The toxicity and carcinogenicity of arsenic depend on its species. Individuals living in Japan consume much seafood that contains high levels of organoarsenics. Speciation analysis of urinary arsenic is required to clarify the health risks of arsenic intake. There has been no report of urinary arsenic analysis in Japan using high performance liquid chromatography with inductively coupled plasma mass spectrometry (HPLC-ICP-MS). We performed speciation analysis of urinary arsenic for 210 Japanese male subjects without occupational exposure using HPLC-ICP-MS. The median values of urinary arsenics were as follows: sodium arsenite (AsIII), 3.5; sodium arsenate (AsV), 0.1; monomethylarsonic acid (MMA), 3.1; dimethylarsinic acid (DMA), 42.6; arsenobetaine (AsBe), 61.3; arsenocholine, trimethylarsine oxide, and unidentified arsenics (others), 5.2; and total arsenic (total As), 141.3 microgAs/l. The median creatinine-adjusted values were as follows: AsIII, 3.0; AsV, 0.1; MMA, 2.6; DMA, 35.9; AsBe, 52.1; others 3.5; and total As, 114.9 microgAs/g creatinine. Our findings indicate that DMA and AsBe levels in Japan are much higher than those found in Italian and American studies. It appears that the high levels of DMA and AsBe observed in Japan may be due in part to seafood intake. ACGIH and DFG set the BEI and BAT values for occupational arsenic exposure as 35 microgAs/l and 50 microgAs/l, respectively, using the sum of inorganic arsenic (iAs), MMA, and DMA. In the general Japanese population, the sums of these were above 50 microgAs/l in 115 (55%) samples. We therefore recommend excluding DMA concentration in monitoring of iAs exposure.