- The potassium status of selected soils from central and eastern Oregon, that have shown a wide range in response from potassium was investigated. The silt and clay minerals were identified using X-ray crystallography. The potassium extracted by water, NH₄OAc, and HNO₃ was evaluated on the total soil and on the clay, silt and sand fractions. A wide range in particle size of the soils included in this study was found, especially in the sand and silt fractions. The sand and silt fractions varied from 87 and 10%, respectively in the Quincy soil to 20 and 61%, respectively in the Nyssa soil. The clay and silt fractions of the Deschutes, Poe and Henley soils were dominated by amorphous materials with minimum indication of other minerals. These soils represented the extremes in NH₄OAc and HNO₃ extractable potassium. The Deschutes soils had 196 and 643 μg/gm of NH₄OAc and HNO₃ extractable potassium, respectively. The Poe soils averaged 484 and 1394 μg /gm of NH₄OAc and HNO₃ extractable potassium, respectively, while the Henley soil, selected for its luxuriant supply of potassium, contained 1384 and 3449 μg/gm of NH₄OAc and HNO₃ extractable potassium respectively. In contrast, the Quincy, Ephrata, Metolius, Agency, Madras, and Nyssa soils contained no amorphous materials. In these soils, smectites and mica were the dominant minerals in the clay fraction and vermiculite, mica, and feldspars were the dominant minerals in the silt fraction. The NH₄OAc extractable potassium was 250 μg /gm in the Quincy soil, 353 μg/gm in the Ephrata soil, 305 μg /gm in the Metolius soil, 427 μg /gm in the Agency soil and 560 μg /gm in the Nyssa soils, while the NH₄OAc averaged 476 μg/gm in the Madras soils. The HNO₃ extractable potassium varied from 975 μg /gm in the Metolius soil to 2232 μg /gm in the Nyssa soil. The NH₄OAc and HNO₃ extractable potassium extracted from any one group of soils, increased with the increase in clay and silt, and the decrease in sand. The exchangeable potassium associated with the clay fraction increased with the presence of smectites. The HNO₃ extractable potassium associated with the silt increased with the presence of mica, vermiculite, and potassium feldspar in this fraction. The soils included in this study also represented a wide range in response from potassium. Marked potassium response was found on the Deschutes soils but not on the Poe and Henley soils with similar texture, sandy loam, and similar mineralogy, mostly amorphous materials. Response from potassium was also found on the Metolius, Agency and Madras (#10) soils but not on the Madras (#12 and 13) with similar texture and mineralogy. The Quincy, Ephrata and Nyssa soils, with similar mineral composition, have shown no response from potassium. Critical potassium levels of about 440 μg /gm for NH₄OAc or 1400 μg /gm for HNO₃ could be established for all soils from central Oregon. However, when these critical levels are used on soils from other areas, the soils on which response from potassium might be measured, are not identified. Separate critical potassium soil test values should be established for both NH₄OAcand HNO₃ when soils from different areas with marked differences in parent material, minerals present, texture, and potassium crop removal are evaluated for response from potassium. Potassium fertilization recommendations should be based primarily on the relationships between a soil test value and potassium response on field experiments on similar groups of soils for each crop.