wild genotypes. Thus, the cultivar F. rubra may facilitate the establishment of other species during grassland restoration, particularly under elevated nutrient conditions.     

A comparison of symbiotic nitrogen fixation in Scotland inAlnus glutinosa andAlnus rubra

Summary Alnus glutinosa andAlnus rubra growing in the field in Scotland show specific nitrogenase activities of the same order of magnitude. The period of maximum potential nitrogenase activity coincides with that of maximum growth in late Spring and Summer. It is suggested that the retention of nitrogenase activity into the Autumn when growth has virtually ceased may be important as a contribution to the nitrogenous reserves of the tree.Bioassay of different Scottish soils, all collected from the locality of natural stands ofAlnus glutinosa, showed wide variation in the nodulation of seedlings, although generally a soil poor for nodulation ofAlnus glutinosa generally gave poor nodulation ofAlnus rubra. Soils of pH 4.5 to 6.5, best suited for growth and nitrogen fixation of the two species, often gave nodules showing highest specific nitrogen fixing activity. Young (2 to 3 year old) plants in glasshouse or controlled environment cabinet, inoculated withAlnus glutinosa endophyte, differed from mature field grown plants, however, sinceAlnus rubra required a much larger (up to 2.5 times) mass of root nodules to fix a unit quantity of N. Microscopic comparison of the nodules of glasshouse plants showed that the proportion of cells containing the vesicular (nitrogen fixing) form of the endophyte was only slightly lower inAlnus rubra than inAlnus glutinosa and it is suggested that the differences in specific nitrogen fixing activity between the two species may reflect some incompatibility of function of theAlnus glutinosa endophyte when in symbiosis withAlnus rubra.     

Limited evidence for ecosystem‐level change on reefs exposed to Haliotis rubra (‘blacklip abalone’) exploitation

It is increasingly recognized that fisheries must take the broader ecosystem into account for sustainable management of marine systems, requiring an understanding of the interaction between fished species and other organisms. This study uses a correlative approach to investigate potential interactions between benthic organisms and Haliotis rubra, a dominant herbivore that is the subject of a large and valuable commercial fishery in south‐eastern Australia. Specific emphasis was placed on understanding associations between H. rubra and understorey organisms, because particular understorey algae (crustose coralline algae) provide critical habitat for H. rubra larval recruitment and juvenile ecology. Broad‐scale surveys along the 6–8 m depth contour (the depth range where H. rubra fishing activity is intense) were conducted across four regions (separated by 10⁴?10⁵ m), including at least 10 sites (separated by 10²?10³ m) within each region. Positive correlations between H. rubra and crustose coralline algae were found, while negative correlations were observed between H. rubra and sessile invertebrates and understorey algae. While significant, these associations were weak and H. rubra abundance generally only explained a small proportion of the variability in the abundance of understorey organisms (r² 0.02–0.30). H. rubra abundance also had a minor influence on community‐level understorey patterns in comparison with differences in community structure attributable to regional variation. Patterns of H. rubra abundance and benthic community structure were also examined in relation to depth at a restricted number of sites. At sites where differences in understorey groups were evident, H. rubra abundance also varied significantly, highlighting the issue of confounding when contrasting patterns of understorey abundance using a correlative approach. Further manipulative experiments are required to confirm causal relationships; however, the available correlative evidence suggests limited ecosystem effects of H. rubra depletion at the scale of individual reefs.     

POLLINATION BIOLOGY,SEED PRODUCTION,AND POPULATION STRUCTURE IN QUEEN-OF-THE-PRAIRIE,FILIPENDULA RUBRA (ROSACEAE) AT BOTKIN FEN,MISSOURI

Queen-of-the-prairie, Filipendula rubra (Rosaceae), is a perennial, rhizomatous herb restricted to calcareous fens, spring seeps, and wet prairies, and is distributed in the northcentral United States from Iowa to New York. Because of the scarcity of adequate habitat, F. rubra is a rare and endangered species in several states. At Botkin Fen and other fens in Missouri, F. rubra possesses limited sexual reproduction. A study was conducted to determine causal mechanisms of its limited seed production that may be barriers to establishing this rare and endangered species in other fens in Missouri. A genetic-ecological hypothesis was supported. F. rubra was found to be self-incompatible. At Botkin Fen, most inflorescences produced very few seeds because most (85.4%) belonged to a single clone. Pollinators (Lassioglossum sp.) had a high probability of transporting pollen with the same self-incompatibility alleles. A few genetically different clones did produce significant quantities of seed. However, this seed, with its significant pool of genetic variability, was not recruited into the population. Apparently, competition from the dense sedges and grasses prevented the establishment of new seedlings.