Although effective therapies for pathological gambling exist, their uptake is limited to 10% of the target population. To lower the barriers for help seeking, the authors tested an online alternative in a randomized trial (N = 66). The participants were pathological gamblers not presenting with severe comorbid depression. A wait-list control was compared with an 8-week Internet-based cognitive behavior therapy program with minimal therapist contact via e-mail and weekly telephone calls of less than 15 min. Average time spent on each participant, including phone conversations, e-mail, and administration, was 4 hr. The Internet-based intervention resulted in favorable changes in pathological gambling, anxiety, depression, and quality of life. Composite between-group effect size (Cohen's d) at posttreatment was 0.83. Follow-ups carried out in the treatment group at 6, 18, and 36 months indicated that treatment effects were sustained (ds = 2.58, 1.96, and 1.98). This evidence is in support of Internet-delivered treatment for pathological gamblers. However, it is not clear how effective the treatment is for more severely depressed individuals. (Contains 1 figure and 1 table.)