Owing to the similar chemistry of phosphorus (P) and arsenic (As), sodium bicarbonate (0.5 N NaHCO 3 ) is commonly used to extract plant accessible As in soil. However, 0.5 N NaHCO 3 is not compatible with the ICP-MS measurement due to the large amount of dissolved solids. This investigation set out to devise a suitable extractant for determining extractable As in soil and measured by the ICP-MS. Paired soil and plant samples were collected from paddy elds in West Bengal, India. Soil was extracted with 0.5 M NaHCO 3 , 0.03 M (0.1 N) and 0.17 M (0.5 N) phosphoric acid (H 3 PO 4 ), 0.05 M (0.1 N) and 0.25 M (0.5 N) sulfuric acid (H 2 SO 4 ), and 0.01 M calcium chloride (CaCl 2 ). This made it possible to measure As by hydride generation-atomic absorption spectrophotometer (HG-AAS), while ICP-MS was used for the determination of As extracted from soil with different concentrations (0.1-1.5M) of HNO 3 . The NaHCO 3 extractable As was 1.45 and 1.23 mg kg − 1 for soil to extractant ratio of 1:20 and 1:5, respectively. Of these extractants, 1.5 N HNO 3 extractable As had the best correlation with As content in rice grain (r = 0.45**) and straw (r = 0.71**), comparable with standard extractant i.e. 0.5 N NaHCO 3 (r = 0.47** and r = 0.64** in case of grain and straw, respectively). A signi cant positive relationship of 1.5 N HNO 3 was obtained with 0.5 N NaHCO 3 . Thus, 1.5 N HNO 3 may serve as an extractant for soil As, which is compatible with ICP-MS analysis.The extractants to remove bioavailable As from soil is 0.5 N NaHCO 3 (1:20) but it is not compatible with ICP-MS estimation of As due to its high salt content.This study evaluated soil properties, different acids and salt extractants, and their correlation with plant As content. Arsenic extracted with most common extractant 0.5 N NaHCO 3 (1:20) was 1.45 mg kg − 1 while 1.5 N HNO 3 extractable As was 1.42 mg kg − 1 . A more or less similar amount of As is extracted from soil by both extractants. Further correlation analysis revealed that pH has a positive correlation while OC content has a negative correlation with grain and straw As content. A relatively higher correlation was observed with grain and straw As content with 0.5 N NaHCO 3 (1:20), followed by 1.5 N HNO 3 . The highest correlation was observed between 0.5 N NaHCO 3 (1:20) extractable As and 1.5 N HNO 3 extractable As. Nitric acid is relatively compatible with ICP-MS. Thus, 1.5 N HNO 3 may be used as a suitable extractant for extracting bioavailable As and subsequent estimation in ICP-MS.