Flower or spikelet? Understanding the morphology and development of reproductive structures in Exocarya (Cyperaceae, Mapanioideae, Chrysitricheae).

Fundamental questions of floral morphology remain unresolved in the grasslike monocots in order Poales, including what constitutes a flower and what constitutes a spikelet. The mapaniid sedges have particularly complex spikeletlike structures, variously interpreted as clusters of flowers or spikelets. Recent phylogenetic studies of Cyperaceae have identified the mapaniid clade as sister to the rest of the family, but the homology of mapaniid reproductive units (RUs) and spikeletlike units (SLUs) to other sedge flowers and spikelets is unclear. We examined reproductive development in the mapaniid Exocarya sclerioides. Inflorescence branches terminated in a SLU with bracts and 1-4 RUs. RUs had four small leaflike structures (LLSs): two lateral LLSs, each associated with a stamen, an abaxial LLS associated with a stamen, and an adaxial LLS. The gynoecium terminated the RU. All RUs were axillary to bracts, and unexpanded bracts and RUs were produced beyond expanded RUs, so SLUs were racemose. RUs developed from a single primordium that initiated two lateral LLSs, then two lateral stamens, then the gynoecium. Initiation of the abaxial LLS and stamen and the adaxial LLS followed. We hypothesize that the RU is a sympodial branch that terminates in a hermaphroditic flower with two stamens and a gynoecium; the two lateral LLSs are halves of a deeply divided prophyll.