TAXONOMIC RELATIONSHIPS OF SALICACEAE AND FLACOURTIACEAE: THEIR BEARING ON INTERPRETATIVE FLORAL MORPHOLOGY AND DILLENIID PHYLOGENY

SUMMARY The manifest relationships between Salicaceae and Flacourtiaceae form the basis of deductions concerning the floral morphology of the Violales (Cistales, Cistiflorae, etc.). The functional reproductive units (conventional flowers) of the Flacourtiaceae, and by inference of at least some other dilleniid families, are frequently anthoidal in nature. This implies that the Dilleniidae cannot possibly be descendants of a magnolioid progenitorial taxon with holanthocormous flowers. The diclinous taxa of the Violales are not secondarily derived from monoclinous representatives of this order, but represent an ancestral condition retained in the diclinous forms (and in the Salicaceae). However, a derivation of the monoclinous forms from extant taxa with unisexual reproductive regions is not acceptable either. Both groups of taxa originated from a common ancestral type of plant in which an original dicliny was partly changing into an incipient ambisexuality. The Salicaceae are more primitive than the Violales in several respects, but more specialised in other ones, particularly in the reductions of the meromonandrial properianth members of the anthoids. The Salicaceae must be classified, as a family in juxtaposition of the Flacourtiaceae, in the Violales, or may even be included in the Flacourtiaceae as a tribe or subtribe. The relations between Populus and Salix are discussed and the generally more basic position of Populus is emphasised; the consequences of the assessment of their degree of phylogenetic advancement are confronted with some palynological and anthecological data.