Polygala
Description
Annual or perennial herbs, shrubs, small trees or rarely lianas.
Leaves alternate or verticillate, sometimes (sub)opposite, rarely caducous, sometimes scalelike or apparently absent.
Inflorescences raceme-like or rarely paniculate, (supra-)axillary, terminal (and then often overtopped by side-branches) or lateral, sometimes in dense heads.
Sepals 5 or the lower pair connate, unequal, caducous or persistent in fruit, the two lateral ones (alae) at least twice as large as the other ones and often petaloid.
Petals 3, unequal, halfway adnate to the staminal tube and sometimes mutually connate into a single 3-lobed petal;
Stamens 8, rarely 6, monadelphous or partly dia- or triadelphous;
Ovary 2-celled, each locule with a single subapical ovule;
Capsule 2-celled or very rarely 1-celled by abortion, compressed contrary to the sept, often more or less winged, sometimes with a double wing, dehiscing by a marginal split, reniform to oblanceolate.
Seeds various, usually at micropylar side with a lobed or un-lobed appendage (aril) and/or with an elongate appendage, at the opposite cha-lazal side sometimes with a variously shaped appendage, rarely appendages at both sides lacking;
Distribution
Asia-Tropical: New Guinea present, Australasia, Eurasia present, North Africa present, North America present, Pantropical present, South Africa present, Southeast Asia present, Tropical Africa present, tropical South and Central America present present
A polytypic genus comprising at least 500 spp. in tropical, subtropical, temperate and montane regions. The majority of species grow in tropical South and Central America, where Polygala also has its greatest infrageneric diversity. Both in North America and in South Africa secondary centres of speciation occur. Eurasia, North Africa, Malesia and Australia are rather poor in species. In Malesia the species can be referred to four little-related sections. Sect. Melchiora is monotypic and endemic to New Guinea, sect. Pseudosemeiocardium is common in Southeast Asia, sect. Chamaebuxus occurs over all continents except tropical South and Central America, and the largest (probably unnatural) sect. Polygala covers the same area as the genus. The majority of species in Malesia have rather restricted areas, with some notable exceptions, e.g., P. persicariaefolia, which also occurs in tropical Africa, and P. paniculata, an American species which is now a pantropical weed. Two species are endemic in Malesia.
Pollination
Self-pollination probably occurs in all species, although the flowers of the majority are attractive to insects. Many large-flowered species show structures which seem to be adaptations to pollinating insects (FAEGRI & VAN DER PIJL, 1979, for P. chamaebuxus), but in those species, too, self-pollination seems to occur frequently.
In a number of species pollen grains are deposited directly on the stigma, often already in the unopened flower (VENKATESH, 1956). In many species the style tip bears special pockets in various ways in which the pollen grains are deposited from the anthers. At that time the stigma can theoretically still be pollinated with foreign pollen grains, as for instance in P. lutea (MILLER, 1971) and P. vulgaris (HEUBL, 1984). According to HEUBL'S observations, however, a visiting insect will effect self-pollination. Yet there are rather reliable records of hybrids of the latter species in nature (HEUBL, l.c.), and Fi-plants of artificially produced hybrids are often fertile. This must mean that cross-pollination (rarely) occurs in P. vulgaris. Only BRANTJES (1982) observed and described cross-pollination in the two Brazilian species P. monticola H.B.K. and P. vauthieri CHODAT. The pollination mechanism of those species is a very precise one, and if cross-pollination fails, the flowers are effectively self-pollinated.
Perhaps such examples of precision cross-pollination can be discovered in many other Polygala species, because it is rather unlikely that the great variation in style tips and stigma forms could have evolved in complete absence of gene-exchange. It is also possible that many species in which formerly cross-pollination was rare, have lost this possibility.
In a number of species pollen grains are deposited directly on the stigma, often already in the unopened flower (VENKATESH, 1956). In many species the style tip bears special pockets in various ways in which the pollen grains are deposited from the anthers. At that time the stigma can theoretically still be pollinated with foreign pollen grains, as for instance in P. lutea (MILLER, 1971) and P. vulgaris (HEUBL, 1984). According to HEUBL'S observations, however, a visiting insect will effect self-pollination. Yet there are rather reliable records of hybrids of the latter species in nature (HEUBL, l.c.), and Fi-plants of artificially produced hybrids are often fertile. This must mean that cross-pollination (rarely) occurs in P. vulgaris. Only BRANTJES (1982) observed and described cross-pollination in the two Brazilian species P. monticola H.B.K. and P. vauthieri CHODAT. The pollination mechanism of those species is a very precise one, and if cross-pollination fails, the flowers are effectively self-pollinated.
Perhaps such examples of precision cross-pollination can be discovered in many other Polygala species, because it is rather unlikely that the great variation in style tips and stigma forms could have evolved in complete absence of gene-exchange. It is also possible that many species in which formerly cross-pollination was rare, have lost this possibility.
Taxonomy
In view of the surprisingly great variation in species-constant characters of diverse flowering and fruiting parts, it has been tried to split this large genus into a number of smaller genera. The large genus concept as adopted by R. CHODAT (), the only author who revised all species, is still adopted, some minor questions regarding the status of some American groups set aside. CHODAT'S infrageneric division of the genus, however, was unsuccessful, as had been pointed out by S.F.BLAKE in his revision of the North American species (1924), and more recently by ADEMA (1966) for the Malesian ones. This is probably due to the fact that two of CHODAT'S major characters to define his sections, viz. the presence or absence of sepals at fruiting state, and of carinal appendages, are not constant in the different lineages: parallel developments (reductions c.q. reversions of character states) will have been developed in many of these lineages. Until a new revision of the tropical American species has been made, it will be impossible to make a balanced reconsideration of the status of the diverse groups of species within and outside the borderlines of this genus.
In the Indo-Australian area a great diversity of species is present, probably caused by diverse migration lines. Perhaps the Australian genus Comesperma should also be included in Polygala; see note under 4. P. papuana.
In the Indo-Australian area a great diversity of species is present, probably caused by diverse migration lines. Perhaps the Australian genus Comesperma should also be included in Polygala; see note under 4. P. papuana.
Citation
BLAKE 1924 – In: N. Amer. Fl.: 305
CHODAT 1891 – In: Monogr.
LINNE 1754: Gen. Pl., ed. 5: 315
ADEMA 1966 – In: Blumea: 256
HASSK. 1863 – In: Miq., Ann. Mus. Bot. Lugd.-Bat. 1: 151
OPIZ 1839: Oekon. Neuigk. Verh.: 526
E. & P. 1896 – In: Nat. Pfl. Fam.: 330
HUTCH. 1967 – In: Gen. Fl. Pl.: 340
MUKHERJEE – In: Bull. Bot. Soc. Beng.: 29
DC. 1824 – In: Prod.: 321
LINNE 1893 – In: Monogr.
BENTH. & HOOK. 1862 – In: Gen. Pl.: 136