A graphite-furnace atomic-absorption spectrometric method, utilizing ultrasonic slurry-sampling has been developed for the determination of Ga in soils. Calibration with aqueous standards and with slurries prepared from a certified soil reference material were both employed. When calibration with soil slurries was used no modifier was needed. Because lower and more variable sensitivity was obtained for Ga in aqueous standards than for Ga in slurry soil samples, external calibration with aqueous Ga standards required a suitable chemical modifier to level out the sensitivity difference. Of the many potential modifiers tested, i.e. Al, As, Co, Mg, Mo, Ni, Pd, Pd+Mg, Se, and Te, Ni was found to be best. When Ni (1.0 mg mL^–1, 10 µL) was injected to the graphite tube with the aqueous standards or slurry samples (10 µL) accurate results were obtained. Both methods of calibration gave acceptable accuracy and precision. The repeatability was ≤4.6% (RSD) for slurry sampling for both methods of calibration. The intermediate precision for analyses performed on three different days was 7.3% and 4.8% when using aqueous standards and slurry standards, respectively. The limit of detection for Ga was 3.0 µg L^–1 in soil slurry solution, equivalent to 0.08 µg g^–1 in the soil samples. Results from slurry sampling were compared with results from microwave-digested samples. Chemical interference was observed when the acid-mixture used to dissolve the samples contained HF. This interference was eliminated when Ni was added as chemical modifier. The repeatability was ≤7.2% (RSD) for analysis of decomposed samples.