Manual removal of cotton (Gossypium hirsutum) squares as a simulated damage method has been adopted for more than a half century, and many of useful results have been obtained. Effects of removing squares at the early stage on cotton yield were variable. The reasons for the highly variable nature of the yield data have been poorly understood so that further utilization of the square removal method was limited.To reveal why there were the highly variable yield data and to try to obtain the possible benefit f...