The Plagianthus Alliance

Introduction | Discussion of Data | Elements | Discussion of Classification | Research Program | References and Bibliography | Feedback

Introduction

The Plagianthus alliance is a group of related genera forming part of subtribe Abutilinae, tribe Malveae and subfamily Malvoideae, of the family Malvaceae sensu APG. The latest revision of Malvaceae, by Kubitzki & Bayer [3], includes 5 genera - Plagianthus, Hoheria, Asterotrichion, Gynatrix and Lawrencia - in the alliance.

These genera are endemic to Australia and New Zealand. However they are not the only elements of tribe Malveae native to Australia - in subtribe Malvinae one species of Lavatera (or Malva) and one or two species of Malvastrum are native, and in subtribe Abutilinae a number of species of SidaSidastrum) and Abutilon are present.

Asa Gray proposed that these genera formed a subtribe, to which he gave the name Plagianthae (which would be Plagianthinae according to current orthographic rules). This proposal has not been widely accepted. Hutchinson [4] places them in Malvinae, presumably on the basis of the generally clavate and introrsely decurrent stigmas. Bentham & Hooker [2] place them in subtribe Sideae (which would be Sidinae according to current orthographic rules), on account of the single-seeded mericarps and pendulous ovules. My opinion is that they belong to subtribe Abutilinae, on the grounds of the lack of an epicalyx and the possession of pendulous ovules - Sidinae is a polyphyletic assemblage defined by the convergent reduction of the number of seeds per mericarp to 1. However, I would not have been surprised to find them to belong in Malvinae, or distinct from either recognised subtribe.

The molecular data (ITS) of Tate et al [5] indicates that these genera form a monophyletic group, possibly with the inclusion of the Australian Sida (sect. Hookerianae) hookeriana, and that a clade consisting of these, plus the North American Sida (sect. Pseudonapaea) hermaphrodita, and, less certainly, the South American genus Sidasodes, is the sister group to the remainder of Abutlinae. Unpublished cpDNA and GBSSI data, presented at the Botany 2005 conference, excludes at least Rhynchosida from the sister group to this clade.

Discussion of Data

Hoheria and Plagianthus are endemic to New Zealand, the former extending to the Kermadec Islands, and the latter to the Chatham Islands. Asterotrichion is endemic to Tasmannia and Gynatrix to south east Australia. Lawrencia is endemic to Australia, being found in all states and the Northern Territory, but is most diverse in Western Australia and Northern Territory.

Hoheria and Plagianthus universally have a diploid chromosome count of 42, thus sharing the base x=7 number common in Malveae, including Malva and Abutilon.

nrDNA data from Tate et al [5] samples most groups of the alliance, but omits Hoheria section Apterocarpa; I do not kniow whether Lawrencia subgenus Panifex was sampled.

Plants of the Plagianthus alliance, with the exception of Lawrencia, are large shrubs and small trees.

The flowers lack epicalyxes and have pendulous ovules. They typically have introrsely decurrent clavate stigmas, but this characteristic is not universal, Hoheria having capitate or peltate stigmas. There is a tendency to produce unisexual flowers, but Hoheria and some species of Lawrencia have hermaphrodite flowers. There is a tendency to have a reduced number of ovary locules (only one in Plagianthus), with the number of seeds further reduced in Asterotrichion and Lawrencia section Selenothamnus; however the number of mericarps in Hoheria and some species of Lawrencia exceeds the number of 5 which I believe to be the basal state in Malvoideae. The petals are typically white, cream or pale yellow, but may be a darker yellow, or green, or red, or purple, in some species of Lawrencia. The staminal column is short, to the degree that the malvaceous nature of the flowers is not always recognisable at a glance.

The leaves are commonly ovate, elliptic or lanceolate. They commonly have palmatipinnate or pinnate venation, rather than the palmate venation typical of Malvoideae.

Elements

Plagianthus J.R. Forster & G. Forster

Tree (Plagianthus regius) or shrub (Plagianthus divaricatus), with divaricate branching at least in juvenile plants; leaves palmatipinnate; flowers unisexual, calyx campanulate, corolla yellowish-white, ovary 1-locular; style branches 2, sometimes 3, clavate, stigmas introrsely decurrent, carpels solitary, subglobose, ±asymetrical, chartaceous.

Two species from New Zealand.

Hoheria section Hoheria

Small trees, indumentum stellate, leaves ±elliptic, palmatipinnate to pinnate, stipules small, caducous, flowers hermaphroditic, solitary or in small cymes, calyx campanulate, corolla white, petals ±asymmetric, staminal column short, ovary 5-8-locular, stigmas capitate or peltate; mericarps dorsally winged, seeds ±flattened.

Hoheria section Apterocarpa

Small trees, indumentum stellate, leaves ±elliptic, palmatipinnate to pinnate, stipules small, caducous, flowers hermaphroditic, solitary or in small cymes, calyx campanulate, corolla white, petals ±asymmetric, staminal column short, ovary 10-15-locular, stigmas capitate or obliquely clavate, mericarps not winged, seeds ±flattened.

Asterotrichion Klotzsch

Dioecious shrub, indumentum stellate, leaves lacking prominent palmate venation, stipules absent, flowers axillary or terminal, staminodes 5, sessile in female flowers, 15, filaments ± as long as the column in the male, calyx tubular-campanulate in female flowers, campanulate in male, ovary 2-locular, style arms clavate, fruit often 1 seeded.

Gynatrix Alefeld

Dioecious shrubs, indumentum stellate, leave with palmatipinnate venation, stipules lanceolate, flowers in ±condensed inflorescences, calyx campanulate, anthers 20, sessile in female flowers, filaments at 1/3 to 1/2 the length of the column, ovary 5-locular, style branches subclavate, fruit schizocarpic, mericarps loculicidally dehiscent.

Lawrencia Hook. (excluding section Selenothamnus)

It would be preferable to cover Lawrencia subgenus Lawrencia section Lawrencia and Lawrencia subgenus Panifex separately, but I do not having sufficent material on the composition and appearance to do so.

Dioecious, androdioecious (L. glomerata), hermaphroditic shrubs or subshrubs, leaves petiolate to sessile, inflorescences of solitary flowers or few-flowered panicles in the leaf axils, sometimes subspicate, flowers sometimes sessile, calyx turbinate, ±deeply 5-lobed, petals sometimes ± pubescent; staminal tube shorter than the petals, anthers 5-30, sterile in the female; ovary 2-11-locular, ± reduced in male; ovules 1 per locule, pendulous; style 0.4-10mm long; stylar branches as many as locules, free to the base or shortly united; stigmas introrsely decurrent, Mericarps 1-seeded, dehiscent or indehiscent, distinct, hyaline to coriaceous, with reticulate thickenings.

Lawrencia subgenus Lawrencia section Selenothamnus

Fleshy-stemmed shrubs, indumentum lepidote, leaves oblanceolate, linear or scale-like, sometimes 3-toothed at the apex, flowers unisexual, axillary; calyx tubular, plicate, contracted at the mouth with short triangular lobes, corolla white, yellow, green, red or purple, ovary 2-4-loculate, style arms linear, stigmatose along their entire length, fruit subglobose, assymetrical, irregularly dehiscent, seeds 1 rarely 2.

Lawrencia squamata is unique among the Plagianthus alliance in its spiny or thorny habit.

Discussion of Classification

As there is no molecular sequence data available, and chromosome counts are available only for Hoheria, we cannot use these to illuminate the classification of the Plagianthus alliance, except to note that the latter are consistent with the unity of Hoheria. Hence we are restricted to morphological, hybridisation and biogeographical data.

Each group has separate suites of ancestral and derived characters, but it is not obvious from these how the various genera are related. The nrDNA of Tate et al [5] has Plagianthus, Gynatrix and Asterotrichion as a trichotomous clade, to which the sister group is Hoheria. Lawrencia, or perhaps Lawrencia+Sida hookeriana, is the sister group to the remainder of the alliance, and section Selenothamnus is nested within the remainder of Lawrencia. Although Hoheria section Apterocarpa was not sampled by Tate et al, the existence of intersectional hybrids, and a lack of recorded hybrids with other genera, suggests that section Apterocarpa is correctly placed in that genus, rather than in Plagianthus or Gaya as has been suggested. The chromosome counts (2n=42) are inconsistent with a placement in Gaya (2n=12). The unity of Plagianthus is additionally supported by the existence of hybrids between its two species.

Research Program

Research on the correct classification of the Plagianthus alliance can take 4 forms.

  1. Production of a character matrix for the morphological traits of the species, and performance of a cladistic analysis.
  2. Production of chromosome counts for species of the alliance.
  3. An experimental hybridisation program, where the viability and fertility of crosses will shed light on relationships.
  4. Production of DNA sequence data for the alliance.

References

  1. Michael Heads, A new species of Hoheria (Malvaceae) from the Poor Knights Islands and Hen and Chickens Islands, New Zealand, New Zealand Journal of Botany 38: 373-377 (2000)
  2. Bentham and Hooker, Genera Plantarum: 195-240 (1865)
  3. Kubitski & Bayer, Malvaceae, in Kubitzki & Bayer, The Families and Genera of Flowering Plants, Vol. 5 (2003)
  4. Hutchinson, The Genera of Flowering Plants. 2: 536-567 (1967)
  5. Tate et al, Phylogenetic Relationships Within The Tribe Malveae (Malvaceae, subfamily Malvoideae) as Inferred from ITS Sequence Data, American Journal of Botany 92(4): 584–602 (2005)

Bibliography

  1. N.S. Lander, Revision of the Australian genus Lawrencia Hook, Nuytsia 5(2): 201-272 (1985)
  2. Melville, Generic delimitation in the Plagianthus complex, Kew Bulletin 20(3): 512-514 (1966)

Feedback

If you have found any errors on this page, or can provide any further information then please contact me at webmaster@meden.demon.co.uk.

© 2005 Stewart Robert Hinsley