The effects of cypermethrin, a commonly used pyrethroid insecticide, were studied in small in situ enclosures situated in an eutrophic lake over an 11-day period. The experimental design used a regression principle that included three untreated controls and a gradient of six unreplicated cypermethrin concentrations, ranging from 0.01 to 6 microg/l. This paper is the second in a series of two and describes the effects on the species composition of the crustacean, rotifer, periphyton and phytoplankton communities. Multivariate ordination technique (redundancy analysis (RDA) combined with Monte Carlo permutation tests) showed that exposure to cypermethrin caused significant changes in the species composition of the communities. Changes in the structure of the communities were observed following exposure to a nominal concentration of 0.13 microg cypermethrin per litre above. The direct acute effect of exposure to cypermethrin was a rapid decrease of many species of crustacean zooplankton. The alterations in crustacean species composition were probably due to variations in susceptibility to the direct toxic effects of cypermethrin. No effects concentration (NEC) for individual zooplankton species were calculated using inverse regression and revealed that copepod nauplii were the most sensitive (NEC=0.01 microg/l) of the crustacean groups examined. The observed alterations of the species composition of the autotrophic communities as well as of the rotifers were most likely caused indirectly by cypermethrin, mediated through the direct negative effects of the insecticide on the crustacean grazers. The results of this experiment provide further knowledge about the direct and indirect effects of pesticide stress on the ecosystem level. They also show that there is a variation in sensitivity between different species of zooplankton under natural conditions and thus exemplify the necessity of multispecies approaches in the risk assessment of pesticides.