During the summer of 1983, a prospective study of swimmingrelated illness was conducted at six Southern Ontario freshwater beaches. Surface water samples collected were analyzed for specific bacteria and for viruses. Some beaches were closed for swimming due to high fecal coliform counts. The overall fecal coliform geometric mean (417;​, and fecal coliform geometric mean for the open (398) , and closed (447), beaches surpassed the Ontario guideline value of 100 fecal coliforms per 100 mL of water. Data were also acquired for 6653 swimmers, 574 waders, and 1193 (n=8420) people who did not enter the water. Crude morbidity data revealed a general trend of greatest morbidity in swimmers, followed by waders, and finally by persons who did not go into the water. For overall illness, the crude symptom rates were:​ 7 6.8 per 1000 swimmers, 41.8 per 1000 waders, and 19.3 per 1000 persons who did not enter the water. For all categories of water exposure, respiratory and gastrointestinal symptoms figured predominantly. Logistic regression modelling (n=8420 people) for overall illness generally revealed that swimmers were at significantly increased During the summer of 1983, a prospective study of swimming-related illness was conducted at six Southern Ontario freshwater beaches. Surface water samples collected were analyzed for specific bacteria and for viruses. Some beaches were closed for swimming due to high fecal coliform counts. The overall fecal coliform geometric mean (417;​, and fecal coliform geometric mean for the open (398) , and closed (447), beaches surpassed the Ontario guideline value of 100 fecal coliforms per 100 mL of water. Data were also acquired for 6653 swimmers, 574 waders, and 1193 (n=8420) people who did not enter the water. Crude morbidity data revealed a general trend of greatest morbidity in swimmers, followed by waders, and finally by persons who did not go into the water. For overall illness, the crude symptom rates were:​ 7 6.8 per 1000 swimmers, 41.8 per 1000 waders, and 19.3 per 1000 persons who did not enter the water. For all categories of water exposure, respiratory and gastrointestinal symptoms figured predominantly. Logistic regression modelling (n=8420 people) for overall illness generally revealed that swimmers were at significantly increased risk of becoming ill in contrast to persons who did not enter the water, but these models must be treated with caution. Logistic regression modelling for swimmers (n=6653) only, for various illness types, generally revealed that there was no evidence to suggest that bacterial count contributed to the prediction of illness in swimmers. Instead, age, contact person, and interviewer, most frequently tended to be important. Whether the bacterial count results are real is unknown, and future researchers must consider the possible effects of various potentially important factors. Logistic regression testing of the American and Ontario fecal coliform guidelines revealed that the latter was more effective in predicting gastrointestinal illness in swimmers. Presently, it is recommended that the Ontario fecal coliform guideline remain unchanged and that fecal coliforms (and Escherichia coli, if desired), total staphylococci, and Pseudomonas aeruginosa (on mPA) be employed as microbial recreational freshwater quality indicators. Future evaluation of a variety of bacterial and viral indicators and pathogens is recommended.