Classification: Malveae

Malvaceae Info (Home)

The tribe Malveae is a division of subfamily Malvoideae of the angiosperm family Malvaceae. It appears to be a well defined group, characterised by absence of staminodes (which are reduced to apical teeth terminating the staminal column elsewhere in Malvoideae), by a schizocarpous fruit (except in Plagianthus, Bastardia and Bastardiopsis), and by the number of branches of the style equalling the number of carpels. (The capsular fruit of Plagianthus might be homologous to a mericarp in other genera of Malveae, arising by reduction of the number of mericarps to 1, rather than to the capsule of Hibisceae and Gossypieae. This is supported by the asymmetrical nature of the capsule.) Malveae has about 1000 species, in 71 currently recognised genera.

Historically the phylogeny of Malveae has been poorly known. Many traits have been used in classification, such as the base chromosome count, whether the schizocarp is capitate or verticillate, whether an epicalyx is present or absent, whether the stigmas are capitate or decurrent, whether the mericarps are dehiscent or indehiscent, whether the mericarps are single or multi-celled, or whether the mericarp cells are single or multi-seeded. The classification obtained depends on the weight given to the individual traits.

I had been investigating the phylogeny myself by combining published DNA sequences, for the internally transcribed spacers (ITS1 and ITS2) and 5.8s RNA loci of the nuclear ribosomal DNA,from the relatively large ITS (nuclear ribosomal DNA) data set for Malveae available at EMBL, arising from several independent studies, but this has been mostly superceded by a paper by Tate el al [3] combining several of those studies and adding additional sequences filling in most of the gaps in coverage. The description given here follows that paper.

Historically Malveae has been divided into several tribes. Three genera - Malope, Kitaibelia and Palaua - were placed in the tribe Malopeae, on the basis of bearing capitate, rather than verticillate, schizocarps. The remaining genera have sometimes been divided between 2 tribes – Malveae and Abutileae – or between 2 or 3 subtribes within Malveae sensu lato. (The subtribes would be Malvinae, Abutilinae and Sidinae, but can be found in older literature as (Eu)Malvae, Abutileae and Sideae.) Malvinae was distinguished from Abutilinae and Sidinae by the presence of decurrent as opposed to apical stigmas, and Abutilinae from Sidinae by the possession of multi-ovular carpels. Kearney introduced a subtribe Corynabutilinae for those genera with bilaterally decurrent stigmas (Corynabutilon and Neobaclea), in which he was followed by Hutchinson [4]. Asa Gray suggested a subtribe Plagianthinae for Plagianthus and related genera.

These divisions are not supported by DNA sequence data. Malopeae is nested within Malveae, and is polyphyletic. Abutilinae, with or without Sidinae, is also polyphyletic, or at best paraphyletic.

The Malveae is also divided informally into smaller groups of genera, or alliances. The alliances given in Kubitzki and Bayer [1], based on earlier work by Bates (1968), Bates and Blanchford (1970), and Fryxell (1988, 1997), and their contents, are as follows.

Abutilon alliance: Neobaclea, Corynabutilon, Tetrasida, Hochreutinera, Abutilon, Billieturnera, Pseudabutilon, Allowissadula, Wissadula, Bastardiastum, Bastardia, Herissantia, Bastardiopsis, Robinsonella, Akrosida, Dendrosida, Allosidastrum, Rhynchosida, Krapovickasia, Malvella, Meximalva, Sidastrum and Sida.
Batesimalva alliance: Batesimalva, Horsfordia, Briquetia, Fryxellia and Bakeridesia.
Malacothamnus alliance: Phymosia, Malacothamnus, Neobrittonia and Iliamna.
Kearnemalvastrum alliance: Kearnemalvastrum.
Malvastrum alliance: Malvastrum.
Sphaeralcea alliance: Sphaeralcea, Tarasa, Fuertesimalva, Calyculogygas, Monteiroa, Calyptraemalva, Napaea, Eremalche, Sidasodes, Acaulimalva, Palaua and Nototriche.
Modiola alliance: Modiola and Modiolastrum.
Gaya alliance: Gaya, Cristaria and Lecanophora.
Anoda alliance: Anoda and Periptera.
Malope alliance: Malope and Kitaibelia.
Anisodontea alliance: Anisodontea.
Malva alliance: Alcea, Althaea, Malva, Lavatera.
Sidalcea alliance: Callirhoe and Sidalcea.
Plagianthus alliance: Hoheria, Lawrencia, Plagianthus, Gynatrix and Asterotrichion.

Within the Malvaceae Pages I retain a division into subtribes Malvinae and Abutilinae, but with different bounds to the classical definitions. The Abutilinae contains the Abutilon, Gaya, Anoda, Plagianthus and Batesimalva alliances. The Malvinae contains the Malacothamnus, Malvastrum, Kearnemalvastrum, Sphaeralcea, Modiola, Malope, Anisodontea, Malva and Sidalcea alliances.

Abutilinae lack an epicalyx (except for Malvella), and have bilaterally decurrent (Corynabutilon, Neobaclea) or capitate stigmas. However these characteristics are insufficient to define the group, as secondary loss of the epicalyx has also occurred several times in Malvinae, and capitate stigmas is an ancestral trait also retained in many taxa in Malvinae. Pendulous ovules are also characteristic of the Abutilinae, but are also found in pluri-ovulate species of Anisodontea.

The Plagianthus alliance, combined with the South American genus Sidasodes and 2 species of Sida, forms the basal lineage in Abutlinae, and is the strongest candidate for recognnition as another sub-tribe. (Asa Gray suggested a sub-tribe Plagiantheae – Plagianthinae in modern orthography – to hold the genera of the Plagianthus alliance, and elsewhere the family Phillipodendreae (Phillipodendraceae) was introduced, presumably for these genera.) They differ in having the style divided for the greater part of its length, and hence also in having a short staminal column. However they agree with Abutilinae in lacking an epicalyx and in having pendulous ovules.

The following observations can be made from the work of Tate et al [3]

  1. The Batesimalva alliance is polyphyletic
  2. The Gaya alliance is polyphyletic, Lecanophora and Cristaria not being closely related to Gaya, the former forming the second basalmost lineage in Abutilinae, and the latter being nested within the Abutilon alliance, together with elements of the Batesimalva alliance.
  3. Sidasodes belongs to Abutilinae, rather than to the Sphaeralcea alliance, and is related to the Plagianthus alliance.
  4. The Abutilon alliance is paraphyletic with respect to Gaya and the Anoda and Batesimalva alliances.
  5. Neobrittonia belongs in Abutlinae, rather than in the Malacothamnus alliance.
  6. Sida and Dendrosida are paraphyletic. There is a well defined group consisting of species of Sida and Dendrosida, but other species of Sida fall into various locations in Abutilinae.
  7. Sidalcea and Callirhoe are not closely related. Their introrsely decurrent stigmas, and tendencies towards the loss of the epicalyx and to gynodioecy or dioecy, would seem to be homoplasious. I interpret the chromosome counts for Sidalcea as representing diploid, tetraploid and hexaploid members of a series with x=5, and those for Callirhoe as representing a tetraploid and octoploid members of a series with x=7, with derived x=15 species, rather than as hexaploid members of a x=5 series, with derived x=14 species. I therefore propose separate Sidalcea and Callirhoe alliances.
  8. The sister group to Sidalcea seems to be Eremalche (which is consistent with distributional and ecological data), and I therefore transfer this from the Sphaeralcea alliance to the Sidalcea alliance. (The shared x=5 base chromosome count in the Sidalcea and Sphaeralcea alliances appears to be convergent.)
  9. The sister group to Callirhoe seems to be Napaea. This is consisted with chromosome number data (I interpret the data for Napaea as x=15), with distributional data, and with Napaea's dioecious habit. I therefore transfer Napaea from the Sphaeralcea to the Callirhoe alliance.
  10. The Malva alliance is paraphyletic with respect to the Malope alliance. Alcea, Althaea are related to Kitaibelia, and Malva and Lavatera to Malope. These two groups may be sister to each, but this is weakly supported, and for example either could be more closely related to the Callirhoe or Anisodontea alliances. I prefer to split the Malva alliance into Malva and Althaea alliances.
  11. Sphaeralcea, Nototriche and Tarasa, form a clade. The former two genera are nested within the last.
  12. Malacothamnus chiliensis is not closely related to the remaining species of Malacothamnus, but instead to 3 species of Tarasa. All 4 species have a base chromosome count of 12 (other species of Tarasa have a count of 10). The genus Andeimalva [2] has been introduced for these 4 species. Introducing this genus resolves the paraphyly of Tarasa with respect to Sphaeralcea, but Nototriche remains nested within Tarasa.
  13. Urocarpidium albiflorum” nests with Fuertesimalva, rather than Tarasa, which would make Urocarpidium the correct name for this genus, rather than Fuertesimalva, unless it be divided into two genera. (Urocarpidium albiflorum” is relatively isolated compared with the remaining species of Fuertesimalva.)

The following other observations can be made.

  1. The traditional division of species between Lavatera and Malva is incorrect. This is discussed further on the page on the Malva alliance.
  2. Hybridisation observations suggest that section Hirsutae of Althaea belongs in the Malva, rather than the Althaea alliance. A poster at IOPB 2004 presents ITS data confirming this, which is discussed further on the page on the Malva alliance.

I propose the following revised set of alliances :-

Plagianthus alliance: Hoheria, Lawrencia, Plagianthus, Gynatrix, Asterotrichion, Sidasodes, and sections Hookerianae and Pseudonapaea of Sida.
Cristaria alliance: Cristaria and Leconophora.
Abutilon alliance: Neobaclea, Corynabutilon, Tetrasida, Hochreutinera, Abutilon, Billieturnera, Pseudabutilon, Allowissadula, Wissadula, Bastardiastum, Bastardia, Herissantia, Bastardiopsis, Robinsonella, Akrosida, Dendrosida, Allosidastrum, Rhynchosida, Krapovickasia, Malvella, Meximalva, Sidastrum, Sida, Batesimalva, Horsfordia, Briquetia, Fryxellia and Bakeridesia, Anoda, Periptera, Neobrittonia, Gaya, Bordasia and Spirabutilon.
Malva alliance: Malva, Lavatera and Malope
Althaea alliance: Althaea, Alcea and Kitaibelia
Callirhoe alliance: Callirhoe and Napaea
Sidalcea alliance: Sidalcea and Eremalche
Anisodontea alliance: Anisodontea.
Sphaeralcea alliance: Sphaeralcea, Tarasa, Andeimalva, Fuertesimalva, Calyculogygas, Monteiroa, Calyptraemalva, Acaulimalva, Palaua and Nototriche.
Modiola alliance: Modiola and Modiolastrum.
alliance: Phymosia, Malacothamnus and Iliamna.
Kearnemalvastrum alliance: Kearnemalvastrum.
Malvastrum alliance: Malvastrum.


  1. K. Kubitzki and C. Bayer, The Families and Genera of Vascular Plants, Vol. 5 (2003)
  2. Tate, Jennifer A., Andeimalva, a new genus of Malvaceae from Andean South America, Lundellia 6: 10-18 (2003)
  3. 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)
  4. Hutchinson, The Genera of Flowering Plants (1967)


  1. Bates, D.M., Generic Relationships in the Malvaceae, tribe Malveae. Gentes Herb. 10: 117-135 (1968)
  2. Bates, D.M., Blanchford, O.J., Jr., Chromosome Numbers in the Malvaceae. II. ..., Am. J. Bot 57: 927-934 (1970)
  3. Fryxell, P.A., Malvaceae of Mexico, Syst. Bot. Monogr. 25: 1-522 (1988)
  4. Fryxell, P.A., The American Genera of Malvaceae. II. Brittonia 49(2): 204-269 (1997)
  5. Krapovickas, Bordasia Krapov., nueva genero de Malvaceas, Bonplandia (Corrientes) 12(1-4): 133-135 (2003)

Malvaceae Info (Home)

© 2004, 2006, 2007, 2009 Stewart R. Hinsley