Atrazine (2-chloro-4-ethylamino-6-isopropylamino-s-triazine) and N-deethylated atrazine (2-chloro-4-amino-6-isopropylamino-s-triazine) were monitored (1974 and 1975) in five rivers which drained agricultural areas in the Yamaska river basin of Québec. Water samples were collected frequently from April to December each year from sites near the outflow of each river. The water samples were extracted with benzene, chloroform, or ethyl acetate and the extracts were analyzed by gas chromatography using a Hall electrolytic conductivity detector and an alkali-flame detector. Atrazine and N-deethylated atrazine residues ranged in concentration from 0.01 to 26.6 microgram/L and less than 0.01 to 1.34 microgram/L, respectively, over the monitoring period. The highest levels of atrazine were observed in July each year and they coincided with the herbicide spraying season in this region and with occasional heavy rainfall events. Discharges (kg/yr) of atrazine from the five rivers were related to corn-growing area in each watershed. Losses of atrazine ranged from 0.1 to 2.9% of the atrazine that was estimated to have been applied in each watershed.

Methoxychlor was applied to 125-m3 limnocorrals at nominal concentrations of 3 and 300 pg L-' in 1981 and of 5 and 50 pg L-' in 1982. The half-life of methoxychlor in water ranged from 6 to 13 d; however, methoxychlor was strongly adsorbed to and slowly released from the polyethylene liners and the sediments on the bottom of the enclosures. Forced mixing of methoxychlor in the water resulted in greater adsorption to the sediments than did surface application. Methoxychlor had little effect on water chemistry parameters, although nitrite and dissolved inorganic carbon concentrations were higher in the treated than in the control corrals in each year of the study.

Permethrin (3-phenoxybenzyl(1RS)-cis,trans-3-(2,2-dimethy[-3-dichlorovinyl)-2,2-dimethylcyciopropanecarboxylate) applied to approximately 100-m3 enclosures (limnocorrals) in a small mesotrophic lake in Southern Ontario (47°51′25″N; 77°25′30″W) at concentrations of 500, 50, 5, and 0.5 μ∙L−1 dissipated from the water rapidly and approximated first-order kinetics in the first 8–12 d. Time taken for 50 and 90% dissipation ranged from 1.65 and 3.65 d, respectively, at 0.5 μ∙L−1 to 3.5 and 6.75 d, respectively, at 50 μ∙L−1. Inter- and intra-seasonal replication of dissipation patterns was good. Rate of dissipation varied slightly with depth, normally being slower at greater depth. Absorption of permethrin to sediments was rapid, penetration shallow, and disappearance slow. Permethrin had no effect on water chemistry but there was an increase in the Secchi disk depth in the treated limnocorrals. Dissolved inorganic carbon decreased in all limnocorrals, including controls after treatment, suggesting precipitation of calcium carbonate which may act as a scavenging agent for permethrin in the water. Limnocorrals are a useful tool for evaluating the behavior of pesticides in the aquatic system.

Methoxychlor was applied to 125‐m3 limnocorrals at nominal concentrations of 3 and 300 μg L−1 in 1981 and of 5 and 50 μg L−1 in 1982. The half‐life of methoxychlor in water ranged from 6 to 13 d; however, methoxychlor was strongly adsorbed to and slowly released from the polyethylene liners and the sediments on the bottom of the enclosures. Forced mixing of methoxychlor in the water resulted in greater adsorption to the sediments than did surface application. Methoxychlor had little effect on water chemistry parameters, although nitrite and dissolved inorganic carbon concentrations were higher in the treated than in the control corrals in each year of the study.