ИСТИНА

Presence or absence of terminal flowers is traditionally used as one of the major features of inflorescence morphology. Several angiosperms that are characterized by racemose inflorescences (i.e., lacking a terminal flower) sporadically possess terminal flower-like structures (TFLS). These can occur in mutants such as the Centroradialis mutant of Antirrhinum. Flower-like terminal structures similar to those described in laboratory-induced mutants of model organisms can be found in plants from natural populations of many different angiosperms. In particular, flower-like terminal structures are found in inflorescences of early-divergent monocots, such as Alismatales. Flowers of Ruppia are normally arranged into an open two-flowered spike, but sometimes the two lateral flowers are congenitally united with each other and form a terminal flower-like structure. TFLS show considerable variation in number of tepals, stamens and carpels. Organ number can be higher or lower than is normally present in lateral flowers of the same species, such as Potamogeton, Triglochin. In these genuses filamentous and tubular structures that are clearly different from other floral organs occur at the tip of some inflorescences. We study anatomy of these filamentous and tubular structures. In Potamogeton the presence of several surrounding bracts suggests that each TFLS could be morphologically interpreted as a pseudanthium, i.e., a product of congenital fusion of the more or less reduced uppermost lateral flowers. If so, pattern formation in the inflorescence is strictly acropetal and the TFLS does not represent a real terminal structure, being an amalgamation of reduced lateral flowers. We test this hypothesis using data from spatial arrangement of vascular bundles in inflorescences of Potamogeton, Triglochin, Ruppia. At least some TFLS show patterns of vascularization that cannot be explained by the pseudanthial model. We suggest that TFLS could behave as truly terminal entities.