We have developed a method for direct determination of arsenic in geological samples using ICP-MS by reduction of interferences, without preconcentration, separation and use of the hydride generation technique. Concentrations of HNO₃ have a significant effect on the arsenic signal. This type of interference cannot be corrected by internal standards (Rh and In) because the signal suppression due to HNO₃ is apparently dependent on the first ionization potential of elements. Addition of 4% (v/v) ethanol to 1–10% (v/v) HNO₃ was found to be an excellent method for reducing this type of matrix effect from 30–40% to less than 5% for high first ionization potential elements ⁷⁵As (9.81 eV), ⁸²Se (9.75 eV), and ¹²⁶Te (9.01 eV). Direct determination of arsenic in geological samples by ICP-MS is often complicated by the presence of Nd²⁺, Sm²⁺and Eu²⁺ interferences, in addition to the well known interference of ArCl⁺, and the high first ionization potential of As (9.81 eV) also results in relatively low analytical sensitivity in ICP-MS. It is shown that both problems can be overcome by a combination of a 4% ethanol matrix modifier with nebulizer gas flow rate adjustment. For example, the interference from doubly charged ions of a rare earth element (Ce²⁺) is reduced by a factor of 30 with the addition of 4% ethanol at a nebulizer gas flow rate of 1.00 l min⁻¹ and rf power of 1350 W, while the signal intensity of As is similar in both solutions. A nebulizer gas flow rate of 0.94 l min⁻¹, an rf power of 1350 W and 4% ethanol modifier were chosen in practical sample analysis. Under these conditions, the interference of doubly charged ions of the rare earth element (Ce²⁺) was reduced by a factor of 6.5 and the signal intensity of As was improved by a factor of 3 relative to that in 3% (v/v) HNO₃ solution at the corresponding optimum nebulizer gas flow rate of 0.98 l min⁻¹ and rf power of 1350 W. The arsenic equivalent concentration caused by ArCl⁺ interference was reduced by a factor of 10 under our given experimental conditions in the presence of 4% (v/v) ethanol. The developed method was applied to the direct determination of arsenic in a series of international geological reference materials. Most of the results were found to be in reasonable agreement with the reported values in the literature, particularly for those having recommended values. This simple method shows a great potential for the direct determination of arsenic in geological and environmental samples.