1) We tested the hypothesis that greening of the floral (involucral) bracts of Dalechampia scandens blossoms after pollination (when bracts are white) increases carbon assimilation and provides photosynthate to developing seeds. 2) We investigated the importance of the involucral bracts for the process of seed development in two ways. First, we removed or shaded bracts of hand-pollinated blossoms to prevent their photosynthesis and tested the effects of these manipulations on seed development. Secondly, we measured the photosynthetic rate of blossoms with white vs. green bracts and compared these rates with those of leaves. 3) After four weeks of development, seeds from blossoms with bracts removed or shaded were lighter than those produced by unmanipulated blossoms. Furthermore, although the area- based photosynthetic rate of green bracts was much lower than that of leaves, it was much greater than that of white bracts. Estimates of the daily carbon budget based on these mea- surements indicate that photosynthesis in green bracts is sufficient to meet the respiratory demand of the whole blossom, but not so in white bracts. 4) Our results support the hypothesis that colour change in D. scandens bracts allows carbon assimilation that contributes to the carbon demand of nearby developing seeds.