---

• Uchida K and Ushimaru A. (2014) Journal of Applied Ecology x:x-x.
1. Although traditional land-use and management practices are known to enhance environmental heterogeneity in agricultural lands, loss of heterogeneity resulting from recent land-use changes can lead to spatiotemporal β-diversity losses (biotic homogenization). However, patterns and causes of β-diversity loss, especially at a within-field scale, have remained unexplored. 2. We examined β-diversity declines in plants, butterflies and orthopterans resulting from land abandonment and intensification in semi-natural grasslands on paddy margins at a within-field scale in relation to changes in disturbance frequency and surrounding landscapes by surveying eight abandoned, thirteen traditional and ten intensive paddy terraces. Three indices, the additive partitioning of species richness (βadd) and the turnover (βtu) and nestedness (βne) components of Jaccard dissimilarity of plants and herbivores, were used to evaluate different aspects of β-diversity. We hypothesized that biotic homogenization due to abandonment and intensification would be expressed as low spatiotemporal βadd and βtu, low spatial βne and high temporal βne. 3. The spatial and temporal βadd of plants and herbivores were higher in traditional terraces than in others. Changes in mowing frequency caused direct decline in butterfly βadd and indirectly decreased herbivore βadd through diminished plant βadd in abandoned and intensified terraces. 4. Neighbouring forests increased plant spatial and butterfly temporal βadd and βtu. Abandoned terraces had higher orthopteran spatial and plant temporal βtu than traditional terraces. Mowing frequency generally did not influence spatial and temporal βtu of plants and herbivores. In intensive terraces, temporal βne of plants were higher than in traditional terraces, suggesting dominance by plant species sharing similar phenologies. 5. Synthesis and applications. These results suggest that traditional mowing practices are essential for maintaining plant and herbivore βadd, and consequently, species pools within terraces. The higher spatial or temporal species turnover of plants and orthopterans in abandoned terraces and higher temporal plant nestedness in intensive than in traditional terraces were not due to increases in species pools within terraces. Because within-terrace environmental heterogeneity is thought to remain in abandoned and intensive terraces, we recommend reduction in mowing frequency in intensive terraces and reintroduction of mowing in abandoned terraces for biodiversity restoration.

---

• Ohara RG and Ushimaru A. (2015) Applied Vegetation Science x:x-x.
Questions: Are there positive edge effects on β- and α-diversity of plant species across seminatural grassland-secondary forest gradients located within a traditional agricultural landscape? Is β-diversity (dissimilarity in plant species composition between two given plots) well explained by environmental differences for both grasslands and forests? Do gradients of environmental factors created by forest edges influence spatial variation in α-diversity from edges to grasslands and forests? Location: Seminatural grasslands and secondary forests (or conifer plantation) around paddy fields in a traditional agricultural landscape, western Japan. Methods: Vegetation survey and environmental (soil water content and sunlight) measurements were conducted on 11 transects passing through grassland-forest ecosystems. Spatial variations in plant β and α-diversity and environmental variables and their relationships were examined. Results: We quantitatively demonstrated common positive edge effects on β-diversity in both grasslands and forests. Our results suggest that among-edge variations in resource conditions, which likely depend on edge aspect and forest type, promote high grassland plant β-diversity in the vicinity of edges. However, the high β-diversity of forest-floor communities in the proximity of edges cannot be explained by this mechanism. Alpha-diversity tended to be lower within grassland plots located around edges than in grassland interiors, while the richness of the forest community increased toward the edge. This α-diversity pattern was suggested to be attributable to resource gradients created by the forest edges. Conclusions: Although plant α-diversity may not be consistently high in the vicinity of edges, β-diversity in the communities bordering grassland–forest edges will generally be high in the traditional agricultural landscape. Thus, maintenance of environmental heterogeneity among grassland–forest edges in agricultural ecosystems should be prioritized for local biodiversity conservation.

---

• Fukumoto S, Ushimaru A and Minamoto T. (2015) Journal of Applied Ecology 52:358-365.
1. To prevent the invasion of exotic species causing a decline in an endangered endemic spe- cies, it is important to determine the distribution of both species at an early stage, when the density of the exotic species is still low, and to manage the invasion immediately. However, dis- tinguishing between closely related species is difficult because they share similar characteristics. 2. The identification of DNA fragments sampled from a body of water (environmental DNA) has become a popular technique for rapidly determining the distribution of a target species. In this study, we analysed environmental DNA in water samples from 37 sites across the Katsura River basin in Japan. We used TaqMan real-time PCR to distinguish the Japa- nese giant salamander Andrias japonicus from the closely related Chinese giant salamander Andrias davidianus, which is known to invade Japanese rivers and hybridize with the Japanese species. 3. In environmental samples, we detected mtDNA of the endemic species at 25 sites and mtDNA of the exotic species at nine sites. The DNA detection sites were concentrated in the upstream region. The exotic species DNA was found beyond the limits of an earlier capturing survey. 4. Synthesis and applications. Using environmental DNA to monitor the two salamander spe- cies requires less time and effort than traditional surveys, so a wide-ranging survey can be conducted rapidly. Our results showed that performing three environmental DNA surveys for each site between autumn and winter is desirable for giant salamanders. Further collection of environmental DNA, in combination with conventional population surveys, will provide valu- able information that can help protect rare endemic species in a variety of aquatic ecosystems and can help monitor the invasion of exotic species.

---

• Uchida K and Ushimaru A. (2014) Ecological Monographs 84:637-658.
Declines in plants and herbivorous insects due to land-use abandonment and intensification, have been studied in agricultural areas worldwide. We tested four hypotheses, which were complementary rather than mutually exclusive, to understand the mechanisms driving biodiversity decline due to abandonment and intensification. These predict that biodiversity decline is caused by a decline in resource diversity, changes in disturbance regime, surrounding landscape conversion and a decrease in biomass production. We compared plant richness and butterfly and orthopteran richness and diversity among three land-use types in semi-natural grasslands: abandoned, traditional, and intensified terraces. Then, we examined effects of changes in resource (plant) richness, frequency of disturbance (mowing), and surrounding landscapes on butterfly and orthopteran diversity to understand the mechanisms driving decline after land abandonment and intensification. Plant and herbivore richness and diversity were significantly lower in abandoned and intensified grasslands than in traditional grasslands. This trend was consistent throughout the seasons in both years of study. Changes in mowing frequency and the surrounding landscape explained plant richness decline as a consequence of land abandonment and intensification. Declines in herbivorous insects were explained by plant richness declines and changes in mowing frequency but not by landscape changes. Plant and herbivore richness were maximized at an intermediate mowing frequency (ca. twice per year), a typical practice on traditional terraces. This is the first report demonstrating that the intermediate disturbance hypothesis well explained the biodiversity declines in agricultural ecosystems. The richness and diversity responses of herbivore functional groups to plant richness, mowing frequency, and surrounding landscapes were generally inconsistent with predictions. We found significant trends in which butterfly and orthopteran species with low abundance in traditional terraces were lost in abandoned and/or intensive terraces. This may suggest that the number of individuals of most herbivorous species decreased randomly with respect to life-history traits following a decline in plant richness due to changes in disturbance frequency. This study demonstrates that declines in herbivorous insects can be explained by multiple factors and provides a unified explanation for biodiversity decline in both abandoned and intensified lands, which have often been studied separately.

---

• Osawa T, Mitsuhashi H and Ushimaru A. (2014) Plant Ecology 215: 597-612.
Although several studies have demonstrated that disturbance contributes to species diversity, little emphasis has been placed on the identification of species coexistence mechanisms related to life history traits. In this study, we compared species richness and components of plant communities around river confluences to explore how disturbance promotes the coexistence of species with differing life history traits. Sites upstream and downstream of confluences are ideal for such comparisons because they draw on the same species pools and have similar ambient conditions, but differ markedly in the extent of flooding disturbance. We compared sites upstream and downstream of confluences by calculating species richness and community similarity indices for several life history traits in both summer and spring. In summer, the richness of all species combined, of annuals, and of summer flowering species was higher downstream from confluences than upstream, but this was not the case for perennials. Similarity analyses suggested that plant communities are constructed according to a neutral process, whereby interactions between coexisting species are neutral. However, in spring, species richness was similar upstream and downstream of confluences for all life history traits. Similarity analyses suggested that under these circumstances, the communities were constructed through a species sorting process; i.e., each life history trait had a distinct habitat preference. Thus, the relative strengths of different community assembly processes may change seasonally. We concluded that species groups differing in their responses to disturbance may coexist in a single community. Thus, community structuring following disturbance may involve two processes: a neutral and a species-sorting process. The relative importance of each may vary between species life history traits and between seasons, and the interaction may account for current community structures.

---

• Ushimaru A, Kobayashi A and Dohzono I. (2014) American Naturalist184: 258-267.
Although land-use changes such as urbanization have dramatically altered plant-pollinator interactions, little is known about their effects on pollen limitation and floral traits. In this study, we examined pollinator visit frequency, reproductive success, and floral trait measurements in 12 populations of the annual andromonoecious Commelina communis in an urban–rural area. Pollinator and mate availability decreased significantly with developed land area around the study site. Most urbanized populations suffered from significant pollinator-limited male and/or female reproductive success. High fruit set in urbanized populations may suggest the presence of high reproductive assurance by selfing. The stigma height and degree of herkogamy significantly decreased with increased pollinator limitation. Petal length, anther height, and/or the pollen:ovule ratio tended to be low in pollinator- and mate-limited urban populations. One urban population with high pollinator availability had flowers with higher herkogamy and stigma height compared to rural populations. These results suggest that urbanization may provide diverse selective forces that could affect the phenotypic variation in floral traits.

---

• Uematsu Y and Ushimaru A. (2013) Ecological Applications23:1357-1366.
Examining the causes of interspecific differences in susceptibility to bidirectional land-use changes (land abandonment and use-intensification) is important for understanding the mechanisms of global biodiversity loss in agricultural landscapes. We tested the hypothesis that rare (endangered) plant species prefer wet and oligotrophic areas within topography- and management-mediated resource (soil water content, nutrient, and above-ground biomass) gradients, making them more susceptible to both abandonment and use-intensification of agricultural lands. We demonstrated that topography and management practices generated resource gradients in semi-natural grasslands around traditional paddy terraces. Terraced topography and management practices produced a soil moisture gradient within levees and a nutrient gradient within paddy terraces. Both total and rare species diversity increased with soil water content. Total species diversity increased in more eutrophied areas with low above-ground biomass, whereas rare species diversity was high under oligotrophic conditions. Rare and common species were differentially distributed along the human-induced nutrient gradient, with rare species preferring wet, nutrient-poor environments in the agricultural landscapes studied. We suggest that conservation efforts should concentrate on wet, nutrient-poor areas within such landscapes, which can be located easily using land-use and topography maps. This strategy would reduce the costs of finding and conserving rare grassland species in a given agricultural landscape.

---

• Fukada C, Kaneko S, Yokogawa M, Kobayashi T, Ushimaru A and Isagi Y. (2012) Conservation Genetics Resources 4:495-497.
Arisaema minus is a vulnerable perennial herb species endemic to Japan. We isolated and characterized ten microsatellite loci in this species. The number of alleles per locus ranged from 3 to 28, with an average of 12.8. The observed heterozygosities ranged from 0.472 to 0.861 with an average of 0.703, and the expected heterozygosities ranged from 0.486 to 0.962 with an average of 0.779. The total exclusionary powers of all ten loci, when no parental information existed or when one parent was known, were 0.99940154 and 0.99998730, respectively, indicating the high resolution of these microsatellite loci. This marker set is potentially useful for conservation genetics, reproductive studies, and the development of effective A. minus conservation programs.

---

• Iwata T, Nagasaki O, Ishii HS, and Ushimaru A. (2012) New Phytologist 193:196-203.
• Despite the wide inflorescence diversity among angiosperms, the effects of inflorescence architecture (three-dimensional flower arrangement) on pollinator behaviour and mating success have not been sufficiently studied in natural plant populations. • We investigated how inflorescence architecture affected inter- and intra-plant pollinator movements and consequent mating success in a field population of Spiranthes sinensis var. amoena (S. sinensis). In this species, the flowers are helically arranged around the stem, and the degree of twisting varies largely among individuals. The large variation in inflorescence architecture in S. sinensis results from variation in a single structural parameter, the helical angle (the angular distance between neighbour-flower directions). • The numbers of visits per inflorescence and successive probes per visit by leaf-cutting bees decreased with helical angle, indicating that individual flowers of tightly twisted inflorescences received less visitations. As expected from pollinator behaviour, pollinia removal and fruit set of individual flowers decreased with helical angle. Meanwhile, geitonogamy decreased in tightly twisted inflorescences. • Our novel findings demonstrate that natural variation in inflorescence architecture significantly affects pollinator behaviour and reproductive success, suggesting that inflorescence architecture can evolve under pollinator-mediated natural selection in plant populations. We also discuss how diverse inflorescence architectures may have been maintained in S. sinensis populations.

---

• Tsuji M, Ushimaru A, Osawa T, and Mitsuhashi H. (2011) Landscape and Urban Planning 103:318-325.
Major drivers of amphibian declines via urbanization include land-use changes that cause loss, fragmentation, splits, and degradation of habitat. The effects of these changes in habitat conditions on the persistence of populations are expected to differ among species depending on their dispersal habits: species with strong site fidelity would likely be more affected by habitat loss and degradation, whereas species with highly dispersive habits would be more threatened by habitat fragmentation and split (the dispersal-dependent-decline hypothesis). To test this hypothesis, we examined the distribution patterns of two paddy-associated frog species (Pelophylax nigromaculatus and Hyla japonica) with different dispersal habits along a rural–urban gradient of the Osaka–Kobe metropolitan area, Japan. Our results partially support the dispersal-dependent-decline hypothesis in that the species with strong site fidelity, i.e., P. nigromaculatus, was threatened by habitat-area (agricultural field) decline and habitat-quality degradation (prevalence of concrete levees) rather than by habitat fragmentation. To the best of our knowledge, our study is the first report that suggests paddy-associated amphibian declines via urbanization in the Asian region. However, the second half of the hypothesis, i.e., that the dispersive species (H. japonica) would be more strongly affected by habitat fragmentation via roads and habitat split via declines in surrounding forests was not supported. The lack of support for this portion of the hypothesis may be due to the high adaptability of H. japonica to artificial landscapes. We discuss the value of the dispersal-dependent-decline hypothesis for conservation planning in agricultural lands of urban areas.

---

• Osawa T, Mitsuhashi H, Uematsu Y and Ushimaru A. (2011) Ecological Informatics 6:270-275.
Species occurrence data sets tend to contain a large proportion of zero values, i.e., absence values (zero-inflated). Statistical inference using such data sets is likely to be inefficient or lead to incorrect conclusions unless the data are treated carefully. In this study, we propose a new modeling method to overcome the problems caused by zero-inflated data sets that involves a regression model and a machine-learning technique. We combined a generalized liner model (GLM), which is widely used in ecology, and bootstrap aggregation (bagging), a machine-learning technique. We established distribution models of Vincetoxicum pycnostelma (a vascular plant) and Ninox scutulata (an owl), both of which are endangered and have zero-inflated distribution patterns, using our new method and traditional GLM and compared model performances. At the same time we modeled four theoretical data sets that contained different ratios of presence/absence values using new and traditional methods and also compared model performances. For distribution models, our new method showed good performance compared to traditional GLMs. After bagging, area under the curve (AUC) values were almost the same as with traditional methods, but sensitivity values were higher. Additionally, our new method showed high sensitivity values compared to the traditional GLM when modeling a theoretical data set containing a large proportion of zero values. These results indicate that our new method has high predictive ability with presence data when analyzing zero-inflated data sets. Generally, predicting presence data is more difficult than predicting absence data. Our new modeling method has potential for advancing species distribution modeling.

---

• Osawa T, Mitsuhashi H, Niwa H and Ushimaru A. (2011) Aquatic Conservation: Marine and Freshwater Ecosystems 21 :358-363.
1. While flooding creates diverse habitats for plants and animals in riparian ecosystems, it presents a nuisance to human settlement. Thus, such areas that are frequently disturbed by flooding are more likely to experience strong artificial-control measures. 2. This hypothesis was tested and a method was developed to identify these areas using river maps. The effects of two terrain components that could potentially influence flooding frequency (the number of confluences and the degree of meandering per unit area) on both the distribution of threatened species and the degree of artificial modification were examined. These effects were examined at two different scales (grid sizes) using two riparian data sources. 3. The numbers of both threatened plant species and artificial constructions increased with the number of confluences and the degree of meandering per unit area. The number of threatened plant species was not correlated with the number of artificial constructions. 4. Additionally, the results suggest that confluences and meanderings provide different habitat conditions for threatened plant species. Confluences may provide more frequently disturbed habitats, whereas meanderings may provide relatively stable habitats. 5. Based on these results, through flooding, both confluences and meanderings are important for creating and maintaining the diverse habitats used by threatened species, but these habitats are also currently threatened by artificial controls. We propose that a per-unit-area confluence density and river length map is a helpful tool for locating local biodiversity hot spots. The identification of major reservoirs of biodiversity that are currently threatened can improve conservation and management planning.

---

• Ushimaru A and Genkai-Kato M. (2011) Evolutionary Ecology25:1161-1169.
An individual of modular organisms, such as plants and fungi, consists of more than one module that is sometimes physically and physiologically connected with each other. We examined effects of translocation costs, resource–fitness relationships and original resource conditions for modules on the optimal resource translocation strategy for reproductive success in modular organisms with simple models. We considered two types of translocation cost: amount-dependent and ratio-dependent costs. Three optimal resource translocation strategies were recognized: all resource translocation (ART), partial resource translocation (PRT), and no resource translocation (NRT). These strategies depended on the translocation cost, shape of resource–fitness curve, and original resource condition for each module. Generally, a large translocation cost and a concave resource–fitness relationship promoted NRT or PRT. Meanwhile, a small translocation cost and convex resource–fitness relationship facilitated ART. The type of translocation cost did not strongly affect the optimal resource translocation patterns, although ART was never an optimal strategy when the cost was ratio-dependent. Resource translocation patterns found in modular plants were discussed in the light of our model results.

---

• Osawa T, Mitsuhashi H, Niwa H and Ushimaru A. (2010) Ecological Research 25:1161 -1169.
In riparian areas, the distribution patterns of plant species are generally considered to depend on their flooding tolerance. Areas around river confluences are known to experience frequent and/or strong flooding events and provide diverse habitats for plants in riparian areas. However, the degree to which hydrophilic vegetation types increase their distribution around confluences may depend on their flooding tolerance. To test this hypothesis, we compared patch numbers and total areas of 10 vegetation groups between confluences and single-flow areas. The vegetation groups were classified on the basis of life form and morphology of dominant species. Additionally, we compared total area of natural bare ground (an index of flooding disturbance) between confluences and single-flow areas. We found that patch numbers of annual grass, forb, and vine, perennial grass and forb, and riparian forest vegetation, as well as total areas of annual forb and vine, perennial grass and forb, bamboo and riparian forest vegetation, and natural bare ground, were greater around river confluences than in single-flow areas. On the other hand, patch numbers of shrub vegetation and total areas of annual grass, perennial vine, willow, and shrub vegetation decreased around confluences. These results suggest that confluences enhance diverse, but not all, types of habitat for hydrophilic vegetation. Thus, river confluences are a key element in maintaining diverse riparian vegetation.

---

• Osawa T, Mitsuhashi H, Niwa H and Ushimaru A. (2010) The Open Ecology Journal 3: 48-58.
Although dendritic networks within ecosystems have typically been considered a special case of network topology, they have attracted a great deal of attention in recent years. These systems exhibit unique features in that both the nodes and branches provide distinct habitats. Within a river discontinuum context, river confluences, which are nodes of dendritic river networks, are hypothesised to have particular hydrodynamic traits that create heterogeneous habitats through a unique disturbance regime, although this hypothesis has not yet been tested. We tested this hypothesis using a vegetation data set collected from 14 river basin systems in Hyogo Prefecture, Japan. We compared vegetation-patch diversity between confluence and single-flow areas using hierarchical Bayesian models. Our results demonstrated greater vegetation-patch diversity in confluence areas compared to single-flow areas. Our findings support the hypothesis that confluences result in highly heterogeneous habitats. To the best of our knowledge, this is the first empirical report to demonstrate that river confluences have high vegetation-patch diversity. We conclude that network nodes play an important role in maintaining the biodiversity of river networks.

---

• Osawa T, Mitsuhashi H and Ushimaru A. (2010) Plant Ecology 209: 95-108.
In riparian zones along the banks of streams and rivers, flooding often causes large changes in environmental conditions immediately downstream of confluences. In turn, spatial heterogeneity in flooding along rivers and streams likely affects local species diversity. Furthermore, flooding during the plant growing season can strongly affect plant survival. In this study, we hypothesized that confluences have impacts on plant species diversity, and that these impacts are larger during the plant growing season. To test this hypothesis, we measured plant species diversity and the extent of natural bare ground at 11 river confluences during two different seasons (summer and spring) within the Mukogawa River basin system, Japan. Species diversity was highest at down-confluence areas in the summer. We linked the pattern of species diversity to that of bare ground creation by floods around the confluences and to the seasonality of annual plant recruitment. The extent of bare ground was significantly greater at down-confluence areas than at up-confluence areas. The recruitment of annual species was higher in the summer than in the spring and included rapid occupancy of bare ground in the summer. We suggest that within river systems, spatial and seasonal differences in patterns of flooding function together to regulate plant species diversity.

---

• Uematsu Y, Koga T, Mitsuhashi H and Ushimaru A. (2010) Agriculture, Ecosystems & Environment135:304-309.
Although traditional agricultural land use maintains biodiversity, recent land-use changes involving abandonment or use intensification have rapidly reduced the biodiversity of agricultural landscapes. Organisms living in agricultural landscapes are likely to respond differently to these changes, with some species declining rapidly and others remaining unchanged. However, few studies have focused on this interspecific difference in susceptibility to land abandonment and intensification in agricultural landscapes. We hypothesize that rarer herb species are more susceptible to both abandonment and intensification than are common herbs due to habitat preferences in the semi-natural grasslands of agricultural landscapes. To test this hypothesis, we examined the distributions of two pairs of congeneric grasslands species on abandoned and consolidated (production-intensified) paddy fields to assess differences in vulnerability to paddy abandonment and consolidation between the rarer and the more common species in an agricultural landscape. We found that higher, steeper fields farther from roads in the upper areas of paddy terraces were more frequently abandoned in our study area. The two rarer species had significantly more overlap with the distribution of fields at risk of abandonment than did the two more common congeneric species. In addition, the two rarer species were significantly less widely distributed in consolidated fields. Thus, both land abandonment and intensification appear to asymmetrically decrease habitats of rarer species. In light of our findings, we also discuss biodiversity conservation in agricultural landscapes with changing land use.

---

• Ushimaru A, Dohzono I, Takami Y and Hyodo F. (2009) Oecologia160:667-674.
Zygomorphic flowers are usually more complex than actinomorphic flowers and are more likely to be visited by specialized pollinators. Complex zygomorphic flowers tend to be oriented horizontally. It is hypothesized that a horizontal flower orientation ensures effective pollen transfer by facilitating pollinator recognition (the recognition-facilitation hypothesis) and/or pollinator landing (the landing-control hypothesis). To examine these two hypotheses, we altered the angle of Commelina communis flowers and examined the efficiency of pollen transfer, as well as the behavior of their visitors. We exposed unmanipulated (horizontal-), upward-, and downward-facing flowers to syrphid flies (mostly Episyrphus balteatus), which are natural visitors to C. communis. The frequency of pollinator approaches and landings, as well as the amount of pollen deposited by E. balteatus decreased for the downward-facing flowers, supporting both hypotheses. The upward-facing flowers received the same numbers of approaches and landings as the unmanipulated flowers, but experienced more illegitimate landings. In addition, the visitors failed to touch the stigmas or anthers on the upward-facing flowers, leading to reduced pollen export and receipt and supporting the landing-control hypothesis. Collectively, our data suggested that the horizontal orientation of zygomorphic flowers enhances pollen transfer by both facilitating pollinator recognition and controlling pollinator landing position. These findings suggest that zygomorphic flowers which deviate from a horizontal orientation may have lower fitness because of decreased pollen transfer.

---

• Ushimaru A, Ishida C, Sakai S, Shibata M, Tanaka H, Niiyama K and Nakashizuka T. (2008) Journal of Apicultural Research 47(4) : 296- 303.
We examined the relationship between flower preference and the spatial distribution of two Bombus species in a traditional agro-forestry Satoyama landscape where human activities maintain a mosaic of landscape units, including old-growth and secondary forests, conifer plantations, cutover land, and paddy fields. Bombus ardens prefers shallow woody flowers, whereas B. diversus tends to forage on herb flowers with long corolla tubes. Bumble bees were collected by window traps at seven types of landscape units. Most B. ardens individuals (77%) were caught at upper forest strata of old-growth stands. B. diversus queens displayed a similar preference for old-growth stands, whereas approximately half of workers were captured in open lands, such as cut-over lands and paddy fields. These findings suggest that B. ardens mainly uses old-growth forests, where woody species diversity and woody flower abundance is high, whereas B. diversus visits forest herbs and shrubs until early summer and then spreads into open lands to seek autumn-blooming herbs.

---

• Ushimaru A, Kikuchi S, Yonekura R, MaruyamaA, Yanagisawa N, Kagami M, Nakagawa M, Suzuki M, Kohmatsu Y, Hatada A, Kitamura S and Nakata K. (2007) Nordic Journal of Botany 24: 593-598.
We compared the amount of variation in flower size between autogamous and insect-pollinated species to examine the hypothesis that pollinator-mediated selection stabilizes flower size in plant populations. One would expect the flower size variation to be larger in selfing species that are less affected by pollinator-mediated stabilizing selection than in insect-pollinated species. The results of phylogenetic comparisons between autogamous and insect-pollinated flowers supported the pollinator-mediated stabilizing selection hypothesis, although the non-phylogenetic comparison did not. According to our results, we discuss the factors influencing the flower size variation.

---

• Ushimaru A, Watanabe T and Nakata K. (2007) American Journal of Botany 94: 249-258.
Visual floral guides such as colored anthers, lines, dots, and UV-absorption patterns on petals are commonly observed in insect-pollinated angiosperms. Floral guides that are known to enhance foraging efficiency of visitors on flowers thus promote return visits (foraging facilitation hypothesis, which predicts that visitors will discriminate against flowers with inferior floral guides). In this study, we experimentally examined the hypothesis that floral guides also prevent pollen-theft behavior by floral visitors (theft prevention hypothesis), which has rarely been tested. Nectarless flowers of Commelina communis have three types of brightly colored floral organs: large blue petals, rewarding yellow anthers, and nonrewarding yellow anthers. Colored floral organs were removed artificially from plants in two natural populations of C. communis. Removal of the nonrewarding yellow anthers diminished hoverings in front of flowers and tended to reduce the number of total floral visitor (syrphid fly) landings, supporting the foraging facilitation hypothesis. Additionally, removal of the rewarding yellow anther decreased the frequency of legitimate landings on flowers and the legitimate landing-to-total landing ratio, which is consistent with the theft prevention hypothesis. The nonrewarding anthers and the rewarding yellow anthers were shown to play an important role in increasing visitor landings and orienting floral visitors toward a landing point appropriate for pollination, respectively. We also showed that the absence of yellow anthers decreased both pollen dispatch from brown anthers and receipt by stigmas in C. communis. These findings support both the foraging facilitation hypothesis and the theft prevention hypothesis.

---

• Ushimaru A, Kawase D and Imamura A (2006) Functional Ecology 20:585-591.
1. An animal-pollinated plant living on a slope should orientate its flowers down-slope towards the more open space if by doing so it receives more pollinator visits and thereby achieves increased reproductive success. 2. We measured flower orientation relative to slope direction on individuals of ten species of forest-floor herbs in cool temperate forests in Japan. For one of these species, Erythronium japonicum, we also manipulated flower orientation to test experimentally for effects on both male and female reproductive function. 3. In all ten species, flowers were preferentially orientated down-slope. This pattern was more pronounced in plants growing on steeper slopes. 4. Our manipulative field experiment in Erythronium japonicum demonstrated that pollen dispatch was highest in flowers orientated down-slope . Additionally, flowers orientated up-slope may have achieved a lower seed set on steep slopes. 5. We conclude that down-slope orientation of flowers was a general phenomenon among the species that we studied, and that this behaviour was adaptive in enhancing plant fitness through pollination.

---

• Ushimaru A and Hyodo F (2005) Evolutionary Ecology Research 7:151-160.
Flower orientation is an important character influencing plant fitness. Zygomorphic flowers are known to orient vertically. We conducted field experiments in which we changed the flower angle of zygomorphic Commelina communis in order to examine how flower orientation affects pollinator behavior. We confirmed that Commelina flowers oriented vertically like other zygomorphic flowers. Then, we artificially prepared control, upward-, and downward-oriented flowers and exposed them to natural pollinators (syrphid flies and bumblebees). We found that the frequency of approach by syrphid flies and bumblebees was not influenced by flower angle, but there were fewer landings on downward-oriented flowers than on control and upward-oriented flowers. Moreover, the upward flower orientation increased illegitimate landings (landing on the flower without touching the stigmas or mating-related anthers) compared with controls. Thus, vertical flower orientation in zygomorphic flowers serves to control pollinator landings. These findings suggest that deviations from vertical orientation may reduce fitness in C. communis by reducing the efficiency of insect-mediated pollen transfer.

---

Top page