New therapies are needed to improve the prognosis of patients with metastatic melanoma. This study evaluates the safety and efficacy of weekly paclitaxel in patients with metastatic melanoma. Patients received paclitaxel at 80 mg/m over 1 h, weekly for 3 weeks, followed by a 1-week rest period. Disease status was assessed every other cycle. Treatment was continued until patients experienced either disease progression or unacceptable toxicity. Twenty-seven patients were enrolled in this phase II clinical trial. Of these patients, two were subsequently determined to be ineligible. All patients, however, were considered to be evaluable for toxicity and all patients were included for response assessment in an intention-to-treat analysis. Patients received paclitaxel for a median of two cycles (range, <1-8). None of the 27 patients showed a response to treatment. Eight patients had stable disease. The median progression-free survival was 1.8 months (95% confidence interval, 1.7-2.5 months) and the median survival was 7.6 months (95% confidence interval, 4.7-9.7 months). The most common grade 3 toxicity was neutropenia (four patients) and one patient had grade 4 neutropenia. Other treatment-related grade 3 toxicities included hypersensitivity reaction (one patient) and diarrhoea (one patient). Of note, five patients had peripheral neuropathy; however, in each case, the neuropathy was only grade 1. Weekly paclitaxel is relatively well tolerated and can maintain disease stability for some metastatic melanoma patients. Unfortunately, the anti-tumour activity of this single-agent therapy is low and additional treatment innovations are needed.