Lakhasly

Phylogenetic analyses of Rhododendron confirm Goetsch et al.'s (2005) findings, but reveal multiple transitions between corolla monosymmetry and polysymmetry within the genus, contradicting Leppik's (1974) assertion of monosymmetry as a derived state. Clade A exhibits the most character state variability and lability, uniquely possessing deflexed styles. Analyses tested correlations between floral monosymmetry and pigment patterns, and monosymmetry's derived status. Maximum parsimony suggests an ancestral monosymmetric Rhododendron with pigment patterns, but corolla symmetry remains uncertain; maximum likelihood provides ambiguous results. The Indian species' phylogenetic placement aligns with traditional subgeneric classification, but sectional classification lacks strong support. Strong evolutionary correlations exist between monosymmetry in different floral whorls (except between bundled stamens and flexed styles). Pigment patterns are closely linked to monosymmetry types in the corolla, androecium, and gynoecium, suggesting co-evolution possibly driven by pollinator selection. Further research in plant ecology, floral development, and genetics is needed. While pollination data is limited, some monosymmetric species with spots/blotches, reflexed stamens, and flexed styles are known to be pollinated by bees or butterflies, while some polysymmetric species are bird-pollinated. Previous studies primarily focused on perianth symmetry or symmetry-governing genes. Fossil evidence indicates floral monosymmetry evolved ~50 Myr ago, coinciding with pollinator diversification. Comparative developmental studies could elucidate the lability of traits influencing floral symmetry. Rhododendron flowers are generally pollinated by bees, bumblebees, butterflies, and potentially flies and birds.

The results of phylogenetic analyses are congruent with those of Goetsch et al. (2005).Within the genus, however, multiple transitions occur between corolla mono- symmetry and polysymmetry, unlike the previous assess- ment that monosymmetry is generally a derived state in Rhododendron (Leppik 1974).The three clades differ in the patterns of evolution--deflexed styles evolved only in Clade A, which also is the most variable in terms of both range of character states (all states for all characters), as well as lability (gains and losses in equal measure).We tested the hypotheses that in Rhododendron: a) flo- ral monosymmetry and pigment patterns show correlated evolution and b) floral monosymmetry is a derived trait.Under MP, the ancestral Rhododendron may have been monosymmet- ric (reflexed stamens and styles) with pigment patterns, but symmetry of the corolla is uncertain.Recon- struction of ancestral states is much more ambiguous using ML than MP; discrepancies between MP and ML character reconstruction are seen in other analyses too (Cunningham et al. 1998; Geeta et al. 2012).Phylogenetic placement of the Indian species is consistent with their traditional place- ment in subgenera Hymenanthes and Rhododendron, but sectional classification is not as well supported.Similarly, we found strongly supported evolutionary correlations in the monosymmetry of different whorls (except between bun- dled stamens and flexed styles).arboreum, R. thomsonii, R. campanulatum and R. macabeanum. In Rhododendron, the occurrence of pigment patterns is tightly linked to the different types of monosymmetry con- tributed by the corolla, androecium and gynoecium.The correlations detected sug- gest that these floral traits are evolving together, perhaps under pollinator-mediated selective pressures (Brothers and Atwell 2014; Davis et al. 2014).Further investigations of related aspects in the fields of plant ecology, floral trait develop- ment, and genetics may give a broader understanding of the evolution of floral forms in angiosperms.Trends in the evolution of subtle monosymmetry in floral traits would reveal whether the pattern of corre- lations we observe here can be extended to angiosperms.We do not have detailed information on pollination of most of the species in this study, but can note that some species, such as R. ponticum L., R. semibarbatum Maxim., R. aureum Georgi and R. calendulaceum (Michx.) Torr., with monosymmetric corolla, spots/blotches, reflexed stamens and flexed style, have been reported to be pollinated by bees or butterflies (Stout et al. 2006; Hirao et al. 2006; Ono et al. 2008; Sugiura 2012; Epps et al. 2015); in some cases, R. macabeanum and R. thomsonii, with polysymmetric corolla, spots/blotches, hooked stamens and hooked style, bird pollination is impli- cated (Basnett 2013; Jing et al. 2015).Most earlier studies on phylogenetic patterns of floral symmetry in angiosperms analysed either symmetry of the perianth (but not other floral whorls), or genes that govern overall floral symmetry (Donoghue et al. 1998; Damerval and Nadot 2007; Citerne et al. 2010; Hile- man 2014). Fossil records show that floral monosymmetry evolved ~ 50 Myr ago and that the diversification of this trait coin- cided with the diversification of insects acting as specialized pollinators (Cronk 2001; Sargent 2004; Soltis et al. 2004; Citerne et al. 2010).The developmental basis of the lability of particular traits leading to variation in floral symmetry may be investigated by comparative developmental studies of sta- men arrangement and flexion of filaments and style, as done on Moringa (Olson 2003).In general, flowers of Rhododendron are said to be pollinated by bees, bumble bees and butterflies, with the additional possibility of flies and birds (Leppik 1974; Sugiura 2012; Georgian et al. 2015)