The filter-feeding clam Corbicula fluminea has widely spread from its Asiatic origin into freshwater habitats on several continents, where it often has a considerable impact on ecosystem processes. The present study was stimulated by the observation that C. fluminea can experience mass mortality during warm summers, even when temperatures are still far below the lethal level. We hypothesized that starvation due to low food quantities during summer is a main factor in this context. In order to test nutritional conditions in an environment where summer mortality occurred, the clams' body mass was tracked in river bypass systems installed at the Lower River Rhine (Germany and The Netherlands). Two food levels were adjusted in the bypass channels: one corresponding to the original chlorophyll level in the river (ambient food) and one with a chlorophyll a level reduced by about 50% (low food). The clams kept at the ambient food level increased their shell length during summer, although growth rates decreased at low food levels in the River Rhine in late summer. In contrast to shell length, body mass decrease in late summer cumulated in 94% reduction from August until October. This trend was enhanced by the experimental food reduction, i.e., clams kept in the low food level treatments weighed 60% less than the clams in the ambient food treatment at the end of summer. However, mortality was low in both treatments. The data demonstrate a high plasticity in the body mass of C. fluminea. The corresponding possibility to respire body mass can be seen as one strategy to survive longer starvation periods.

No abstract

1. The impact of burrowing larvae of Ephoron virgo (Ephemeroptera, Polymitarcidae) on sediment microbiology has not been previously investigated because of difficulties in sampling the sediment of large rivers under in situ conditions. Therefore, we conducted experiments in the on-ship Ecological Rhine Station of the University of Cologne (Germany), in which ambient conditions of the River Rhine can be closely mimicked. 2. In two consecutive seasons, experimental flow channels were stocked with Ephoron larvae and continuously supplied with water taken directly from the River Rhine. Sediment from the immediate vicinity of Ephoron burrows (i.e. U-shaped cavities reaching 10-80 mm deep into the sediment) and bulk sediment samples were analysed for (i) particulate organic matter content, (ii) microscale in situ distribution of O 2 , NO À 3 , and NH þ 4 , and (iii) potential activities of exoenzymes. 3. Sediment surrounding the Ephoron burrows had markedly higher organic matter contents and exoenzyme activities compared with the bulk sediment. Microsensor measurements demonstrated that local O 2 and NO À 3 penetration into the sediment were greatly enhanced by larval ventilation behaviour. Volumetric O 2 and NO À 3 turnover rates that were calculated from steady state concentration profiles measured directly in the burrow lining were considerably higher than at the sediment surface. 4. In the sediment of the fast flowing River Rhine Ephoron burrows are preferential sites of organic matter accumulation and dissolved oxidant penetration. Our data suggest that the burrows are surrounded by a highly active microbial community that responds to the inputs from the water column with elevated O 2 and NO À 3 turnover, and release of exoenzymes into the sediment pore water. Especially during periods of mass occurrence, the larvae of E. virgo may thus significantly contribute (i) to the ecological connection between the water column and the sediment and (ii) to biogeochemical processing of organic matter in the riverbed.

1. New methods for the observation of moulting-, pupation-, and emergence-rhythms under natural and experimental conditions are described. 2. The larvae feed during all hours of the day. Rhythmic changes in their behaviour were not observed. 3. The drift rate of medium and large larvae is very low, if the animals are not disturbed (0.05% per day,n=5,500). The drift rate of young larvae seems to be much higher, but could not be determined quantitativly. Drift rhythms must not be concerned with circadian rhythms in Simuliidae. 4. The larvae moult and pupate during all hours of the day. The number of moults and pupations reaches its maximum in the afternoon. 5. The duration of the pupal stage depends on temperature and the hour of pupation. In the stream Kaltisjokk it lasts about 6-7 days. 6. The seasonal emergence of 9 species is compared in two years. 7. All species (10) emerge during the daylight hours. The rhythm persists in the continous light of the subarctic summer and under artificial LL and DD. 8. Emergence is timed by changes in light and temperature. Within ecological limits light is the dominant Zeitgeber. 9. Under LL and DD the Simuliidae emerge two hours after a rise in temperature. 10. Under natural illumination and low temperatures the emergence is concentrated round noon. In relatively high temperatures the distribution becomes broader and sometimes biphasic. There were no specific differences between species and sexes. 11. Simuliidae fly mainly in the morning and evening to a CO-trap.Helodon ferrugineus was caught in light traps during the night.

Chironomus thummi emerged from cold water (9-12°C) during daylight, from warm water (16-25°C) mainly after dusk. At medium temperatures no intermediate peak occurred, but the pattern became biphasic. The midges switched from one eclosion time to the other without any transition. This happened in constant as well as in falling and rising temperatures. The biphasic pattern is explained by some intraspecific variability of the critical temperature. Midges bred in cold water changed the pattern at the same temperature level as those bred in warm water, indicating that there was no long-term adaptation. Both eclosion times were synchronized with the light-dark cycle (LD), and continued in constant light (LL) as well as in constant darkness (DD) with a free running circadian rhythm. A single step from LL to DD given to arhythmic populations resulted in rhythmic eclosion persisting for up to two weeks. The phase and shape of the peaks depended on the temperature level and resembled the two alternate patterns in LD.The duration of development in this polyvoltine species is scarcely affected by the photoperiod. Since eclosion is not restricted to a short suitable season, the variable pattern may be an adjustment to different seasonal temperature levels.