Malva (incl. Lavatera)
In the present treatment Malva and Lavatera are merged. The traditional generic distinction between Lavatera and Malva is based on fusion or non-fusion of the epicalyx but this character is very critical and turns out to be highly artificial and untenable (see for instance Ray 1995, Banfi & al. 2005). The relationships inferred from molecular data strongly contrast with this traditional classification (Escobar García & al. 2009). There are species in Lavatera (L. mauritanica Durieu and L. triloba L.) in which fusion of epicalyx segments differs between populations (Hinsley 2010). Some species traditionally included in Lavatera are more closely related to for instance Malva sylvestris than are some other species of Malva. These species of Lavatera – the so-called Malvoid group (Ray 1995) – were recently already transferred to Malva (Ray 1995, Ray 1998, Banfi & al. 2005, Molero Briones & Montserrat Martí 2005, Banfi & al. 2011). The remainder of the genus (the Lavateroid group) is still upheld today (Mabberley 2008), in part based on an alternative distinction of both genera on mericarp characters (i.e. malvoid mericarps with dorsal faces flat versus lavateroid mericarps with dorsal faces rounded; see Ray 1998, Escobar García & al. 2009). However, this separation is blurred by the existence of taxa with intermediate morphology (some Malva species possess fruits that are not assignable to any of the two types; Escobar García & al. l.c.). Both genera are therefore better merged or Lavatera further divided into four or up to 12 independent genera (Banfi & al. 2005, Escobar García & al. 2009, Hinsley 2010). Malva having priority, all species currently accommodated in Lavatera are better transferred to Malva (see also Stace 2010). A disadvantage of this transfer would be that Malva would become a morphologically very diverse and hardly diagnosable genus (Escobar García & al. 2009). However, the advantages of such a classification clearly prevail since Malva would then represent a single monophyletic group (Hinsley 2010). On exactly the same grounds Anderberg & al. (2007) recently merged (among others) Anagalis, Glaux and Trientalis and included them in a broadly circumscribed genus Lysimachia. See also: Verloove & Lambinon (2011).
Althaea was traditionally distinguished from Malva and related genera by six (or more) epicalyx segments. However, recent studies have shown that epicalyx characters are not informative at the generic level (Ray 1995, Davis 2010). Escobar García & al. (2009) demonstrated that the annual species of Althaea doubtlessly belong with the Malva-complex. Thus, Althaea hirsuta is here also included in Malva (see also Stace 2010). The perennial species (incl. native Althaea officinalis) are upheld in Althaea.
Previous segregates of Malva should, of course, also be merged. Some authors segregated, for instance, Bismalva Med. (see Slavík 1992), distinguished by having its flowers solitary in the leaf axils or in a congested terminal raceme (a character shared with numerous species of former Lavatera!). This distinction was obviously very artificial. In Belgium, two species were concerned: Bismalva alcea (L.) Med. (syn.: Malva alcea) and B. moschata (L.) Med. (syn.: M. moschata).
In this new circumscription, Malva counts ca. 35-40 species, some of which have become nearly cosmopolitan weeds. Four species are native in Belgium (or at least archaeophytic): Malva alcea L., M. moschata L., M. neglecta Wallr. and M. sylvestris L. At least three of these (M. alcea, M. moschata and M. sylvestris) are also grown for ornament and are regularly observed outside their native distribution range in Belgium. Especially Malva moschata is a common garden plant and often escapes (incl. white-flowered variants, cv ‘Alba’). A putative hybrid between Malva alcea and M. moschata (M. xintermedia Bor.) was recorded in 2008 in the verge of a maize field in Gutsenhoven (Heers). This nothotaxon most likely is of artificial origin and is probably never produced in the wild. It is more vigorous than either parents, intermediate in leaf shape and indumentum. Flowers occur from June till December but fruits are not produced (see also Radcliffe 2004). It is possibly overlooked. An aberrant, probably non-native form of Malva alcea is known since 2004 from a railway track in the port of Gent. It has all leaves shallowly lobed in stead of divided and probably is var. ribifolia (Viv.) Paol. Finally, alien variants of Malva sylvestris are also frequently encountered, especially var. mauritiana (L.) Thellung (syn.: M. mauritiana L., M. sylvestris subsp. mauritiana (L.) Boiss.). This taxon usually has wider and darker petals, multi-flowered inflorescences but hardly deserves taxonomic recognition.
Malva alcea var. ribifolia
1. Epicalyx segments 6-10 === Malva setigera
1. Epicalyx segments 3 === 2
2. Carpophore much enlarged in fruit, disc-like and entirely or partially concealing the mericarps === 3
2. Carpophore not enlarged in fruit, never disc-like and entirely or partially concealing the mericarps === 4
3. Carpophore only partially concealing the mericarps (distal part of the mericarps visible) === M. maroccana
3. Carpophore entirely concealing the mericarps (mericarps invisible) === M. trimestris
4. At least uppermost leaves usually very deeply divided. Petals bright pink or (more rarely) white, 20-35 mm long. Mericarps smooth or faintly reticulate. Lowermost flowers always solitary in the leaf axils (native) === Malva alcea, M. moschata
4. Leaves usually lobed but not deeply divided. Petals variously coloured, 4-30 mm long. Mericarps smooth to strongly reticulate. Lowermost flowers solitary in the leaf axils or in fascicles === 5
5. Most flowers solitary in the leaf axils. Epicalyx segments distinctly fused at base (ca. for half their length). Shrub-like or herbaceous perennial, often with woody branches, rarely annual or biennial === 6
5 Most flowers in fascicles in the leaf axils (rarely some solitary). Epicalyx segments free at base or indistinctly fused. Annuals, biennials or perennials but never shrub-like === 8
6. Herbaceous annual or biennial. Fruiting peduncle up to 15 cm long. Carpels 10-17 (usually ca. 14), transversely rugose. Stems usually flushed purple, sparsely covered with stellate hairs === M. punctata
6. Shrub-like or more or less herbaceous perennial. Fruiting peduncle much shorter, at most 8 cm long (usually much shorter). Carpels 20-23, smooth. Stem usually green, densely covered with stellate hairs === 7
7. Mericarps (if formed) slightly but distinctly pubescent. Shrub-like perennial with woody branches === M. xclementii
7. Mericarps glabrous. Herbaceous perennial === M. thuringiaca
8. Petals about as long as the sepals or only slightly longer (up to twice as long in M. verticillata but then with calyx distinctly closed at maturity, obscuring the mericarps) === 9
8. Petals obviously longer than sepals, usually 2-3x as long. Mericarps never obscured by closed calyx === 11
9. Calyx tightly enclosing the mericarps at maturity. Stem erect. Flowers in multi-flowered fascicles. Petals lilac. Leaf margin sometimes crispate. Fruiting pedicel at most twice as long as calyx === M. verticillata
9. Calyx open at maturity, leaving the mericarps exposed. Stem often procumbent, rarely erect. Flowers in few-flowered fascicles (usually 2-6). Petals whitish, pinkish, bluish or lilac. Leaf margin never crispate. Fruiting pedicels much longer than the calyx === 10
10. Calyx strongly accrescent at maturity. Fruiting pedicels usually less than 10 mm long. Sepals usually glabrous at margin (rarely with some hairs < 0,5 mm long). Angles of mericarps winged. Claw of petals glabrous === M. parviflora
10. Calyx not accrescent at maturity. Fruiting pedicels usually more than 10 mm long. Sepals long-ciliate at margin. Angles of mericarps sharp but not with a wavy wing. Claw of petals ciliate === M. pusilla
11. Petals 18-25 mm long, pinkish purple or lilac === 12
11. Petals 7-14 mm long, pale mauve or white === 13
12. Perennial. Indumentum sparse, hairs simple, bifid and stellate. Mericarps strongly reticulate. Petals bright pinkish-purple with distinct dark stripes, often more than 20 mm long. Epicalyx segments free at base (native) === M. sylvestris
12. Annual or biennial. Young branches with dense stellate hairs. Mericarps smooth to slightly reticulate. Petals dull lilac, with less distinct stripes, up to 20 mm long. Epicalyx segments fused at their very base === M. pseudolavatera
13. Mericarps strongly reticulate. Epicalyx segments usually ovate, not more than 3x as long as wide. Petals always pale lilac === M. nicaeensis
13. Mericarps smooth. Epicalyx segments linear, much more than 3x as long as wide. Petals white to pale lilac (native) === M. neglecta
References
Anderberg A.A., Manns U. & Källersjö M. (2007) Phylogeny and floral evolution of the Lysimachieae (Ericales, Myrsinaceae). Willdenowia 37: 407-421.
Banfi E., Galasso G. & Soldano A. (2005) Notes on systematics and taxonomy for the Italian vascular flora. 1. Atti Soc. It. Sci. Nat. Museo Civ. Stor. Nat., Milano 146(2): 219-244.
Banfi E., Galasso G. & Soldano A. (2011) Notes on systematics and taxonomy for the Italian vascular flora. 2. Atti Soc. It. Sci. Nat. Museo Civ. Stor. Nat., Milano 152(2): 85-106.
Davis C.J.S. (2010) Malva aethiopica, a new name for Lavatera abyssinica (Malvaceae): an endemic species of the Ethiopian Highlands. Phytotaxa 13: 15-18.
Escobar García P., Schönswetter P., Fuertes Aguillar J., Nieto Feliner G. & Schneeweiss G.M. (2009) Five molecular markers reveal extensive morphological homoplasy and reticulate evolution in the Malva alliance (Malvaceae). Molecular Phylogenetics and Evolution 50: 226-239.
Fernandes R.B. (1968) Contribuçoes para o conhecimento do género Lavatera L. I. Notas sobre algumas espécies. Collectanea Botanica 7: 393-448.
Hill S.R. (2015) Lavatera. In: Flora of North America Editorial Committee (eds.), Flora of North America North of Mexico, vol. 6. Oxford University Press, New York: 276-278. [available online at: http://www.efloras.org/florataxon.aspx?flora_id=1&taxon_id=117757]
Hill S.R. (2015) Malva. In: Flora of North America Editorial Committee (eds.), Flora of North America North of Mexico, vol. 6. Oxford University Press, New York: 286-293. [available online at: http://www.efloras.org/florataxon.aspx?flora_id=1&taxon_id=119571]
Hinsley S.R. (2010) Names in Malva. BSBI News 115: 51.
Mabberley D.J. (2008) Mabberley’s plant-book (3th ed.). Cambridge University Press, Cambridge: XVIII + 1021 p.
Molero Briones J. & Montserrat Martí J.M. (2005) Nomenclatura de algunas especies del género Malva Linnaeus (Malvaceae). Fontqueria 55: 285-292.
Radcliffe B. (2004) A hybrid mallow with garden potential. BSBI News 95: 54-57.
Ray M.F. (1995) Systematics of Lavatera and Malva (Malvaceae, Malveae) – a new perspective. Pl. Syst. Evol. 198: 29-53.
Ray M.F. (1998) New combinations in Malva (Malvaceae: Malveae). Novon 8(3): 288-295.
Slavík B. (1992) Malvaceae. In: Hejný S. & Slavík B. (eds.), Kvĕtena České Republiky, vol. 3. Academia, Praha: 282-316.
Stace C. (2010) New flora of the British Isles, 3th ed.: XXXII + 1232 p. Cambridge University Press.
Verloove F. & Lambinon J. (2011) The non-native vascular flora of Belgium: new combinations and a new variety. New Journ. Bot. 1(1): 38-42.