Gunneraceae
Herbs, perennial, monoecious or polygamomonoecious [dioecious], terrestrial or amphibious, acaulous, armed, clonal or not. Roots fibrous, rhizomes with [without] triangular scales (cataphylls). Stems absent. Leaves basal, alternate, simple; stipules scalelike or absent; petiolate; blade palmately lobed [toothed], venation prominent, margins irregularly toothed, abaxial surface with stiff, hard prickles. Inflorescences ± from rhizome, terminal [from axils of distal leaves], panicles of spikes; bracts absent. Flowers often pistillate proximally, staminate distally, bisexual between, sessile or pedicellate; perianth and androecium perigynous; sepals 2 [0]; petals 0 [2]; stamens 2 [1] epipetalous; anthers basifixed, extrorse, pollen-grains 3 (–5) -aperturate, colpate, 2-nucleate; pistil 2-carpellate; ovary inferior, 1-locular; placentation apical; styles 2; stigmas 2, papillate; ovule 1, anatropous, bitegmic, crassinucellate. Fruits drupes. Seeds with oily endosperm, embryo straight.
Discussion
Genus 1, species 35–50 (1 in the flora): introduced, California; Mexico, Central America, South America, w Europe, se Asia, s, e Africa, Pacific Islands (Hawaii), Australia.
The systematic position of Gunneraceae has been uncertain for a long time. Gunnera has sometimes been included in the mostly aquatic Haloragaceae, which also have reduced flowers with an inferior ovary. Molecular studies (combining rbcL, atpB, and 18S data) suggest that Gunnera occupies an isolated position among the core “eudicots.” Some of these studies have also identified the African shrub Myrothamnus Welwitsch as the closest relative to Gunnera. Other work, based on leaf morphology and pollen, suggests a close relationship with Saxifragaceae (D. G. Fuller and L. J. Hickey 2005). Gunnera is remarkable for its symbiosis with cyanobacteria of the genus Nostoc (B. Osborne et al. 1991).
Species of Gunnera subg. Panke (Molina) Schindler are unique in having large, triangular scales between the leaves on the rhizomes. The morphological significance of these scales has been debated, and they have been interpreted as stipules, ligules, or cataphylls. Comparisons with features of the rhizomes in other subgenera support the interpretation of these structures as cataphylls (L. Wanntorp et al. 2003).
Selected References
None.