1. Three experimental series of Biloxi soybeans were given photoinductive treatments of from two to ten cycles, consisting of 8 hours of natural daylight followed by 16 hours of darkness. After induction, series I and II were placed on cycles of long photoperiod, consisting of 21 hours of light and 3 hours of darkness, and series III on cycles of 16 hours of light and 8 hours of darkness. 2. Three types of controls were employed. One group of plants was kept on long photoperiod. None of these flowered. A second group was kept on natural photoperiod and a third on short photoperiod. These two groups flowered. 3. Floral buds from controls and from the three experimental series were collected progressively in order to obtain various stages in development. In the second and third groups of controls, floral development was normal and meiosis occurred as has been described by previous workers. In all the experimental groups the floral structures developed normally until the differentiation of sporocytes had occurred in the anthers. Following this, various abnormalities in meiosis were observed. 4. In plants given five or fewer photoinductive cycles, the sporocytes might begin to enlarge, accompanied by marked vacuolation of the cytoplasm and ultimate disintegration of their nuclei. Only a few of the sporocytes ever reached the metaphase stage of meiosis I. 5. In plants receiving six or more photoinductive cycles of short photoperiod, degeneration of microsporocytes in some anthers may occur as already described, but in others the sporocytes undergo apparently normal meiosis, and four groups of chromosomes are formed. In some instances degenerative changes occur at this point before cleavage into microspores takes place; in others cleavage occurs followed by degeneration before development of normal microspores has been accomplished; while in a very small percentage of instances apparently normal microspores are formed. Even those plants receiving as many as ten photoinductive cycles of short photoperiod showed a high percentage of degenerated microspores. 6. The long postinductive photoperiod employed in these experiments appears to have played a significant role in bringing about the degenerative changes observed and in suppressing the development of mature flowers.