Search

We investigated the effects of the presence of bladderwort plants on survival of early instar larvae of one coenagrionid and two libellulids in laboratory experiments. In all three species survival...

Recent studies suggest insect declines in parts of Europe; however, the generality of these trends across different taxa and regions remains unclear. Standardized data are not available to assess large-scale, long-term changes for most insect groups but opportunistic citizen science data is widespread for some taxa. We compiled over 1 million occurrence records of Odonata (dragonflies and damselflies) from different regional databases across Germany. We used occupancy-detection models to estimate annual distributional changes between 1980 and 2016 for each species. We related species attributes to changes in the species’ distributions and inferred possible drivers of change. Species showing increases were generally warm-adapted species and/or running water species while species showing decreases were cold-adapted species using standing water habitats such as bogs. We developed a novel approach using time-series clustering to identify groups of species with similar patterns of temporal change. Using this method, we defined five typical patterns of change for Odonata – each associated with a specific combination of species attributes. Overall, trends in Odonata provide mixed news – improved water quality, coupled with positive impacts of climate change, could explain the positive trend status of many species. At the same time, declining species point to conservation challenges associated with habitat loss and degradation. Our study demonstrates the great value of citizen science data for assessing large-scale distributional change and conservation decision-making.

AIM: Land use change, such as urbanization and intensification of agricultural practices, poses major threats for biodiversity. We examined whether the composition of freshwater species differs between landscapes dominated by urban, agricultural or more natural structures in a region with a long history of landscape transformation. We determined the differences and characteristics of the dragonfly fauna in relation to major land use categories. We particularly focused on urban land use, as it generally induces a high level of landscape transformation. LOCATION: Germany. METHODS: We used distribution data of dragonflies derived from a citizen science database, which was compiled for the German distribution atlas. To analyse whether different land use categories are associated with distinct species compositions, we performed the classification method random forest. RESULTS: Based on dragonfly abundance per grid cell, we identified distinct land use‐related assemblages that were separated by two gradients, that is, the level of transformation and a gradient from open land to forest. In particular, urban landscapes possessed a distinct species composition with specific species, all of which were opportunistic and supported by warm climate. They also comprised significantly higher α‐diversity than intensive agricultural landscapes. Similar values of γ‐diversity over all land use categories implied a countrywide homogenized dragonfly fauna, which is probably caused by historic land use. MAIN CONCLUSIONS: Our results indicate that land use is a relevant driver for the composition of freshwater species at the landscape scale. Urban landscapes maintain species diversity better than agricultural landscapes, but they modify the species composition. Additional research on the effects of land use change is required to understand species responses and predict future distributions in a changing world in order to plan sustainable conservation strategies.

We report cases of long-distance dispersal in Odonata, some of which were directly observed by identifying single individuals of riverine species in unsuitable habitat, mostly desert, far distant from reproduction habitats. The shortest possible linear distances of the observation points to reproduction habitats were measured. Furthermore, established populations of riverine species were recorded in artificial lakes in central and southern Namibia far distant from the next regular reproduction sites. Our records demonstrate that single individuals of riverine species were probably covering distances of several hundred kilometres over arid landscape without any intervening possible reproduction habitat. Although it is likely that only small numbers of individuals of the river populations may disperse long distances, relatively recent colonizations of artificial habitats suggest that a few, or even single, dispersing individuals may lead to large-scale-range expansions.

African freshwater ecosystems are increasingly being impacted by humans, requiring an effective tool to assess these impacts for future conservation action. Such a tool, the Dragonfly Biotic Index (DBI), was earlier developed to assess the quality of South Africa’s freshwater ecosystems and is based on combining the scores of three sub-indices (geographical distribution, threat status, and habitat sensitivity) for each South African dragonfly species. The sum of the DBI scores for all the species recorded at assessed sites indicates the relative quality of these sites. The International Union for the Conservation of Nature/Species Survival Commission (IUCN/SSC) has assessed the threat status of certain aquatic taxa in Africa, including dragonflies. These assessments, coupled with the latest information on the geographical distribution of each species, makes it possible here to geographically expand the South African DBI into a continental-scale assessment index (the African Dragonfly Biotic Index (ADBI)) by adapting the South African DBI sub-indices. We develop this continental index here. However, there are challenges when undertaking an assessment at the continental scale compared to a national scale. In particular, the habitat sensitivity sub-index of the South African DBI is a relative, quantitative measure based on numbers of individual dragonflies recorded from natural versus human-modified or artificial freshwater systems. While the data for the two sub-indices, species’ geographical distribution and Red List threat statuses, are available across the continent, this is not the case for the habitat sensitivity sub-index at this large spatial scale. This meant that an alternative sub-index measure was required. We overcame this challenge by exploring an alternative sub-index, i.e. the ‘species vulnerability sub-index’, based on knowledge of the vulnerabilities of the species to certain types of landscape transformation. Then, the species vulnerability sub-index scores were calculated and combined with the geographical distribution and Red List threat status sub-index scores to develop ADBI scores for a core of 604 dragonfly species with adequate data across the African continent. These ADBI scores provide a workable framework and baseline for determining freshwater quality, both lotic and lentic, relative to human disturbance at a continental spatial scale. The ADBI enables the monitoring of quality changes, for better or worse, over the continent in years to come. Overall, the ADBI also has the potential to help identify threats to, and sensitivities of, African freshwater ecosystems, leading to conservation action.

This study investigates whether cities have the potential of hosting high species diversity of dragonflies (Odonata), a target group in freshwater conservation. We reviewed the dragonfly fauna of 30 cities in Central Europe and analysed their species richness compared to the regional species pools in the hinterlands, i.e. estimated the amount of regional diversity represented in cities. In particular, we examined the occurrence of species of conservation concern at the European scale. Results revealed that 92.6% of all 81 Central European dragonfly species occurred in cities, as well as 85.7% of 14 species of conservation concern. As expected, assemblages of city species were subsets of the regional species pool and city species richness increased with regional species numbers. Some cities hosted the complete regional species pool. Ten species of conservation concern established autochthonous populations and six of them, such as Aeshna viridis and Ophiogomphus cecilia, were abundant at least in single cities. We conclude that there is good potential for cities to host high dragonfly diversity and even to promote species of conservation concern. To exploit this potential we recommend city planners to focus on the needs of regionally characteristic species.

Summary 1. Increasing temperature and invading species may interact in their effects on communities. In this study, we investigated how rising temperatures alter larval interactions between a naturally range-expanding dragonfly, Crocothemis erythraea, and a native northern European species, Leucorrhinia dubia. Initial studies revealed that C. erythraea grow up to 3.5 times faster than L. dubia at temperatures above 16 °C. As a result, we hypothesised that divergent temperature responses would lead to rapid size differences between coexisting larvae and, consequently, to asymmetric intraguild predation at higher ambient temperatures. 2. Mortality and growth rates were measured in interaction treatments (with both species present) and non-interaction controls (one species present) at four different temperature regimes: at an ambient temperature representative of central Germany, where both species overlap in distribution, and at temperatures increased by 2, 4 and 6 °C. 3. The mortality of C. erythraea did not differ between treatment and control. In contrast, mortality of L. dubia remained similar over all temperatures in the controls, but increased with temperature in the presence of the other species and was significantly higher there than in the controls. We concluded that L. dubia suffered asymmetric intraguild predation, particularly at increased temperature. Reduced growth rate of L. dubia in the interaction treatment at higher temperatures also suggested asymmetric competition for prey in the first phase of the experiment. 4. The results imply that the range expansion of C. erythraea may cause reduction in population size of syntopic L. dubia when temperature rises by more than 2 °C. The consequences for future range patterns, as well as other factors that may influence the interaction in nature, are discussed.

3 This is the first continent-wide overview of insect diversity and status sufficiently fine-scaled to be used in con- servation planning. We analyze patterns of richness and the conservation status of African dragonflies and dam- selflies (Insecta: Odonata), commonly referred to as dragonflies, to determine threats to species and freshwater habitats, location of diversity hotspots, necessary conservation actions, and research gaps. Major centers of drag- onfly diversity in Africa are tropical forest areas that include highlands. Most threatened species - as classified by the International Union for Conservation of Nature global Red List - are concentrated in highlands from Kenya to South Africa (together with the Cape Floristic Region), western Africa (including mountains on the Cameroon-Nigeria border), and Ethiopia. Currently available knowledge can be applied throughout Africa's freshwater systems to help minimize or mitigate the impact of future development actions, allowing dragonflies to act as "guardians of the watershed". The private sector can be advised to safeguard sensitive habitats and species when selecting sites for development. Key sites and species for monitoring can be identified by checking the distribution of threatened species at www.iucnredlist.org.

A new species of Crenigomphus is described and illustrated from a type series of eight males and eight females, all collected along the Okavango River in Namibia during December 2004, three non-type adult specimens and several exuviae (holotype ♂ Namibia, N'Kwazi Lodge, 19 xii 2004, deposited at NMNW). Both sexes lack foliations at S8-9 as occur in some Crenigomphus, but the male is peculiar in having exceptionally long cerci. The latter character is normally present in the genus Paragomphus. Other characters typical of Crenigomphus include all wings having a bright yellow costal border, S10 longer than S9 in males, colouration mostly ochreous with few darker markings, and the strong blackish serration at the posterior end of the cerci. The larval characters based on exuviae, one associated with an emerged male, do not allow clear separation from Paragomphus.

1. Freshwater communities are often structured by predation. In permanent lentic freshwater habitats dragonfly larvae are major predators which, in return, suffer predation by fish. Antipredator traits vary between the dragonfly species, and the dragonfly communities are therefore shaped by the presence of fish. But fish communities vary, and as different fish species affect dragonflies in different ways, the species composition of the fish community may affect the composition of the dragonfly community. 2. We sampled dragonfly larvae in 24 lakes with a known fish stock in south-western Sweden, and explored the impact of fish as well as vegetation structure on dragonfly communities by means of multivariate analyses. 3. We found that the presence of four fish species affected the community structure of dragonflies. The impact strength depended mainly on the abundance of Perca fluviatilis, with which most dragonfly species were negatively correlated. Many dragonfly species were also positively correlated with the occurrence of at least one fish species, which may reflect similar habitat requirements or imply indirect positive effects of these fish species. 4. Of the 24 recorded dragonfly species, four did not occur in lakes dominated by P. fluviatilis, whereas only one species was lacking in lakes dominated by Rutilus rutilus. The dragonfly species diversity was higher in R. rutilus lakes than in P. fluviatilis lakes. 5. Our results suggest that the fish species composition is a major determinant of the dragonfly community, which in turn will influence the lower trophic levels.

1. For modelling the future ecological responses to climate change, data on individual species and on variation within and between populations from different latitudes are required. 2. We examined ...

1. Morphological defence structures evolve against predators but are costly to the individual, and are induced only when required. A well-studied example is the development of longer abdominal spin ...

With a total of 75 species the odonate diversity in the Namibian desert is surprisingly high. Based on their distribution characteristics, invasion patterns, and breeding success, there are six well-defined categories of Odonata: widespread species - (1) permanently living in the desert, and desert biased, (2) permanently living in the desert, but not desert-biased; (3) entering the desert seasonally; (4) entering from neighbouring tropical or temperate regions, whose populations may breed in the desert sometimes or locally. Category (5) consists of species with highly localised breeding populations in the desert, which are widely isolated from potential source populations. The last category (6) consists of species restricted to allochthonous perennial rivers. We discuss these patterns from a geographical and a temporal perspective. On the one hand, there have been different spatial directions from where species have entered deserts. On the other hand, Odonate distribution patterns in the deserts...

It has been shown that life history, behavioural as well as morphological traits vary with the habitats occupied by odonate larvae. Here we ask the following questions: (1) Are the morphological tr ...

Summary 1. Precise models for the phenology of different species are essential for predicting the potential effects of any temporal mismatch of life cycles with environmental parameters under different climate change scenarios. Here we investigated the effects of ambient water temperature on the onset and synchrony of emergence for a widespread European riverine dragonfly, Gomphus vulgatissimus. 2. Long-term field data on the annual emergence from two rivers in northern Germany, and additional data from a laboratory experiment with different temperature regimes, were used to develop a model that predicted the onset of emergence by using mainly the temperature sum (degree days) as a parameter. 3. Model predictions of the onset of emergence fitted the observations well and could be transferred between localities. This was particularly so when weighting early winter temperature data by using a day length and a temperature-response function, implying potential additional control mechanisms for the onset of emergence. 4. We simulated effects of different winter temperature regimes on the emergence curves in order to predict the effects of climate change. These indicated an acceleration of emergence by 6–7 days per 1 °C temperature increase, which is corroborated by the laboratory data and is in the upper range of data published for other dragonflies.

Potential effects of future warming should be reflected in life history patterns of aquatic organisms observed in warmer climates or in habitats that are different in ambient temperature. In the special case of the dragonfly Gomphus vulgatissimus (L.) (Odonata: Gomphidae) previous research suggests that voltinism decreases from south to north. We analysed data on voltinism from 11 sample sites along a latitudinal gradient from about 44°N to 53°N, comprising small streams to medium-sized rivers. Furthermore, to simulate different conditions and to allow projections for future climate change scenarios, we developed a population dynamic model based on a projection matrix approach. The parameters of the model are dependent on temperature and day length. Our field results indicate a decrease in voltinism along the latitudinal gradient from southern to northern Europe and a corresponding increase of voltinism with higher temperatures. An increase in voltinism with width of the running water implies an effect of varying habitat temperature. Under the impact of global warming, our model predicts an increased development speed, particularly in the northern part of the latitudinal gradient, an extension of the northern range limit and changes in phenology of G. vulgatissimus, leading to an extension of the flight season in certain regions along the gradient.

We classified 542 records of voltinism for 275 species and subspecies of Odonata according to three variables: geographical latitude, systematic position and habitat type. We sorted records according to voltinism—categories being three or more generations per year, two generations per year, one generation per year, one generation in two years and one generation in three or more years. We sought to correlate the voltinism of each record with latitude of the study site, thus demonstrating an overall negative correlation between voltinism and latitude. After allowing for phylogenetic similarity a negative correlation remains, although it decreases in strength after removal of taxonomic correlates, mainly between family and genus levels. A negative correlation exists at the species level within most families, with the exception of Lestidae. In genera for which we lacked data for latitudes 0-31°N/S no significant correlation between latitude and voltinism exists. In temporary waters most species compl...

Descriptions and illustrations of the final stadium larvae of Gomphidia bredoi, G. gamblesi and G. quarrei are presented, based on exuviae collected in Ivory Coast, Cameroon, Kenya and Namibia. The three species can be separated by the presence/ absence and shape of an abdominal dorsal spine on segment 9, by the numbers of abdominal lateral spines and by several characteristic processes on the head.

Information on larval biology of Macromia splendens was compiled and supplemented by hitherto unpublished data. Larvae inhabit mainly calm river stretches, sometimes artificial impoundments, and lentic margins of lotic sections. From the majority of records it is concluded that the larvae mainly dwell in sandy substrates in shallow water, which sometimes contains little leaf litter. Larvae occur in smaller numbers on substrates dominated by coarse detritus or on bedrock in deeper water. Larvae of M. splendens are able to burrow in sand, but such burrowing takes a long time, and sometimes parts of their body remain uncovered. Therefore, they are considered shallow burrowers. Based on head-width frequency distributions recorded at the Gardon de Mialet, southern France, the species is believed to require two years per generation.

Identifying and examining traits that influence the distribution of species is crucial to the understanding of community structure. Theory predicts that traits should differ between species that live in temporary and permanent waters because of differing major environmental variables; viz drying out and predator presence, respectively. Species, however, will also be influenced by their evolutionary history, i.e. by the traits of their common ancestors. We studied differences in life history and behaviour traits in a series of laboratory experiments using pairs of dragonfly species out of three genera of Namibian Libellulidae (Odonata) with one species from each type of habitat. As predicted, growth rates were significantly higher in the temporary water species compared to the permanent water species. Activity and foraging, in contrast, differed between the genera, but did not differ between the habitat types. Hence, our study implies that the behavioural traits are influenced by phylogenetic inertia rather than by the habitat variables, while growth rate is adapted to the habitat. We argue that in all three genera one species has diverged recently from a sister species that lives in the original habitat of the genus, which may be temporary waters in Crocothemis Brauer and in Orthetrum Newman, and permanent waters in Trithemis Brauer. The behavioural traits may therefore be less well adapted. Rapid growth may be the more relevant trait because it is crucial to survival in temporary waters.

We studied how egg size, larval size and egg development time are related to oviposition site selection and mate guarding in a dragonfly assemblage of the Namib desert. Species that oviposited mainly in tandem flight (Pantala flavescens and Sympetrum fonscolombii) spread their eggs over several ponds, which we interpreted as spatial risk spreading. Other species (Orthetrum chrysostigma, Trithemis kirbyi, T. annulata and Crocothemis erythraea) performed noncontact guarding and localized their eggs, usually in a single pond, which we interpreted as distinct habitat selection. Because long oviposition durations increase the risk of disruption by rival males, we predicted that in species that localize their eggs to a distinct oviposition habitat the first-laid eggs of a clutch should be the largest, indicating high quality. Species that perform risk spreading should distribute large eggs randomly over all oviposition sites. We tested our hypothesis at artificial ponds using experimental manipulations of oviposition. In O. chrysostigma and T. kirbyi, egg size and larval size decreased significantly with order of laying, and the time to hatch was randomly distributed. As predicted, within the clutches of P. flavescens and S. fonscolombii, egg size was randomly distributed over a clutch and we found no trend of decrease in larval size or increase in the time to hatch in relation to order of laying.

1. Freshwaters form a gradient from small temporary waters to large permanent waters. Identifying and examining traits that restrict the distribution of species along this gradient are crucial to the understanding of community structure in these habitats. 2. Using dragonfly larvae species, differences in traits important for growth and survival were studied. 3. The traits were studied in a series of laboratory experiments using two pairs of dragonfly species that coexist in the Namibian semi-desert. One species pair was from the most temporary part of the water permanence gradient and the other species pair from an intermediate part of the gradient. 4. As predicted, activity, capture rate, and growth rate were significantly greater in the two temporary water species. Contrary to the prediction made in the work reported here, species differences in microhabitat selection were not related to the species' habitat origin. Cannibalism did not differ between species. 5. The results lend support to the hypothesis that selection has favoured certain combinations of trait values and that these traits are important for a successful life in temporary and permanent waters.

In this article we report underwater oviposition and epilithic oviposition in Anisoptera. We observed Notoaeschna sagittata totally submerged ovipositing on bare rock in the fast current of a rapid. Dendroaeschna conspersa oviposited also underwater, but into wood submerged in very clear water.

A written description and illustrations of the final stadium larva of Nesciothemis farinosa are presented, based on larvae and exuviae collected in Namibia. The larvae were reared in the laboratory until emergence. Additionally, information on larval microhabitat, behaviour and development is provided.

A negative relationship between clutch size and egg size is generally expected. However, no such trade off has been reported in Odonata. In this paper we analyse relationships between egg size and clutch size in the dragonfly subfamily of Sympetrinae using material from Norway, Sweden, Germany and France. Clutch size varied within and among the species, and only maximum clutch size was comparable between species. Both egg length and width varied among species. Moreover, mean egg length differed significantly intraspecifically among individual clutches of all species. We found mean egg length to be negatively correlated to clutch size, confirming the trade off between egg size and clutch size across species. Clutch size was positively correlated to female size, but egg size was not.

Libellulid dragonflies lay large egg clutches. When eggs of two or more species are deposited at the same time and patch, one can expect a high density of early-instar larvae. Thus, interspecific interactions should be evident. We studied the interaction of two species, Sympetrum fonscolombii and Orthetrum cancellatum, which typically co-occur in French rice fields. We experimentally simulated the laying of egg clutches of both species at the same time and site. Survival of O. cancellatum was reduced compared with that of controls without S. fonscolombii and also with that of S. fonscolombii in the two-species treatment. At the end of the experiment, mean head width of S. fonscolombii was greater than that of O. cancellatum, which may be one reason for the differential survival. In a second experiment we observed the behaviour of pairs of early-instar larvae of both species in different size combinations. Small larvae of S. fonscolombii reduced locomotory activity in the presence of larger O. cancellatum. In contrast, small O. cancellatum did not do so in the presence of larger S. fonscolombii. This behavioural difference may also cause asymmetric interspecific predation.

The microdistribution patterns of larval Gomphus graslinii and Oxygastra curtisii in three rivers in southern France were studied. While G. graslinii was caught mainly in sandy reaches covered with leaf litter, O. curtisii was found almost exclusively clinging to alder roots at the river margins. Preferences of habitat structure were determined from habitat selection experiments under outdoor conditions with four types of substrates: (I) gravel, (II) sand, (III) sand covered with leaves, and (IV) alder roots. G. graslinii preferred substrate types III and II over I and IV; O. curtisii preferred IV and III over I and II.

I studied predation by larval Cordulegaster boltonii on larvae of two other dragonfly species in a laboratory artificial stream. Onychogomphus forcipatus unguiculatus suffered significantly higher predation than Onychogomphus uncatus. Activity was much lower in O. uncatus, which is interpreted as an antipredation trait against C. boltonii. The higher activity of O. f. unguiculatus makes it vulnerable to the sit-and-wait predator C. boltonii due to increased encounters. Rearing of both Onychogomphus in field cages under the same conditions of temperature, food availability and density revealed lower growth rates of O. uncatus, which may be interpreted as a cost of the antipredation behaviour. To test whether experimental results have relevance for largescale distribution patterns of the two Onychogomphus, samples were taken at 23 sites in river-systems of southern France where both species co-occur. In the field I found that O. f. unguiculatus does not coexist with C. boltonii at sites where the latter occurs in high numbers. In contrast, O. uncatus is established even at high density of C. boltonii.

The emergence of Onychogomphus uncatus from three stretches at two adjacent canals in Southern France was compared in 1993. Whereas 50% of annual emergence from the two stretches at one canal had occurred after 25 and 36 days, respectively, it had occurred after 5 days at the second, where emergence began 14 days earlier. At the second canal most of the larvae of one generation had overwintered in the final instar but at the first canal only half did so. These differences correlate with greater annual fluctuations of temperature at the second canal which are responsible for high temporal synchronisation. Other factors that may cause differences in seasonal regulation, such as drying up and density, are discussed. Differences exist also in sex ratio and abundance of emerging individuals, particularly between the two collection sites at the first canal.

Urbanisation is increasing and it is essential to integrate biodiversity into the spatial planning of urban areas. This requires deeper understanding of biodiversity patterns in cities. We investigated which habitat variables are major determinants of dragonfly diversity and species assemblage structure in the municipal area of Dortmund (Germany). We sampled dragonfly larvae in 33 ponds situated in city parks, commercial, residential and agricultural areas. We recorded 30 autochthonous dragonfly species with species richness ranging from zero to 17. Additionally, we surveyed a set of environmental variables including habitat size, water level, pond structures and vegetation as well as surrounding landscape and potential disturbances like waterfowl and fish. Multivariate methods were used to identify the major determinants of dragonfly diversity, abundance and assemblage structure. Analysis indicated that diversity of aquatic and terrestrial vegetation affected dragonfly diversity positively. City park ponds had low diversity, but Ischnura elegans was obviously promoted by the specific park pond conditions, including high waterfowl density. We found five assemblages mostly determined by generalistic species which were related to different pond types. Moderately disturbed ruderal and pioneer ponds in residential and agricultural areas also contained increased numbers of rare species. Our results indicate that urban ponds may have a great value for maintaining biodiversity, but various disturbances have negative impact. To promote urban biodiversity we suggest a natural design of well-vegetated ponds as well as a high diversity of different pond types and particularly a more-natural redesign of city park ponds.

Anisopteran leg functions change dramatically from the final larval stadium to the adult. Larvae use legs mainly for locomotion, walking, climbing, clinging, or burrowing. Adults use them for foraging and grasping mates, for perching, clinging to the vegetation, and for repelling rivals. In order to estimate the ontogenetic shift in the leg construction from the larva to the adult, this study quantitatively compared lengths of fore, mid, and hind legs and the relationships between three leg segments, femur, tibia, and tarsus, in larval and adult Anisoptera of the families Gomphidae, Aeshnidae, Cordulegastridae, Corduliidae, and Libellulidae, represented by two species each. We found that leg segment length ratio as well as ontogenetic shift in length ratios was different between families, but rather similar within the families. While little ontogenetic shift occurred in Aeshnidae, there were some modifications in Corduliidae and Libellulidae. The severest shift occurred in Gomphidae and Cordulegastridae, both having burrowing larvae. These two families form a cluster, which is in contrast to their taxonomic relationship within the Anisoptera. Cluster analysis implies that the function of larval legs is primarily responsible for grouping, whereas adult behavior or the taxonomic relationships do not explain the grouping. This result supports the previous hypothesis about the convergent functional shift of leg characters in the dragonfly ontogenesis.

1. Substratum selection by the burrowing larvae of the dragonfly Onychogomphus uncatus was examined in an artificial laboratory stream at different larval densities and in the presence of one of three other dragonfly species. 2. The larvae of O. uncatus, as well as those of the other species, clearly preferred gravelly sand substratum rather than gravel or stone. 3. At low larval density (71.4 m–2) in the stream, 83% of the O. uncatus were found in gravelly sand. An increase of larval abundance in the stream to 202.4 specimens m–2 resulted in greater density in all substrata, but this increase was proportionally lowest in gravelly sand. 4. The presence of a second species had various effects on the microdistribution of O. uncatus. In the presence of Gomphus simillimus or Cordulegaster boltoniiimmaculifrons the distribution of O. uncatus changed significantly; their density increased in the normally less preferred substrata. This effect is interpreted as asymmetric interspecific interference. The presence of a third species, Onychogomphus forcipatus unguiculatus, had no effects.

Four types of field experiments were carried out in order to study the effects of fish predation on different size groups of burrowing larval dragonflies Onychogomphus uncatus (CHARP.). (1) The effects of the bottom feeding fish Barbatula (Nemacheilus) barbatula (L.) on three size groups of O. uncatus in three types of substrate: sand, gravel and stones, were recorded using field enclosure cages. The densities of small sized larvae inhabiting cages with stones and gravel were reduced in the presence of fish compared with fish-free controls. No effects due to fish presence were found in larger larvae. (2) The colonisation of sand, gravel and stones in the field by different size groups of O. uncatus was studied using frames filled with substrates. Substrate as well as size specific effects were found. The low density of small larvae in coarse substrates is interpreted mainly as a direct effect of predation by B. barbatula which exclusively colonised these substrates. (3) To test the effect of exclusion of all fish on colonisation of O. uncatus, frames filled with stones were exposed in a running water and half of these were protected by cages. The densities of very small and medium sized larvae in the protected substrates were higher than in those without cages. (4) In cages with a wide mesh size, which allowed a free exchange of dragonfly larvae and B. barbatula but prevented predation by large fish, the density of small larvae of O. uncatus was reduced. B. barbatula were found exclusively inside the cages. It is suggested that predation by B. barbatula using the cages as a shelter against predation by large fish was responsible for this reduction.

Mortality during emergence in two species of dragonfly, Onychogomphus uncatus and Orthetrum coerulescens, was studied at a Mediterranean irrigation canal in France. Overall mortality was 5.2% (n =1901) and 5.7% (n =611), respectively. Predation was low in both species. The prime cause of mortality was incomplete ecdysis resulting from different factors among which heavy wind was the most important in O. coerulescens and lack of suitable emergence supports in O. uncatus. Besides data on mortality the emergence curves of both species are provided.