pigmentation, and keratosis were 10.6, 183.5, and 71.0/1,000, respectively (Tseng et al., 1968). They also found that the male-to-female ratio for skin cancer was 2.9:1 and 1.1:1 for hyperpigmentation and keratosis. The prevalence appeared to increase progressively with age for all three conditions, although there was a decline in cancer and hyperpigmentation in women older than 69 years of age. The prevalence rates for skin cancer, hyperpigmentation, and keratosis showed an ascending gradient which correlated with the arsenic content of the well water. Blackfoot disease had an overall prevalence rate of 8.9/1,000 and, similar to skin cancer, displayed a dose-response relationship with the amount of arsenic in the well water. There was a significantly high association of blackfoot disease with hyperpigmentation, keratosis, and skin cancer.

The risk of bladder cancer in Taiwan was increased with intake of arsenic from water of 10 μg/kg/day (Chen et al., 1992). This increased risk has been confirmed in studies from Japan (Tsuda et al., 1995), Argentina (Hopenhayn-Rich et al., 1996), and Chile (Smith et al., 1998). Studies in U.S. populations exposed to arsenic in drinking water have not identified cancer increases (Morton et al., 1976; Southwick et al., 1981; Valentine et al., 1992).

These epidemiological associations have to some extent been replicated in animal experiments (Simeonova et al., 2000; Yamamoto et al., 1995). However, the mechanisms of arsenic carcinogenesis are not established, but may involve genetic effects (Goering et al., 1999) or perturbation of cellular signaling pathways (Simeonova et al., 2000).

Clearly, high intakes of inorganic arsenic are associated with various toxicities, including increased risks of several cancers with chronic exposure to high levels in drinking water. There is no evidence linking organic arsenic in food to any adverse effect, including cancer. Since there is no evidence available to define the mechanisms of arsenic carcinogenesis and no data to support a threshold, it is not possible to establish a health-based level of inorganic arsenic in drinking water and food. It should be noted that a recent report of the National Research Council recommended a downward revision from the current maximum contaminant level for arsenic in drinking water of 50 μg/L (NRC, 1999). Because organic forms of arsenic are less toxic than inorganic forms, any increased health risk from intake of organic arsenic from food products such as fish is unlikely.