Effects of early square (floral bud) removal on cotton (Gossypium hirsutum L.) development were evaluated in normal and extremely late plantings in southern Arkansas in 2000 (Teague et al 2001). The experiment was repeated in a normal date of planting in Northeast Arkansas in 2001. Squares were removed either by feeding by tarnished plant bug (Lygus lineolaris Palisot de Beauvois) or manually by crushing. Injury treatments were initiated when first squares were visible, 41 days after planting (DAP), and continued at 48 and 55 DAP. All visible squares were crushed on each treatment date. For plant bug treatments, 3 nymphs, 2nd to 3rd instar, were released per plant. Plant response was monitored using COTMAN™ in-season with final plant mapping done using COTMAP. Square shed of 1st position squares at 59 DAP ranged from 8% in uninjured (protected) cotton to 39 and 45% following Bug or Crush treatments, respectively. No. of days to physiological cutout (nodes above white flower (NAWF) = 5) was delayed by 8 days for plants with insect induced injury compared to plants protected by insecticide. Differences in final plant structure and crop compensation following plant bug injury compared to manual square removal were observed. Mean maturity date for plants injured by plant bugs was 6 days later than either manually injured plants or protected plants. Plots were picked 5 times, and results indicated significant differences between treatments in the first 4 harvests. By the final harvest on 19 Oct, there were no statistical differences in final cumulative yield. Data from the 2001study were similar to observations made in 2000 and indicated that crop response to injury from tarnished plant bug feeding was measurably different from response to manually induced injury. Results provide evidence that researchers should use insects in their research protocol rather than manually induced injury, if their goal is to accurately study plant response to feeding by tarnished plant bug.