Increasing costs of inputs in upland cotton (Gossypium hirsutum L.) have highlighted the need for more sophisticated within-season management of the crop. The objective of this study was to investigate the interaction between the timing of water stress and simulated pest damage in their effects on cotton yield and yield components. A field experiment was repeated throughout 5 yr on a Gan Shmuel vertisol (clayey, montmorillonitic, calcareous, thermic Typic Cbromoxerert). Treatments comprised a control [post-emergence irrigations at first flower (f), f+23, and f+38 d]; a delay in this sequence from the first (f+ 8, f+ 33, and f+ 47 d), second (f, f+ 32, and f+ 46 d), or third irrigation (f,f+23, and f+48 d); four irrigations (f−9,f+ 12,f+ 27, and f+ 40 d); and severe mid-season stress (f−10, f+ 23, and f+38 d). Total irrigation amount was constant. The dates of the six irrigation schedules were repeated with pest damage simulated by the early removal of 120 floral buds meter−2. There was a significant stress-damage interaction in flower, boll, and fiber production. In comparison with the undamaged control, boll production was reduced by damage alone and by stress alone. All stress-damage combinations, however, restored boll production to the control level. Only severe mid-season stress without damage, reduced seed cotton yield. Treatment effects were detected despite an underlying conservatism in crop response, whereby flower retention was inversely related to flower production. This resulted in low variability of boll number. The effect of boll number on yield was in turn buffered by boll weight. Consequently, irrigation scheduling and the control of pests that damage fruiting bodies cannot be optimized independently.
Contribution no. 3155 - E, 1990 series, from the ARO, Bet Dagan, Israel.