Between 1 January and 31 July 1992 a cholera epidemic caused 548 reported cases (an incidence of about 8 cases per 1,000 inhabitants) in Riohacha, Colombia. Following an initial review of hospital and laboratory data, a cross-sectional household survey and case-control study were conducted to investigate this epidemic. The cross-sectional survey found an increased risk of cholera between November 1991 and September 1992 among subjects who usually drank unchlorinated piped water from the municipal water system (prevalence odds ratio, POR = 5.7; 95% confidence interval, CI = 1.2-41.1), as well as an increased risk of acute diarrheal disease in the 2 weeks preceding the survey interview among these same subjects (POR = 3.3; 95% CI = 1.1-11.2). The case-control study revealed an association between cholera and drinking unboiled tap water (OR = 7.2; 95% CI = 1.6-32.2), and also between cholera and limited availability of water (< 1,400 liters per week) within the household (OR = 3.6; 95% CI = 0.8-16.4). These findings strongly suggest that most of the Riohacha cholera cases were transmitted by contaminated municipal water, a conclusion supported by descriptive evidence of problems affecting Riohacha's municipal water and sewerage systems.