Chemicals used to control phytoplankton blooms induce the release of p
hytotoxins that increase the potential health risks in drinking water
supplies. To test this hypothesis, the effects of six chemical treatme
nts on the release of the cyanobacterial toxin, microcystin-LR (MCLR;
182-837 mu g g(-1) dry wt) from freshly collected phytoplankton were e
xamined in laboratory experiments. In addition, the integrity of a che
mically-treated culture of Microcystis aeruginosa was examined by both
a scanning electron microscope and a transmission electron microscope
. Chemicals which control cyanobacterial blooms through inhibition of
cell functions (e.g. Reglone A, potassium permanganate, chlorine, and
Simazine) appeared to induce cell lysis and subsequently increased dis
solved MCLR concentration in the surrounding water. In contrast, both
lime and alum treatment (within pH 6-10) controlled the cyanobacterial
blooms mainly by cell-coagulation and sedimentation, without any (lim
e) or only little (alum) increase in dissolved MCLR concentration in t
he water. The estimated half-life of released MCLR from these dense cy
anobacterial blooms ranged from 0.5 (+/- 0.1) to 1.6 (+/- 0.0) d. In c
ontrast, x greater than or equal to 39% of the MCLR remained in decayi
ng phytoplankton for up to 26 d, therefore it is likely that MCLR woul
d persist and decay inside the lime or alum coagulated Microcystis cel
ls, before being released into the surrounding water phase. For these
reasons, lime or to a lesser extent alum, appears to be more suitable
than either algicides or chlorine for the control of microcystin-conta
ining cyanobacterial blooms in drinking water.