Notes on the Evolution of Androecial Organisation in the Magnoliophytina (Angiosperms)

Abstract

This paper aims to summarize briefly and to update our ideas about androecial architecture formulated in earlier publications. Ontogenetic evidence of stamen development, viz. the initiation, arrangement and relationship of stamens to other floral morphomes, can be translated into a semophyletic scheme reflecting the phylogeny of the androecium. The ancestral androecium is discussed in the light of recent theoriesabout angiosperm phylogeny. Two divergent androecial processes are proposed for the angiosperms starting from a spiral androecium with a moderate number of stamens. However, transitions exist between spiral polyandry, numerous stamens in whorls, and chaotic polyandry. From an androecium with several alternating whorls of paired and single stamens, outer stamen pairs are retained following the successive loss of inner stamen whorls. Single stamens instead of pairs occur at the very end of this line and represent a more advanced condition. This line is mostly present in tri‐ and dimerous flowers. From the same starting point diplostemony (with two alternating whorls of single stamens) originated, again giving rise to various states usually present in pentamerous or tetramerous flowers.

Comparative floral structure and systematics in Oxalidales (Oxalidaceae, Connaraceae, Brunelliaceae, Cephalotaceae, Cunoniaceae, Elaeocarpaceae, Tremandraceae)

“…For example, flowers are commonly 5‐merous; the perianth has sepals and petals; sepals are acute, with a broad base and three vascular traces; petals have a broad, rounded plate and a narrow base, and one vascular trace (for other core eudicots, see e.g. Eames, 1931; Eckert, 1966; Endress, 1967, 1994; Rohweder, 1970); stamens are in two whorls; obdiplostemony is common (Dickson, 1864; Eckert, 1966; Gelius, 1967; Rohweder, 1970; Klopfer, 1973; Ronse Decraene & Smets, 1993, 1995, 1996, 1998); the obdiplostemonous condition is correlated with the phenomenon that epipetalous stamens tend to be smaller (shorter, thinner, narrower) than episepalous ones (Eckert, 1966), in the extreme case sterile or even lacking; anthers are dorsifixed and introrse (see, e.g. Endress & Stumpf, 1991); there is an extensive range of carpel union, from completely free to almost completely (congenitally) united, but commonly the carpels are free at least in the uppermost part of the style.…”

“…In contrast, in obdiplostemonous flowers the inner base of episepalous stamens is closer to the floral centre than that of alternisepalous stamens, and carpels (if isomerous) are alternisepalous. The developmental conditions for the two patterns were studied in detail in a range of core eudicots by Eckert (1966), and the systematic distribution was also discussed by Ronse Decraene & Smets (1993, 1995, 1998). In Oxalidales, the presence of two stamen whorls is also common.…”

Evolutionary diversification of the flowers in angiosperms

“…In basal eudicots, they characterize Papaveraceae, e.g., Eschscholzia (androecium) ( Endress, 1987a ); Berberidaceae, Podophyllum , (androecium) ( deMaggio and Wilson, 1986 ;Ronse De Craene, 2010 ); Ranunculaceae (androecium) ( Sch ö ffel, 1932 ); Buxaceae (androecium) ( von Balthazar and Endress, 2002a , b); and Tetracentron (gynoecium) ( Endress, 1986a ;Chen et al, 2007 ). In core eudicots, they have a scattered occurrence ( Ronse Decraene and Smets, 1993aSmets, , b , 1996Smets, , 1998. Examples are Brassicaceae and Cleomaceae (perianth and androecium) ( Endress, 1992 ), Apodanthaceae (perianth) ( Blarer et al, 2004 ) or Fouquieriaceae (androecium) ( Sch ö nenberger, 2009 ).…”

Early floral development and androecium organization in the sarracenioid clade (Actinidiaceae, Roridulaceae and Sarraceniaceae) of Ericales

“…Additional ontogenetic studies and reconstructions of character evolution on a large scale in Ericales are needed to determine unequivocally the origin of the second whorl of stamens in two‐whorled ericalean taxa. If the ancestral state in all (or part of) Ericales is indeed polystemonous androecia, this implies that the various supra‐ and infrafamilial groups of Ericales with two‐whorled androecia evolved independently; for example, Ronse Decraene & Smets (, ) presented a compelling argument that two‐whorled androecia evolved from polystemonous androecia.…”