An experimental protocol using mesocosms was established to study the effect of Microcystis sp. cell abundance on microcystin production. The mesocosms (55 L) were set up in a shallow eutrophic lake and received either no (control), low (to simulate a moderate surface accumulation), or high (to simulate a dense surface scum) concentrations of Microcystis sp. cells collected from the lake water adjacent to the mesocosms. In the low- and high-cell addition mesocosms (2 replicates of each), the initial addition of Microcystis sp. cells doubled the starting cell abundance from 500 000 to 1 000 000 cells mL-1, but there was no detectable effect on microcystin quotas. Two further cell additions were made to the high-cell addition mesocosms after 60 and 120 min, increasing densities to 2 900 000 and 7 000 000 cells mL-1, respectively. Both additions resulted in marked increases in microcystin quotas from 0.1 pg cell-1 to 0.60 and 1.38 pg cell-1, respectively, over the 240 min period. Extracellular microcystins accounted for <12% of the total microcystin load throughout the whole experiment. The results of this study indicate a relationship between Microcystis cell abundance and/or mutually correlated environmental parameters and microcystin synthesis.