We examined water from 182 non-circulating hot spring bathing facilities in Japan for possible Legionella occurrence from June 2005 to December 2006, finding Legionella-positive cultures in 119 (29.5%) of 403 samples. Legionellae occurrence was most prevalent in bathtub water (39.4%), followed by storage tank water (23.8%), water from faucets at the bathtub edge (22.3%), and source-spring water (8.3%), indicating no statistically significant difference, in the number of legionellae, having an overall mean of 66 CFU/100mL. The maximum number of legionellae in water increased as water was sampled downstream:180 CFU/100 mL from source spring, 670 from storage tanks, 4,000 from inlet faucets, and 6,800 from bathtubs. The majority--85.7%--of isolated species were identified as L. pneumophila : L. pneumophila serogroup (SG) 1 in 22%, SG 5 in 21%, and SG 6 in 22% of positive samples. Multivariate logistic regression models used to determine the characteristics of facilities and sanitary management associated with Legionella contamination indicated that legionellae was prevalent in bathtub water under conditions where it was isolated from inlet faucet/pouring gate water (odds ratio [OR] = 6.98, 95% confidence interval [CI] = 2.14 to 22.8). Risk of occurrence was also high when the bathtub volume exceeded 5 m3 (OR = 2.74, 95% CI = 1.28 to 5.89). Legionellae occurrence was significantly reduced when the bathing water pH was lower than 6.0 (OR = 0.12, 95% CI = 0.02 to 0.63). Similarly, occurrence was rare in inlet faucet water or the upper part of the plumbing system for which pH was lower than 6.0 (OR = 0.06, 95% CI = 0.01 to 0.48), and when the water temperature was maintained at 55 degrees C or more (OR = 0.10, 95% CI = 0.01 to 0.77). We also examined the occurrence of amoeba, Mycobacterium spp., Escherichia coli, Pseudomonas aeruginosa, and Staphylococcus aureus in water samples.