Soybean [Glycine max L. (Merr.)] producers across the United States are confronted with significant yield losses every year from severe hail storms. A lot of research has been conducted on the impact on soybean physiology and recovery in response to defoliation. However, little information exists in response to soybean node removal which could influence replant decisions. The objective of this study was to test the ability of soybean recovery to node removal at different growth stages on soybean grain yield and composition. Research was conducted in Iowa and Indiana from 2003 to 2004. The experimental design was a randomized complete block in a split-plot arrangement with four replications. Main plots were date of node removal (V1-V3, V4-V6, and R3 in Iowa and V1-V3 and V4-V6 in Indiana). The sub-plots were six levels of node removal (0, 20, 40, 60, 80, and 100%). Plot size of the sub-plot experimental units was 3 by 7.6 m. Data from the two states were consistent with the greatest seed yield reductions occurring at R3, followed by V4-V6, and with the smallest yield loss at V1-V3 (up to 10%). Plants were not able to recover from a 100% node removal after V3. Despite of plant growth stage, increasing the level of node removal decreased the seed mass, plant height, and final plant population but increased the protein and oil content. The data indicate that node removal affects grain yield and grain composition significantly and should be considered in a replant decision. The project will continue in 2005.