EXTANT SEED PLANTS
Plant woody, evergreen; nicotinic acid metabolised to trigonelline, (cyanogenesis via tyrosine pathway); primary cell walls rich in xyloglucans and/or glucomannans, 25-30% pectin [Type I walls]; lignins rich in guaiacyl units; true roots present, apex multicellular, xylem exarch, branching endogenous; arbuscular mycorrhizae +; shoot apical meristem multicellular, interface specific plasmodesmatal network; stem with ectophloic eustele, endodermis 0, xylem endarch, branching exogenous; vascular tissue in t.s. discontinuous by interfascicular regions; vascular cambium + [xylem ("wood") differentiating internally, phloem externally]; wood homoxylous, tracheids +; tracheid/tracheid pits circular, bordered; sieve tube/cell plastids with starch grains; phloem fibers +; stem cork cambium superficial, root cork cambium deep seated; nodes ?; stomata ?; leaf vascular bundles collateral; leaves megaphyllous [determinancy evolved first, then ad/abaxial symmetry], spiral, simple, axillary buds +[?], prophylls [including bracteoles] two, lateral, veins -5 mm/mm2 [mean for all non-angiosperms 1.8]; plant heterosporous, sporangia eusporangiate, on sporophylls, sporophylls aggregated in indeterminate cones/strobili; true pollen [microspores, i.e. no distal pore for release of gametes] +, grains mono[ana]sulcate, exine and intine homogeneous; ovules unitegmic, crassinucellate, megaspore tetrad tetrahedral, only one megaspore develops, megasporangium indehiscent; male gametophyte development first endo- then exosporic, tube developing from distal end of grain, to ca 2 mm from receptive surface to egg, gametes two, with cell walls, with many flagellae; female gametophyte endosporic, initially syncytial, walls then surrounding individual nuclei; seeds "large", first cell wall of zygote transverse, embryo straight, endoscopic [suspensor +], short-minute, with morphological dormancy, white, cotyledons 2; plastid transmission maternal; two copies of LEAFY gene, PHY gene duplication [N/O//A/C and P//BE lines], mitochondrial nad1 intron 2 and coxIIi3 intron present.
MAGNOLIOPHYTA
Plant woody, evergreen; lignans, O-methyl flavonols, dihydroflavonols, triterpenoid oleanane, non-hydrolysable tannins, quercetin and/or kaempferol +, apigenin and/or luteolin scattered, [cyanogenesis in ANITA grade?], lignins derived from both coniferyl and sinapyl alcohols, containing syringaldehyde [in positive Maüle reaction, syringyl:guaiacyl ratio less than 2-2.5:1], and hemicelluloses as xyloglucans; root apical meristem intermediate-open; root vascular tissue oligarch [di- to pentarch], lateral roots arise opposite or immediately to the side of [when diarch] xylem poles; origin of epidermis with no clear pattern [probably from inner layer of root cap], trichoblasts [differentiated root hair-forming cells] 0; stem with 2-layered tunica-corpus construction; wood fibers and wood parenchyma +; reaction wood ?, with gelatinous fibres; starch grains simple; primary cell wall mostly with pectic polysaccharides; tracheids +; sieve tubes eunucleate, with a sieve plate and cytoplasm with P-proteins, companion cells from same mother cell that gave rise to the sieve tube; nodes unilacunar [1:?]; stomata with ends of guard cells level with pore, paracytic, outer stomatal ledges producing vestibule; leaves with petiole and lamina [the latter formed from the primordial leaf apex], development of venation acropetal, 2ndary veins pinnate, fine venation reticulate, veins (1.7-)4.1(-5.7) mm/mm2, endings free; flowers perfect, polysymmetric, parts spiral [esp. the A], free, development in general centripetal, numbers unstable; P not sharply differentiated, outer members not enclosing the rest of the bud, smaller than inner members; A many, with a single trace, introrse, filaments stout, anther ± embedded in the filament, tetrasporangiate, dithecal, with at least outer secondary parietal cells dividing, each theca dehiscing longitudinally by action of hypodermal endothecium, endothecial cells elongated at right angles to long axis of anther; tapetum glandular, binucleate; microspore mother cells in a block, microsporogenesis successive, walls developing by centripetal furrowing; pollen subspherical, binucleate at dispersal, trinucleate eventually, tectum continuous or microperforate, ektexine columellar, endexine thin, compact, lamellate only in the apertural regions; nectary 0; G free, several, ascidiate, with postgenital occlusion by secretion, few [?1] ovules/carpel, ovules marginal, anatropous, bitegmic, [outer integument often largely subdermal in origin, inner integument dermal], micropyle endostomal, integuments 2-3 cells thick, nucellus at apex of ovule 1-3 cells thick, megasporocyte single, megaspore lacking sporopollenin and cuticle, chalazal, female gametophyte four-celled [one-modular, nucleus of egg cell sister to one of the polar nuclei], stylulus short, hollow, cavity not lined by distinct epidermal layer, stigma ± decurrent, dry [not secretory]; P deciduous in fruit; seed exotestal; pollen germinating in less than 3 hours, siphonogamy, tube elongated, growing at 80-600 µm/hour, with callose plugs and callose-based walls, penetrating between cells, penetration of ovules within ca 18 hours, distance to first ovule 1.1.-2.1 mm; tube moves between nucellar cells, double fertilisation +, endosperm diploid, cellular [first division oblique, micropylar end initially with a single large cell, chalazal end more actively dividing], copious, oily and/or proteinaceous, embryo cellular ab initio, minute; germination hypogeal, seedlings/young plants sympodial; Arabidopsis-type telomeres [(TTTAGGG)n]; whole genome duplication, single copy of LEAFY and RPB2 gene, knox genes extensively duplicated [A1-A4], AP1/FUL gene, paleo AP3 and PI genes [paralogous B-class genes] +, with "DEAER" motif, SEP3/LOFSEP and PHYA + C/PHYB + E gene pairs.
Evolution. Possible apomorphies for flowering plants are in bold. Note that the actual level to which many of these features, particularly the more cryptic ones, should be assigned is unclear, because some taxa basal to the [magnoliid + monocot + eudicot] group have been surprisingly little studied, there is considerable variation between families in particular for several of these characters, and also because details of relationships among gymnosperms will affect the level at which some of these characters are pegged. For example, if reticulate-perforate pollen is optimized to the next node on the tree (see Friis et al. 2009 for a discussion), it effectively makes the pollen morphology of the common ancestor of all angiosperms ambiguous... For other features such a a nucellus only one (Nymphaeales) to three cells thick above the embryo sac and a stylar canal lacking an epidermal layer, although plesiomorphous for basal grade angiosperms (Williams 2009), where on the tree a thicker nucellus and a stylar epidermal layer are acquired has not yet been indicated.
NYMPHAEALES [AUSTROBAILEYALES [[CHLORANTHALES + MAGNOLIIDS] [MONOCOTS [CERATOPHYLLALES + EUDICOTS]]]]: vessels + [one position], elements with elongated scalariform perforation plates; axial parenchyma diffuse or diffuse-in-aggregate; tectum reticulate-perforate [here?]; ?genome duplication; "DEAER" motif in AP3 and PI genes lost, gaps in these genes.
AUSTROBAILEYALES [[CHLORANTHALES + MAGNOLIIDS] [MONOCOTS [CERATOPHYLLALES + EUDICOTS]]]: ethereal oils in spherical idioblasts [lamina and P ± pellucid-punctate]; tension wood 0; tectum reticulate-perforate [here?], nucellar cap + [character lost where in eudicots?]; 12BP [4 amino acids] deletion in P1 gene.
[CHLORANTHALES + MAGNOLIIDS] [MONOCOTS [CERATOPHYLLALES + EUDICOTS]] : benzylisoquinoline alkaloids +; P more or less whorled, 3-merous [possible position], carpels plicate; embryo sac bipolar, 8 nucleate, antipodal cells persisting; endosperm triploid; ?germination.
MONOCOTS [CERATOPHYLLALES + EUDICOTS]: (veins in lamina often 7-17mm/mm2 or more [mean for eudicots 8.0]; stamens opposite [two whorls of] P; pollen tube growth fast).
[CERATOPHYLLALES + EUDICOTS]: ethereal oils 0.
EUDICOTS: myricetin, delphinidin scattered, asarone 0 [unknown in some groups, + in some asterids]; root epidermis derived from root cap [?Buxaceae, etc.]; nodes 3:3; stomata anomocytic; flowers (dimerous), cyclic; K/outer P members with three traces, "C" with a single trace; A few, (polyandry widespread, from few initial [5, 10, ring] primordia), filaments fairly slender, anthers basifixed; microsporogenesis simultaneous, microspore walls developing by centripetal furrowing; pollen with endexine, tricolpate; G with complete postgenital fusion, stylulus/style solid [?here]; seed coat?
[[SABIACEAE + PROTEALES] [TROCHODENDRALES [BUXALES + CORE EUDICOTS]]]: (axial/receptacular nectary +).
TROCHODENDRALES [BUXALES + CORE EUDICOTS]: benzylisoquinoline alkaloids 0; euAP3 + TM6 genes [duplication of paleoAP3 gene: B class], mitochondrial rps2 gene lost.
BUXALES + CORE EUDICOTS: ?
CORE EUDICOTS: ellagic and gallic acids common; micropyle?; PI-dB motif +, small deletion in the 18S ribosomal DNA common.
ROSIDS ET AL. + ASTERIDS ET AL.: root apical meristem closed; (cyanogenesis also via [iso]leucine, valine and phenylalanine pathways); flowers rather stereotyped: 5-merous, parts whorled; calyx and corolla distinct, the former enclosing the flower in bud [with three or more traces, both bracteal in origin?]; stamens = 2x K/C, in two whorls developing internally/adaxially to the corolla whorl and successively alternating, (numerous, but then usually fasciculate and/or centrifugal); pollen tricolporate; [G 5], [3] also common, compitum +, placentation axile, style +, stigma not decurrent; endosperm nuclear; fruit dry, dehiscent, loculicidal [when a capsule]; euAP1 + euFUL + AGL79 genes [duplication of AP1/FUL or FUL-like gene], PLE + euAG [duplication of AG-like gene: C class], SEP1 + FBP6 genes [duplication of AGL2/3/4 gene]; RNase-based gametophytic incompatibility system present.
ROSIDS ET AL. = DILLENIALES [SAXIFRAGALES [VITALES + ROSIDS]]: nodes 3:3; stipules + [usually apparently inserted on the stem].
SAXIFRAGALES [VITALES + ROSIDS] = ROSANAE Takhtajan: ??
VITALES + ROSIDS: anthers articulated [± dorsifixed, transition to filament narrow, connective thin].
ROSIDS: embryo long; genome duplication; chloroplast infA gene defunct, mitochondrial coxII.i3 intron 0.
MALVIDAE = [[GERANIALES + MYRTALES] [CROSSOSOMATALES [PICRAMNIALES [SAPINDALES [HUERTEALES [MALVALES + BRASSICALES]]]]]]: ?
CROSSOSOMATALES [PICRAMNIALES [SAPINDALES [HUERTEALES [MALVALES + BRASSICALES]]]]: ?
PICRAMNIALES [SAPINDALES [HUERTEALES [MALVALES + BRASSICALES]]]: 2 apical pendulous ovules/carpel.
SAPINDALES [HUERTEALES [MALVALES + BRASSICALES]]: flavonols +; vessel elements with simple perforation plates; (cambium storied); petiole bundle(s) annular; inner integument thicker than outer, style +; endosperm scanty.
HUERTEALES [MALVALES + BRASSICALES]: ?
Evolution. The age of the crown group was estimated as (94-)89(-85) or (80-)74(-68) million years; Bayesian relaxed clock estimates were slightly different, to 96 million years (Wang et al. 2009), while Magallón and Castillo (2009) estimated ages of ca 91.9 and 92.1 million years for relaxed and constrained penalized likelihood datings for the age of the crown group, and ages of 105.2 and 105.6 million years for the stem.
Phylogeny. These two orders were linked in a whole chloroplast genome analysis by S.-B. Lee et al. 2006), although no chloroplast genomes of other rosid II taxa were then available. This position is supported in the more recet analyses of Worberg et al. (2009).
MALVALES Dumortier Main Tree, Synapomorphies.
(Cyclopropenoid fatty acids +), flavones, myricetin +; mucilage cells +; C contorted, disc 0, few ovules/carpel, style long; exotegmen much thickened and lignified, palisade. - 10 families, 338 genera, 6005 species.
Evolution. Wikström et al. (2001: relationships are [Brassicales [Malvales + Sapindales]]]) date the origin of stem Malvales to (84-)80(-76) million years before present, diversification beginning (71-)67(-63) million years before present. The age of crown group Malvales was estimated as (80-)78(-76) and (76-)74(-72) million years (two penalized likelihood dates), the stem group age being (93-)89(-85) or (78-)74(-70) million years; Bayesian relaxed clock estimates were slightly different, to 88 and 96 million years respectively (Wang et al. 2009), while Magallón and Castillo (2009) estimated ages of ca 91.9 and 92.1 million years for relaxed and constrained penalized likelihood datings for the age of stem Malvales, and ages of a mere 33.9 million years for the crown group.
Malvales contain ca 3.2% eudicot diversity (Magallón 1989). Timing the separation of clades in this group is of considerable interest given the distributions of many of the families; Ducousso et al. (2004) suggest that Dipterocarpaceae and Sarcolaenaceae had a common ancestor some 88 million years before present prior to the split of India and Madagascar, but unfortunately relationships in this area are unclear.
Taxa with ectomycorrhizal associations are common in this order, e.g. Tilioideae and ex Sterculiaceae, Dipterocarpaceae and Cistaceae (Smith and Read 1997), Sarcolaenaceae (Ducousso et al. 2004); it will be interesting to see just how widely distributed such mycorrhizae are.
Chemistry, Morphology, etc. For the distribution of cyclopropenoid fatty acids, which are also scattered outside Malvales, see Badami and Patil (1981), Gaydou and Ramanoelina (1983) and Bayer et al. (1999). However, as Kubitzki and Chase (2002: Table 1) show, the evolution of this and other features common in the order are difficult to understand. Species with stratified phloem always have wedge-shaped phloem rays, Sarcolaenaceae perhaps excepted (Kubitzki & Chase 2002). The androecium is possibly basically oligomerous, and the earliest initiated or innermost members are oppositisepalous with centrifugal or lateral polyandry. Starchy endosperm may be an apomorphy for the group.
General information, esp. carpel orientation, is taken from from Nandi (1998b) and Kubitzki and Chase (2002); for wood anatomy, see i.a. den Outer and Vooren (1980) and den Outer and Schütz (1981).
Phylogeny Several clades within Malvales are quite well established, but relationships between them, as well as the position of one or two families, remain unclear. Fay et al. (1998a) and Bayer et al. (1999) discuss general relationships. These may be represented as Muntingiaceae [[[Thymelaeaceae + Sphaerosepalaceae] [Neuradaceae [Sarcolaenaceae [Dipterocarpaceae + Cistaceae]]]] [Bixaceae + Malvaceae]], but few of these relationships have much support, even after successive weighting. Dayanandan et al. (1999) and Morton et al. (1999) relationships within Dipterocarpaceae, and Alverson et al. (1998), Bayer et al. (1999) and especially Alverson et al. (1999) those within Malvaceae s.l. Neuradaceae may be sister to the rest of the order (see also Soltis et al. 2007a). Although Bixaceae are expanded here, it has also been suggested that Cochlospermum and relatives are also close to Diegodendron and Sphaerosepalaceae, less to Bixa (Johnson-Fulton & Watson 2008). The relationships of Sphaerosepalaceae within Malvales are unclear; in Bayer et al. (1999) they are weakly associated with Thymelaeaceae and in Alverson et al. (1998) with Bixaceae and Cochlospermaceae. The relationships between Sarcolaenaceae, Dipterocarpaceae, and Cistaceae are also uncertain (Ducoussu et al. 2004).
Relationships of Cytinaceae with Malvales have been suggested (Nickrent 2002), and these appeared in all analyses in a more recent study (Nickrent et al. 2004). However, the only Malvales included in Nickrent et al. (2004) were Cistaceae and Malvaceae, and the placement of Cytinaceae remained somewhat provisional. Recently Nickrent (2007), with much better sampling, found that Cytinaceae are sister to the poorly-known Muntingiaceae with moderate (maximum likelihood) to strong (maximum parsimony) support (nuclear small-subunit [SSU] r-DNA was the nuclear gene used in the placement). Both Cytinaceae and other Malvales have exotegmic seeds, and aspects of the perianth of Cytinaceae and Malvaceae can perhaps be compared. The mitochondrial genes cox1 and matR showed considerable divergence, but not the atp1 gene (Barkman et al. 2007). Apodanthaceae, here included in Cucurbitales, with which they show gynoecial similarities in particular, both having a connate androecium and extrose anthers, inferior ovary, parietal placentation, etc., are also somewhat similar morphologically to Malvales (Nickrent et al. 2004).
Classification. Elaeocarpaceae, previously usually included in (Cronquist 1981) or near Malvales, are placed unambiguously - if unexpectedly - in Oxalidales. Most Malvales as delimited here are included in Takhtajan's (1997) Malvanae; the core of Malvales in the past was the families here all included in Malvaceae.
Includes Bixaceae, Cistaceae, Cytinaceae, Dipterocarpaceae, Malvaceae, Muntingiaceae, Neuradaceae, Sarcolaenaceae, Sphaerosepalaceae, Thymelaeaceae.
Synonymy: Aquilariales Link, Bixales Lindley, Bombacales Kunth, Byttneriales link, Cistales Reichenbach, Cytinales Dumortier, Daphnales Lindley, Dipterocarpales Blume, Neuradales Doweld, Sterculiales Berchtold & J. Presl, Thymelaeales Wilkomm, Tiliales Caruel - Malvidae Thorne & Reveal - Malvanae Takhtajan - Cistopsida Bartling, Daphnopsida Meisner, Malvopsida R. Brown, Thymelaeopsida Endlicher
NEURADACEAE Link, nom. cons. Back to Malvales
Annual (perennial) herbs/subshrubs; cyclopropenoid fatty acids +, ellagic acid?, tannins?; cork?; cambium storying?; pits not bordered; sieve tube plastids with protein crystalloids and starch; nodes ?; petiole anatomy simple; cuticle waxes 0; hairs unicellular; leaves spiral, toothed to pinnatifid, 2ndary veins subpalmate, stipule ?; inflorescence modified cymose; hypanthium +, short; K valvate, C also imbricate, distinctively coloured when dry; A 10; pollen syncolpate; G [5-10], ± inferior, opposite petals, 2-4 adaxial carpels infertile, 1-2 apotropous ovules/carpel, styluli +, stigmas capitate; ?micropyle; fruit dry, indehiscent, K accrescent, styles persistent, forming spines or not; exotegmic cells also tangentially elongated, crystalliferous, other tegmic cells persistent; endosperm ?development, 0, embryo bent; n = 7.
3[list]/10: Grielum (5). Africa to India, dry or desert areas. [Photo - Flower]
Chemistry, Morphology, etc. The plant probably lacks stipules, the stipule-like structures that are sometimes seen in fact being a prophyll that develops later than the foliage leaf (Bayer 2002); there are indeed structures that look very like stipules, so presence/absence of stipules should be confirmed, and this will be of relevance to understanding nodal anatomy. There appears to be no epicalyx, although the outside of the ovary may have spines which become conspicuous in fruit. The corolla of members of Neuradaceae changes colour on drying, as in some Malvaceae (Airy Shaw 1966) - see Huber (1993a). Goldberg (1986) notes there may be two ovules/carpel; both he and Takhtajan (1997) describe the latter as dehiscing ventrally.
For general information, see Bayer (2002); the family needs work. I am grateful to Z. Rogers for comments.
Previous Relationships. Neuradaceae have previously placed been elsewhere, as in Rosales in Cronquist (1981) and Takhtajan (1997), Hutchinson (1973) even included it in the Rosaceae, and the floral anatomy of the two is similar (Ronse Decraene & Smets 1996b).
Synonymy: Grielaceae Martynov
Thymelaeaceae, Sphaerosepalaceae, Bixaceae, [Cistaceae + Sarcolaenaceae + Dipterocarpaceae], Muntingiaceae, Malvaceae: pits vestured; phloem stratified, phloem rays wedge-shaped.
THYMELAEACEAE Jussieu, nom. cons. Back to Malvales
Cyclopropenoid fatty acids +; wood often fluoresces; vascular tracheids +; secondary phloem fibers unlignified; nodes 1:1; pericyclic fibers 0 [Edgeworthia, Tepuianthus]; crystal sand +; petiole bundle arcuate; epidermal cells (massively) mucilaginous; stomata cyclocytic; hairs simple; leaves spiral, supervolute (conduplicate), stipules 0 or minute; inflorescence cymose; flowers (3-)4-5(-6)-merous; K and C imbricate; pollen trinucleate; 1 epitropous ovule/carpel, ovules with endostomal micropyle, antipodal cells persist, many, obturator from near base of stylar canal; (testa fleshy), exotegmen with brown contents, endotegmen with brown contents, reticulately thickened and lignified; embryo white, cotyledons large.

46-50[list]/891 - 3 groups below. World-wide, esp. trop. Africa and Australia (map: from Domke 1934; Meusel et al. 1978). [Photo - Flower Flower, Fruit.]
1. Tepuianthus
Trees or shrubs, bark bitter; phloem rays narrow; resiniferous cells +; plant androdioecious; flowers small, K and C free, C ± clawed; 5-10 glandular scales; A 5, opposite sepals, or 12-22, in groups opposite petals[?], connective produced or not; pollen colp(or)ate; G [(2)3], styles separate, bifid [parastyles?]; seeds with an angled raphe; endosperm slight[?]; n = ?
1/7. Guayana highlands, northeast South America.
Synonymy: Tepuianthaceae Maguire & Steyermark
Octolepidoideae + Thymelaeoideae: C 0, pollen oligo- to polyporate, minutely spinulose, style single, stigma ± capitate, dry; endotegmen with stripes on the inner surface.
2. Octolepidoideae Gilg
Trees, shrubs or lianes; secretory cavities + [leaves punctate]; (stomata anomocytic - Octolepis); hypanthium +, at most short, or 0; K (3-)5(6), valvate or imbricate; glandular scales 4-40; A 8-80, not adnate to tube, (connate), anthers usu. recurved and hippocrepiform, connective well developed, (basal layer with pendulous internal processes); G (2-)3-5(-8), (clavate or subglobose parastyles +; stigma punctate); seed with a raphal aril, or angled at the raphe, or funicle swollen; chalazal fold on ventral side only [distribution elsewhere in family?], nucellar tracheids +, (tegmen multiplicative); endosperm copious (or not?); n = ?
8/49: Gonystylus (20). Tropical Africa and Madagascar (Octolepis), Malesia to Australia (Queensland), New Caledonia, Fiji.
Synonymy: Gonystylaceae van Tieghem, nom. cons.
3. Thymelaeoideae Burnett
Trees, shrubs, lianes or herbs; phorbol ester diterpenes [largely orthoesters and 1-alkyldaphnane derivates], cyclopropenoid fatty acids, chelidonic acid +, myricetin, tannins 0; internal phloem +; vascular bundles bicollateral; (stomata anomocytic); leaves often opposite; inflorescence often capitate; flowers 4-5-merous, "hypanthium" long (0), petaloid appendages to 2 x P or 0, A 2-5, opposite (alternate with) P, or 10; pollen crotonoid; nectary disc +, morphology various, (long-tubular), or 0; G 2, one locule often not developed and style then excentric; fruit a drupe or achene; (seed with a chalazal fold, and/or caruncle +); (nucellar tracheids +; testal cells enlarged; palisade exotegmen 0; tegmen multiplicative); endosperm 0 (quite copious - e.g. Daphne, Lachnaea); n = (7-)9(10), much polyploidy.
37/690: Gnidia (140), Pimelea (110), Daphnopsis (55), Daphne (ca 100), Lachnaea (40). World-wide, esp. trop. Africa and Australia.
Synonymy: Daphnaceae Ventenat, Phaleriaceae Meisner
Economic Importance. A number of taxa scattered through the family produce gaharu or agarwood. This is developed from the heartwood often after wounding and perhaps also infection by fungi; gaharu is much esteemed as source of incense and medecines, and some species that produce it have been decimated in the wild (Eurlings & Gravendeel 2005).
Chemistry, Morphology, etc. Microsemma (= Lethedon, in Gonystyloideae) has cyclopentenoid cyanogenic glycosides (Spencer & Seigler (1985. Other aspects of the chemistry of Thymelaeaceae, including the presence of phorbol ester diterpenes, are similar to the chemistry of Euphorbiaceae (Seigler 1994).
At least some Thymelaeoideae have a lignified, torus-bearing, pit membrane (Coleman et al. 2004; Dute et al. 2010). The petiole anatomy of Gonystylus needs to be confirmed; the bundle is perhaps unlikely to be arcuate.
Floral morphology in the family is poorly understood. The vasculature of the perianth in Thymelaeaceae was studied by Heinig (1951). The vascular bundles supplying the structures inside the calyx, whether paired and more or less opposite the sepals or single and in the petaline position, came from lateral branches of the sepal vasculature; equating these structures with stipules seems unlikely. Dicranolepis (Octolepidoideae: Africa) has large "petals" that are variable in number but paired and opposite the petals, and they are sometimes serrate or laciniate; the anther connectives are massive. Other taxa have single structures alternating with the lobes of the perianth tube, while in Lachnaea there are paired structures in the perianth tube borne below the instertion of the two whorls of stamens (Herber 2002b). Here I have been conservative, calling the major tubular structure is a perianth, I am agnostic about the occurence of petals and also whether a hypanthium in the strict sense is present or not. Synandrodaphne lacks a floral tube. The pollen of many Thymelaeaceae-Thymelaeoideae is similar to that of Euphorbiaceae-Crotonoideae. The micropyle of Gnidia is zig-zag. Eckardt (1937) discussed gynoecial variation in members of Thymelaeoideae.
The testa of Tepuianthus is about 6 cells thick and unlignified, then there is a layer of lignified palisade cells, the exotegmen, and immediately underneath apparently a layer of low, lignified cells, i.e., the seed coat is similar to that of other members of the family. Reports of a small embryo (Maguire & Steyermark 1981) need to be confirmed. The base of the lamina joins the petiole on the adaxial side, and so the lamina is almost peltate. Illustrations in Maguire and Steyermark (1981) suggest that there are colleters at the base of the calyx. See Roth and Lindorf (1990) for anatomy of the genus.
Some information is taken from Guérin (1916: ovule and seed), Domke (1934: general), Ding Hou (1960: general), Evans and Taylor (1983: phorbol esters), Weberling and Herkommer (1989: inflorescence morphology), Kubitzki (2002: Tepuianthaceae), Herber (2002a: palynology, 2002b: general), and Horn and Wurdack (ms.: general).
Phylogeny. For the phylogeny of the family, see van der Bank et al. (2002). They found the following set of relationships [Gilgiodaphne (= Synandrodaphne), Gonystyloideae [Aquilarioideae + Thymelaeoideae]]. These relationships, other than the position of Gilgiodaphne, were well supported: genera like Tepuia and Octolepis were not included, while Gnidia was highly polyphyletic (see also the much more detailed study of Beaumont et al. 2009). Savolainen et al. (2000b) placed Ploiarium (here in Bonnetiaceae - Malpighiales) within Thymelaeaceae. Morphologically and anatomically this position would seem rather unlikely.
The inclusion of Tepuianthus in this clade makes eminent morphological sense (Wurdack & Horn 2001; Horn & Wurdack, ms.). It has a number of apomorphies of other Thymelaeaceae, while the features in which it differs from them are mostly plesiomorphies, i.e., they are similarities to other Malvales. Tepuianthus has both a well-developed calyx and corolla and also scales outside the androecium, perhaps suggesting that the corolla scales of Gonystylus, and perhaps those of the rest of the family, are homologous with these glandular scales. Furthermore, although it is described as having three separate styles, the stigmas being bilobed or not, these "styles" may be similar to similar processes on top of the ovary in Gonystyloideae, which are called parastyles and are associated with the styles proper. Distinctive epidermal columns in the palisade mesophyll of the leaf of Tepuianthus are found in other Thymelaeaceae such as Solmsia, and its resin cavities may be compared with the secretory cells of Octolepidoideae. The bark of Tepuianthus is described as being bitter, while Thymelaeoideae are well known for often being rather poisonous, unfortunately, the chemistry of Octolepidoideae is poorly known. Finally, the well developed parallel venation of Tepuianthus is very like that of other Thymelaeaceae, and Solmsia (New Caledonia: Octolepidoideae) is vegetatively remarkably similar to Tepuianthus down to details of the base and mucronate apex of the lamina.
Classification. On hold at present. The old Aquilarioideae are monogeneric and where the monogeneric Gilgiodaphnoideae (van der Bank et al. 2002) are to go is unclear. Daphne is probably to include Wikstroemia (Halda 2001); many generic limits will need reconsideration because of the fragmentation of Gnidia, yet Gnidia s.l. is not necessarily "maximally stable" given the poor support for the clade it represents (cf. Beaumont et al. 2009: 413). For general information, see Zachary Rogers's A World Checklist of Thymelaeaceae (2009 onwards). I am grateful to Z. Rogers for comments.
Previous Relationships. The pollen of many Thymelaeaceae-Thymelaeoideae is similar to that of Euphorbiaceae-Crotonoideae, and the chemistry is also similar to that of Euphorbiaceae, including the presence of phorbol ester diterpenes (Seigler 1994); Takhtajan (1997) places his Thymelaeales immediately after Euphorbiales. Microsemma (= Lethedon - Gonystyloideae) has cyclopentenoid cyanogenic glycosides; Spencer and Seigler (1985) suggested that because of this, it should be placed in Flacourtiaceae (see Achariaceae here). Thymelaeaceae were included in Myrtales by Cronquist (1981).
Sphaerosepalaceae, Bixaceae, [Cistaceae + Sarcolaenaceae + Dipterocarpaceae], Muntingiaceae, Malvaceae: hairs often stellate; stipules often well developed; A many, developing centrifugally, from 5 or 10 (15) bundles, when 5 often opposite the petals, several [³6] ovules/carpel, micropyle bistomal.
SPHAEROSEPALACEAE Bullock Back to Malvales
Deciduous trees; ellagic acid +; cambium storying ?0; pits not vestured; true tracheids +; calcium oxalate crystals +; rays uniseriate; secretory canals +; resin-filled cells outside veins; petiole bundles cylindrical (with adaxial plate; medullary bundles +); (stomata cyclocytic); hairs simple; leaves spiral or two-ranked, conduplicate, (2ndary veins palmate; fine venation closely raised), stipule intrapetiolar, broad [± encircling stem], deciduous, petiole pulvinate; inflorescences with subumbelliform cymules; flowers usu. 4-merous, K usu. 2 + 2, caducous, outer median (persistent), inner larger, C (3-)4(-9), when 4 opposite sepals, clawed, aestivation various, with many short resinous lines, caducous; A with broad connective, pollen usu. ± spinose, pollen with endoapertures larger than ectoapertures, short gynophore +, nectary on top, G [2(-5)], placentation basal, 2-9 epitropous ovules/carpel, micropyle endostomal, style continuous to gynobasic, stigma punctate or obscurely lobed; fruit ± baccate, muricate to finely verrucose, ± deeply lobed; G separate, or G on one side not developed, 2-9 ovules/carpel, 1 seed/carpel; seeds with funicular aril/0, ruminate or not, testa 6-20 cells across, exotesta mucilaginous, exotegmen conspicuously incurved on either side of hypostase (not), operculum +; endosperm ?development, moderate, ruminate or not, starchy, cotyledons cordate and bilobed apically; n = ?

2[list]/18. Madagascar. [Photos - Collection]
Chemistry, Morphology, etc. Secretory cavities are widespread, and the carpels produce an exudate when cut. The rays are not storied (den Outer & Schütz 1981), and Jansen et al. (2000a) did not find vestured pits (cf. den Outer & Schütz 1981). Takhtajan (1997) described the stipules as being extrapetiolar and the endosperm as being copious. The lateral sepal bundles are commissural, as in Thymelaeaceae. There are androecial trunk bundles opposite the petals. The apparently terminal style may be modified from the gynobasic condition (Horn 2004).
For more information, see Capuron (1962), Huard (1965), Bayer (2002) and Horn (2004).
Synonymy: Rhopalocarpaceae Takhtajan
Phylogeny. [Bixaceae [Cistaceae + Sarcolaenaceae + Dipterocarpaceae]] form a group morphologically united as follows: plant with secretory canals; K imbricate; seed with bixoid plug, the exotegmen curved inwards in chalazal region, hypostase plug with core and annulus. This bixoid chalazal plug forms a water gap through which water enters the hard seeds with such plaugs, so causing the breaking of the physical dormancy of the such seeds (Baskin et al. 2000). Bixaceae + Cistaceae may form another group: leaf teeth with a single vein proceeding to opaque deciduous apex; embryo long, cotyledons thin, curved or folded, radicle short, stout. Molecular phylogenies are largely silent about the first grouping, but suggest that the second is incorrect.
BIXACEAE Kunth, nom. cons. Back to Malvales
Plant with secretory canals; resin-filled cells outside veins; hairs glandular, not tufted or stellate; leaf teeth with a single vein proceeding to opaque deciduous apex; inflorescence terminal, flowers large; K imbricate, anthers porose, many ovules/carpel, funicles long, micropyle zig-zag, stigma at most slightly lobed; exotegmen curved inwards in chalazal region, hypostase plug with core and annulus, outer hypostase forming the core; embryo long, cotyledons spatulate, thin, curved or folded, radicle short, stout.
4/21. Pantropical. Three groups below.
Trees to ± herbs; ellagic acid +; gums +; cork ?; young stem with continuous vascular ring; sieve tube plastids with protein crystalloids and starch; petiole bundles 3, annular; glandular hairs when young; leaves spiral, palmately lobed, conduplicate, margins toothed or entire, stipules narrow; bracteoles 0; flowers ± obscurely monosymmetric, C contorted, G [3-5], opposite petals or odd member adaxial, (placentation parietal); fruit a septicidal capsule, endocarp ± separating from the mesocarp; seeds hairy, tegmen with enlarged outer hypodermal cells; endosperm oily, embryo curved, ?color, cotyledons spathulate; n = 6.
2/15: Cochlospermum (12). Pantropical (map: from Poppendieck 1980, 1981, Cochlospermum religiosum widely naturalised from Java to India). [Photo - Habit, Flower.]
Synonymy: Cochlospermaceae Planchon, nom. cons.
Bixa + Diegodendron: glandular hairs peltate; petiole bundle cylindrical, with medullary strands; stipules ensheathing bud; G [2-4]; fruit muricate.

2. Bixa
Trees or shrubs; flavones, flavonols, ellagic acid, flavonoid sulphates +, proanthocyanins 0; leaves spiral, often palmate to palmately lobed, conduplicate, margins toothed to entire, 2ndary veins palmate; (flowers vertically monosymmetric); K with basal abaxial glands, C imbricate; A with trunk bundles, folded horizontally; G [2-4], opposite sepals, placentation parietal, inner integument 4-5 cells across; testa pulpy, endotegmen with ± thickened cells, hypodermal layer of hour-glass cells; endosperm starchy, embryo white; n = 7.
1/5. Tropical America. [Photo - Flower, Fruit, Fruits.]

3. Diegodendron
Evergreen tree; cork?; nodes ?; mucilage cells?; (stomata cyclocytic); leaves two-ranked, punctate, stipules intrapetiolar, deciduous; K unequal, C caducous, anthers with slits, disc ?inconspicuous; G [2(-4)], orientation ?, 2 basal epitropous ovules/carpel, ?micropyle, ?funicle, style gynobasic; fruit with small glands, indehiscent; seed with a glutinous outer layer, coat thin, no palisade layer, inwardly-curving exotegmen or hypostase plug, etc.; endosperm 0; n = ?
1/1: Diegodendron humbertii. Madagascar.
Synonymy: Diegodendraceae Capuron
Economic Importance. The orange colouring of Bixa orellana, annatto, is used as a food colouring, e.g. for margarine.
Chemistry, Morphology, etc. In Cochlospermum vitiifolium the median sepal is abaxial, there are no bracteoles, and the sepals are of unequal size (or three "true" sepals + two bracteoles?). Flowers of Cochlospermum are monosymmetric in bud, and the floral vasculature is monosymmetric. The androecium has five or six bundles, and development is centrifugal. Carpel orientation needs to be checked if the flower is inverted. There is no obvious nectary. Amoreuxia has obliquely (?tranversely) monosymmetric flowers, the positionally "upper" stamens being much shorter than the lower ones and differently coloured; the four "upper" petals are bicolored.
The gums of Cochlospermum and those of Sterculia (Malvaceae-Sterculioideae) are similar, both containing acetylated acidic polysaccharides.
See Keating (1970, 1972) and Poppendieck (1980, 2002) for more details.
For general information on Bixa, see Poppendieck (2002). The wood anatomy of Diegodendron is very like that of Sphaerosepalaceae (Dickison 1988), but the genus is otherwise poorly known (see also Bayer 2002).
Classification. From molecular data (Fay et al. 1998a), Diegodendron is very similar to Bixa in particular, and the absence of any distinctive seed coat anatomy in Diegodendron is probably because the fruit is indehiscent. Although Diegodendron does seem morphologically rather different from the other two groups (Kubitzki & Chase 2002, Table 1), nevertheless, all three have much in common. Cochlospermaceae and Diegodendraceae can optionally be placed in Bixaceae s.l. (see A.P.G. II 2003), and this seems reasonable.
Previous Relationships. Diegodendron was included in Ochnaceae by Cronquist (1981), but excluded by Amaral (1991); Diegodendraceae were placed in Malvales by Takhtajan (1997).
Cistaceae + Sarcolaenaceae + Dipterocarpaceae: ectomycorrhizae +; tracheids +; ellagic acid +; plant with secretory canals; K imbricate, two outer members often different from the rest, filaments not articulated, ovules both anatropous and straight; exotegmen curved inwards in chalazal region, hypostase plug with core and annulus; endosperm starchy.
Phylogeny. This grouping is strongly supported in molecular studies, albeit the sampling is poor, indeed, both Sarcolaenaceae and Cistaceae may be embedded within Dipterocarpaceae as currently circumscribed (Kubitzki & Chase 2002; Ducousso et al. 2004). Fancy the nomenclatural brouhaha that will result if this is confirmed! Nandi (1998b) noted several similarities between Cistaceae and Sarcolaenaceae (hollow style, stigma morphology, carpel number and indumentum). In both Fumana (Cistaceae) and Dipterocarpoideae the outer integument is elongated, and ectomycorrhizae are common (Appanah 1998; Ducoussu et al. 2004). Clearly, future morphological studies may well strengthen the characterisation of the whole clade, but equally clearly, clade limits within it are unclear.
CISTACEAE Jussieu, nom. cons. Back to Malvales
Herbs to shrubs; (flavonoid sulphates +); root hairs 0; cambium storying?; phloem not stratified; nodes 1:1; mucilage cells 0?; petiole bundles arcuate; cuticle waxes 0 (platelets, annular rodlets); hairs glandular, or simple, clustered, or stellate, each cell with a basal internal compartment; leaves opposite (spiral), ± conduplicate-curved, margin toothed, 2ndary veins pinnate or palmate, stipules + or 0; K and C ± opposite, K 5, 2 outer smaller than the others (3), C (3) 5, (imbricate), crumpled in bud; A (3-)many and centrifugal; pollen often starchy; G [3, 5(-10)], opposite petals or median member abaxial, style hollow, stigmas (sessile), small to capitate and/or lobed, dry, with multicellular multiseriate papillae; placentation parietal (-axile), (1-)2-many straight ovules/carpel, (anatropous, funicles short - Fumana), funicles long; testa often mucilaginous; embryo ± strongly curved, long, cotyledons thin, curved or folded, radicle short, stout; n = 5, 7, 9-12, 16, etc.

8[list]/175: Helianthemum (80-110), Crocanthemum (24), Cistus (18). More or less worldwide, if scattered, often temperate or warm temperate, esp. Mediterranean region (map: from Meusel et al. 1978). [Photos - Collection]
Evolution. Many Cistaceae have stamens that are sensitive to touch, moving and dusting the insect with pollen when it disturbs them.
Chemistry, Morphology, etc. For the absence of root hairs, at least in seedlings, and the fungal associations of the plant, see references in Arrington and Kubitzki (2002); Dickie et al. (2004) also mention ectomycorrhizae in Helianthemum. Wood rays are uniseriate and xylem parenchyma is almost absent (Keating 1966). Corolla initiation in Cistaceae tends to be retarded (Nandi 1998b: it is not retarded in Dipterocarpaceae, Kocyan 2005). The androecium has ten bundles, each member of the oppositisepalous whorl supplies a group of stamens while the traces of the inner whorl usually supply a single stamen only; Saunders (1936) suggested that in Cistus there are five oppositipetalous groups. At least some Cistaceae have starchy pollen grains. The embryo is green (1 record) or white (Nandi 1998a). Fumana has n = 16 (Guzmán & Vargas 2009).
For floral diagrams, see Eichler (1878), for floral development, see Nandi (1998b), for ovule and seed anatomy, Nandi (1998a), for general information, Arrington and Kubitzki (2002, revised in Arrington 2004). For more information on the web, see R. Page's Cistus and Halimium website.
Phylogeny. Fumana and Lechea are successively sister to the remainder of the family with 100% posterior probabilities but mediocre maximum parsimony support (Guzmán & Vargas 2009); the former in particular has a number of features that are plesiomorphic in the family (Arrington 2004). Both have only three petals, the former, alone in the family, has staminodes. for the phylogeny of Cistus, see Guzmán and Vargas (2005).
Previous relationships. Takhtajan's Cistales included Cistaceae, Bixaceae and Cochlospermaceae. Corner (1976 1: 97) described Cistaceae as being "little more than variations on a single generic theme", and noted similarities between the three families mentioned.
Synonymy: Helianthemaceae G. Meyer
Sarcolaenaceae + Dipterocarpaceae: petiole anatomy complex; stipules usu. well developed.
Phylogeny. Perhaps sister taxa (e.g. Alverson et al. 1998), at least if Pakaraimaeoideae are excluded from Dipterocarpaceae (Kubitzki & Chase 2002).
SARCOLAENACEAE Caruel, nom. cons. Back to Malvales
Woody, usu. evergreen; cyclopropenoid fatty acids +; cork?; wood not storied; true tracheids +; sclereids +; secretory canals?; stomata ?; hairs stellate or not; leaves two-ranked, involute [Keller 1996], stipules caducous; inflorescence various, involucre of varying morphology subtending 1 or 2 flowers, (bract [largely two stipules] enclosing flower), K 3(-5), when 5, outer 2 smaller, C 5 (6), disc +; A (10 [Leptolaena]-)many, ± connate into 5-10 bundles or not, of 2 or 3 lengths, anther basi- or dorsifixed, pollen in tetrads; G (1) [2-4 (5)], densely hairy, placentation basically axile, (1-)2-6(-many) ovules/carpel, style hollow, stigma capitate and/or ± lobed, with multicellular papillae; fruit (indehiscent), with persistent/accrescent bracts or cupule, endocarp hairy; seed hairy or not, often ruminate; endosperm copious (0), cotyledons cordate; n = 11.

8[list]/60. Madagascar, mostly E. and C. [Photos - Collection]
Evolution. Fossil pollen of Sarcolaenaceae is known from the Tertiary of South Africa (Coetzee & Muller 1984).
Chemistry, Morphology, etc. Sarcolaenaceae, the family with a "fleshy outer tunic". For details of cyclopropenoid fatty acid distribution, see Gaydou and Ramanoelina (1983), for ectomycorrhizae, see Ducousso et al. (2004).
Synonymy: Rhodolaenaceae Bullock, Schizolaenaceae Barnhart
DIPTEROCARPACEAE Blume, nom. cons. Back to Malvales
Trees, (ectomycorrhizal); triterpenoid dipterocarpol, sesquiterpene oleoresins +; cork also outer cortical; cambium storied; (vessel elements with scalariform perforation plates); tyloses +; cortical bundles +; secretory cavities in pith; nodes also 5:5; petiole geniculate; stomata ?; hairs tufted, peltate, etc.; leaves spiral and two-ranked, conduplicate(-plicate); inflorescence axillary, often branched; K (slightly connate); A initiation centrifugal, anthers ± versatile, with prolonged connective, median carpel abaxial, ovules apical, stigma slightly lobed or not; K thinnish, enlarging somewhat in fruit; seed usu. 1, testa vascularized; endosperm 0 (+), cotyledons often folded.

17[list]/680 - three groups below. Tropical, but overwhelmingly W. Malesian in species diversity (map: from Gottwald & Parameswaran 1966; Ashton 1982).
1. Monotoideae Gilg
Rays usu. uniseriate; adaxial gland at base of lamina; androgynophore +; G [3 (4)], (1-)2 (atropous) ovules/carpel, exostome prolonged; endosperm not starchy; n = ?
3/30: Monotes (26). Africa, Madagascar, South America (Colombian Amazon: Pseudomonotes) (map: above, area in green).
Synonymy: Monotaceae Kostermans
2. Pakaraimaeoideae Maguire, Ashton & de Zeeuw
Included phloem +; rays usu. biseriate; C imbricate, shorter than K; G [5], (2-)4 ovules/carpel; n = ?
1/1: Pakaraimaea roraimae. The Guaianan Highlands, South America (map: above, area in blue).
3. Dipterocarpoideae Burnett
Rays usu. multiseriate; resin ducts pervasive in wood; at least lateral bundles leaving central cylinder well before they enter the leaf; K imbricate [Shorea] or valvate; A (= and opposite sepals)-15, etc., anthers ± basifixed, pollen tricolpate, G [(2) 3] (inferior), 2(-3) ovules/carpel, micropyle endostomal; fruit usu. 1-seeded, usu. a nut, endocarp hairy, K enlarging unequally and thickish; (palisade exotegmen 0); endosperm 0 (+ - Dipterocarpus), not starchy; n = (6-)7, (10-)11(-13).
13/650: Shorea (360), Hopea (105: these two should perhaps be merged), Dipterocarpus (70), Vatica (60). Seychelles, Sri Lanka, India, South East Asia to New Guinea, but mostly W. Malesian, often dominating in mixed-species stands (map: above, area in red). [Photo - Flower, Fruit.]
Evolution. Dipterocarpaceae are large trees that are often dominant in the tropical lowland forests from India and Sri Lanka to West Malesia, where they grow both in seasonal and everwet forests. They are the most diverse family in the latter, at least (Gentry 1988); elsewhere in the range of the family members are also often described as being locally abundant. Individual species tend to be mast fruiters, and in the everwet forests of S.W. Sri Lanka and West Malesia all taxa in the family tend to flower and especially fruit at the same time, apparently in response to climatic changes induced by El Niño events; this kind of behaviour is very uncommon in the tropics. Pollination of species of Shorea section Mutica and of other dipterocarps in Sarawak is by chrysomeild and curculionid beetles; in peninsula Malaysia species of section Mutica are apparently pollinated by thrips (Sakai et al. 1999). Fruit dispersal is predominantly by wind, but the fallen fruits are eaten. The ectomycorrhizal habit of the family has been implicated in the distinctive phenological behaviour, and/or predator satiation that occurs in the family (Janzen 1974a, no mention of fungi; Ashton 1982, 2002).
For ectomycorrhizae, see Smits (1994) and Appanah (1998); roots may lack root hairs. For mycorrhizae in Pakaraimaeoideae, see Moyersoen (2006). Dipterocarpaceae lack parasitic rust fungi (Uredinales), unlike many other ectomycorrhizal groups (Malloch et al. 1980); this observation should be confirmed - or not - for other members of this clade.
Hemipteran coccoid Beesonidae form distinctive galls on Dipterocarpoideae, although details of the association are poorly known (Gullan et al. 2005).
Economic Importance. Dipterocarpoideae, with their long, straight and clean boles and gregarious habits, are a major source of commercial hardwood. In the mid 1980s they comprised 25% of the world trade in tropical hardwoods, and 80% of that was made up by timber from Shorea. Furthermore, both oleoresins and hard resins (dammar) are collected from a number of Dipterocarpoideae, and they also a source of lac (exudate produced by Coccoidea), butter fat from the fruits, etc. (Ashton 1982; Smits 1994 and references).
Chemistry, Morphology, etc. For additional information on Monotoideae see Catalina Londoño et al. (1995), Morton (1995). For additional information about Pakaraimaea, see Maguire et al. (1977) and Maguire and Ashton (1980).
Bergenin, a derivative of gallic acid, is widespread. The stipules of Stemonoporus are extremely caducous. The calyx in many Shoreeae is imbricate in fruit, while the corolla is basally connate in many Dipterocarpeae. Dipterocarpus has vascular bundles in the inner integument.
For additional information, see Gottwald and Parameswaran (1966: wood anatomy), Ashton (1982: general, 2002: general), Kajita et al. (1998), and Kocyan (2005: floral development).
Phylogeny. See Morton et al. (1999) and Dayanandan et al. (1999) for relationships.
Classification. Ashton (1982) is a convenient account covering most of Dipterocarpodeae. Generic limits within Dipterocarpaceae need attention (Yulita et al. 2005), Neobalanocarpus possibly even being an intergeneric hybrid (Shorea sp. X Hopea sp., see Kamiya et al. 2005).
Cytinaceae + Muntingiaceae: ellagitannins +; ovules numerous; fruit a berry, seeds numerous.
Chemistry, Morphology, etc. For general information about this family pair, see Nickrent (2007). Synapomorphies for the clade depend on more detailed knowledge of all aspects of the poorly-known Muntingiaceae and a comprehensive phylogeny of both families.
CYTINACEAE A. Richard Back to Malvales
Echlorophyllous endophytic root parasites; vessels 0; phloem +; sieve tube plastids without starch or protein inclusions; stomata?; leaves spiral; plant monoecious or dioecious; inflorescence racemose, capitate or spicate; P 4-9, basally connate; staminate flowers: A 6-20+, extrorse, connate, monothecal, (P joined by dissepiment to both A and stylodium), nectariferous cavities between stamens, connective massive, with terminal (branched) appendage (0), pollen 2-4-porate or 3-colpate, (in tetrads), stylodium +; carpellate flowers: staminodes 0, nectariferous cavities near base of style; G [5-14], inferior, placentation intrusive parietal, style +, swollen towards the apex, stigma capitate-radiate, commissural; ovules uni- or bitegmic straight tenuinucellate ovules/carpel, nucellar epidermis persists, micropyle endostomal; seeds embedded in pulp; exotegmic cells thickened all around [?both genera?]; endosperm +, embryo undifferentiated; n = 12, 16.

2/10. Mexico, Mediterranean, South Africa and Madagascar (map: from Jalas & Suominen 1976; Alvarado-Cárdenas et al. 2009). [Photo - Collection of Cytinus, Bdallophytum - Staminate Flower, Carpellate Flower.]
Evolution. Cytinus is quite often parasitic on Cistaceae (same order!) in the Mediterranean region, but on a variety of non-Malvalean families, perhaps especially Asteraceae, in Africa (Smythies & Burgoyne 2010), and in America Bdallophytum is most commonly found on species of Bursera (Alvarado-Cárdenas et al. 2009).
Pollination of Cytinus hypocistus is by ants, other members of the genus are pollinated by mammals (de Vega et al. 2009; Smythies & Burgoyne 2010).
The seeds become embedded in mucilaginous material derived from the placentae and funicles (Nickrent 2007).
Chemistry, Morphology, etc. The elagitannin is isoterchebin (Hegnauer in Meijer 1997). In Cytinus Harms (1935a) reports a nectary at the base of the style and the staminal tube. The outer integument, when present, is much reduced. The seeds of Bdallophytum have a blunt projection at both ends (Alvarado-Cárdenas et al. 2009).
For general information see the Parasitic Plants website (Nickrent 1998 onwards) and also Heide-Jørgensen (2008), de Vega et al. (2008) for population differentiation in the western Mediterranean, and de Vega et al. (2007) for the anatomy of the endophytic portion of the host.
Classification. For a monograph of Bdallophytum, see Alvarado-Cárdenas et al. (2009).
MUNTINGIACEAE C. Bayer, M. W. Chase & M. F. Fay Back to Malvales
Trees; ellagic acid +; pits not vestured; tracheids +; non-dispersive protein bodies?; mucilage canals 0; petiole bundle annular, no pericyclic fibers; stomata ?; hairs stellate or tufted; leaves two-ranked, conduplicate-subplicate [Muntingia], margins toothed, 2ndary veins pinnate to palmate, stipules 0, but prophylls basal, heteromorphic; flowers fasciculate, extra-axillary, (4-)5-merous; K valvate, basally connate, C imbricate, shortly clawed, crumpled in bud; anther wall development the basic type; pollen small, ca 10 µm, triporate; nectary on [inside of] broad disc; G [5(-7)], or inferior, opposite petals, with numerous septae, placentation axile-laminar, or with bilobed axile-pendulous placentae, style stout, stigma conical, 5-ridged, or ± capitate; ovules with multicellular archesporium, micropyle zig-zag, funicle long [Muntingia], embryo sac monosporic, tetranucleate, spore micropylar [Muntingia]; K persistent or deciduous; seed with mucilaginous funicle; exotesta mucilaginous, endotesta crystalliferous, cells of exotegmen shortly elongated; endosperm +, diploid, starchy [details of seed taken from Muntingia alone]; n = 15.

3 [list] (Dicraspidia, Muntingia, Neotessmannia)/3. Tropical America. [Photo - Flower]
Chemistry, Morphology, etc. Some characters of Muntingiaceae (lack of stipules; pits not vestured) might suggest that the family may be rather basal in Malvales; characters of the young secondary tissue of Muntingia, such as widely flaring rays, stratified phloem, etc., are like those of most other Malvales. Petenaea (?Elaeocarpaceae) may also be associated with this clade (Bayer et al. 1999).
Petenaea is described as having minute stipules (Bayer 2002). However, although Muntingiaceae appear to have stipules, this may not to be the case. Dicraspidia has strikingly asymmetric prophylls; on the adaxial side of the branch they are orbicular, foliaceous and persistent, while on the abaxial side they are linear, thin and caducous. In Muntingia only an adaxial prophyll is present, and it is narrow (Karima Gaafar, pers. comm.: the situation in Neotessmannia is unknown). Sensarma (1957) suggested that the nodes of Muntingia are trilacunar, he interpreted the prophyll as a stipule, nevertheless, nodes indeed appear to be trilacunar. Given that stipules are common in Malvales, the situation in Muntingiaceae needs to be clarified.
Muntingia has a superior, ovary, caducous calyx, and pendulous placentae, the two other genera have inferior ovaries, laminar placentation, and a persistent calyx (?: Neotessmannia). Muntingia has erect uniseriate hairs in addition to its tufted hairs. Stamens, etc., are borne on a massive, almost disc-like structure towards the inside of which are dense hairs; the inner side of this disc as it faces the ovary seems to be a nectariferous region.
Some information is taken from Benn and Lemke (1991) and Venkata Rao (1952); the latter suggested that there were glandular, nectar-secreting hairs in Muntingia, but the hairs seem eglandular to me. Bayer (2002) gives a general account of the family. For relationships, see Bayer et al. (1998c), for carpel orientation, see Ronse Decraene (1992), and for anatomy, see Carlquist (2005a). I am grateful to Lucia Lohmann for sending me material of Muntingia.
Previous relationships. Muntingiaceae were placed in Tiliaceae - Neotessmannioideae by Takhtajan (1997).
MALVACEAE Jussieu, nom. cons. Back to Malvales
Shrubs to trees (herbs); cyclopropenoid fatty acids, terpenoid-based quinones +; (cork cambium outer cortical); wood commonly fluoresces; pits not vestured; tile cells common; sieve tubes with non-dispersive protein bodies; leaves spiral or two-ranked, usu. conduplicate(-plicate), margins entire or toothed, single vein running to the non-glandular apex, 2ndary venation palmate; inflorescence made up of modified cymose units ["bicolor units"]; (epicalyx +), K valvate, (C imbricate; 0), (androgynophore +); A (5-)many, in five groups, but fundamentally obdiplostemonous, extrorse; G [(3-)5(-many)], variable in orientation, micropyle zig-zag or endo- or exostomal, (nucellar cap +), stigma usu. dry; fruit a capsule (berry, schizocarp, etc.; muricate); endotesta crystalliferous; endosperm often starchy, embryo often green; sporophytic self-incompatibility system present [in "Sterculiaceae"].
243[list]/4225+ - 9 groups below. Largely tropical, also temperate. [Photos - Collection]
Grewioideae + Byttnerioideae: ?
1. Grewioideae Hochreutiner
K ± free, no nectary, C adaxially with various epidermal modifications, (basally with nectariferous hairs), androgynophore + (0), (nectariferous); A usu. free, first A antesepalous, additional A centrifugal, antesepalous A + (0), staminodes 0; pollen prolate; G [2-10]; ovule with micropyle exo- or endostomal; n = 7-9(10).
25/770: Grewia (290), Triumfetta (150), Corchorus (40-100), Microcos (60). Pantropical (warm temperate).
Synonymy: Grewiaceae Doweld & Reveal, Sparmanniaceae J. Agardh, nom. cons.
2. Byttnerioideae Burnett
Petiole bundle ± incurved-arcuate; leaves (palmate - Herrania), spiral or distichous; K usu. connate, C broad at the base [hooded; with inrolled margins], then limb clawed; A epipetalous, 5(-30), in antepetalous fascicles alone, antesepalous staminodes + (0), petaloid, fascicles + staminodes forming a tube, false plasmodial tapetum [contents of cells resorbed], nectary exuding through stomata [?level], style apically ± branched; n = (5-7) 10(-13).
26/650: Byttneria (135), Hermannia (100), Ayenia (70), Melochia (55: A 5), Theobroma (20). Pantropical, esp. South America. [Photo - Flower, Flower, Fruit.]
Many taxa have only five stamens, a derived condition, even although developmental work might suggest that the higher numbers are derived by doubling (Whitlock et al. 2001b).
Synonymy: Byttneriaceae R. Brown, nom. cons., Cacaoaceae T. Post & Kuntze, Hermanniaceae Durande, Lasiopetalaceae Reichenbach, Melochiaceae J. Agardh, Theobromataceae J. Agardh
Sterculioideae + Tilioideae + Dombeyoideae + Brownlowioideae + Helicteroideae + [Malvoideae + Bombacoideae]: 21 bp deletion in ndhF gene.
3. Sterculioideae Burnett
Petiole bundle annular, with medullary bundle; leaves spiral, often palmately compound [basal?]; plant monoecious, inflorescence axillary, paniculate, lacking obvious bicolor units, epicalyx 0; K petaloid, C 0, androgynophore +, (nectariferous hairs at the base), filaments connate, anthers ± sessile, wall development the basic type? [5-6 cells across], staminodia 0; G largely free, micropyle endostomal, styles +; fruit a follicle (indehiscent), (endocarp pubescent); n = (15, 16, 18) 20 (21, etc.). commonly.
12/430: Sterculia (150), Cola (125). Pantropical. [Photo - Staminate flowers, Fruit.]
Wilkie et al. (2006) suggest relationships within Sterculioideae and discuss evolution of fruit types, dispersal mechanisms, etc.; leathery follices seem to be the basal condition in the group.
Synonymy: Sterculiaceae Salisbury, nom. cons.
4. Tilioideae Arnott
Plant ectomycorrhizal; stachyose, raffinose + [phloem exudate - Tilia]; some species with siliceous leaves; petiole bundle annular, with medullary phloem strands and inverted bundles; leaves distichous, horizontally conduplicate [?always]; K free; A free, staminodia and stamens antepetalous, antesepalous sector empty; G opposite sepals; (seeds arillate); cotyledons folded; n = 41.
3/50: Tilia (23). N. temperate, Central America (map: from Meusel et al. 1978; Hultén & Fries 1986).
Mortoniodendron is to be included in this clade (Nyffeler et al. 2005).
Synonymy: Tiliaceae Jussieu, nom. cons.
5. Dombeyoideae Beilschmied
Leaves spiral; epicalyx + (0); K connate basally to free, nectary at base, A connate (free), (5-)10-10(-30), elongated antesepalous staminodes + (0), staminodes forming a short tube, (anther wall development the basic type); (tapetum amoeboid); pollen often porate, spinulose; G [(2-)5(-10)]; ovules with (multicellular archesporium; nucellar cap +), outer integument 3-4 cells across, inner integument 4-5 cells across, micropyle bitegmic; endocarp often pubescent; seed with umbonate sarcotestal projections; cotyledons bifid; n = 19, 20, 30, etc.
21/381: Dombeya (225), Melhania (60). Old World tropics, St Helena, esp. Madagascar and Mascarenes.
Synonymy: Dombeyaceae Desfontaines
6. Brownlowioideae Burret
Indumentum often lepidote; inflorescences axillary; K connate, campanulate, splitting irregularly into 2-3 lobes; A ca 30, in antepetalous bundles, anther thecae sagittate, often broadly so, staminodia antesepalous (0), petaloid, ca 2 ovules/carpel, style or styles +; K persistent; n = 10.
8/68: Pentace (25). Tropical, esp. Old World.
Synonymy: Berryaceae Doweld, Brownlowiaceae Cheek
7. Helicteroideae Meisner
(Indumentum lepidote); (secondary veins pinnate - Durioneae); K connate, petals often with lateral constrictions, androgynophore +; A usu. with short tube and/or fascicles, pollen baculate or microverrucate to suprareticulate; (seeds arillate); n = 9, 14, 20, 25, etc.
8-10/95: Helicteres (40), Durio (27). Tropical, esp. Southeast Asia and W. Malesia. [Photo - Fruit.]
Synonymy: Durionaceae Cheek, Helicteraceae J. Agardh, Triplochitonaceae Schumann
Malvoideae + Bombacoideae: leaves spiral; K connate, adaxially at base with multicellular clavate nectariferous hairs, C adnate to base of A, staminal tube +, (locellate [= "polysporangiate": some "basal" members]), anther thecae sessile ["basal" members], unbranched synlateral [± oppositisepalous] vascular bundle +.
8. Malvoideae Burnett
Petiole bundle annular; (epicalyx +), median K often abaxial; A from 5 antepetalous primordia, (dividing into two), centrifugal, tube often with 5 apical teeth, (anthers bisporangiate/monothecal), [staminodes in fascicle; antesepalous member], (synlateral bundle 0), tapetum plasmodial, pollen often spiny, 7+ porate; G (1[2-)3-many], styles often +, stigmas decurrent [e.g. Malva] to capitate, hairy; (fruit schizocarpic); (seeds hairy); embryo curved, cotyledons folded; n = 5-20(+).
78/1670: Hibiscus (580, inc. Pavonia, etc.), Sida (200), Abutilon (100), Nototriche (100), Cristaria (75), Gossypium (40). Temperate to tropical (map: from Hultén & Fries 1986; Meusel et al. 1978). [Photo - Flower]
Synonymy: Chiranthodendraceae A. Gray, Hibiscaceae J. Agardh, Pentapetaceae Sprengel
9. Bombacoideae Burnett
Trunk often stout, with parenchymatous water-storage tissue, so soon becomes punky when cut, stout prickles, bark thin, often green; leaves palmate(ly lobed - e.g. Ochroma); flowers axillary, 1-2 together, (K imbricate - Ochroma), filaments usu. fasciculate, (A 5), (anthers bisporangiate/monothecal), wall development the basic type? [5-7 cells across], pollen ± flattened, triangular, staminodes usu. 0; ovule with nucellar cap; endocarp pubescent; embryo curved; n = 36(-46).

16/120: Pachira (24), Pseudobombax (20). Tropical, esp. America and Africa (map: from Aubréville 1974). [Photo - Flower], Seeds.]
Synonymy: Bombacaceae Kunth, nom. cons.
Evolution. Wood attributable to Malvaceae is known from the late Maastrichtian (Cretaceous) ca 68 million years ago. It has simple perforation plates in radial miltiples and storied wood; tile cells were not reported (Wheeler et al. 1994). Pollen of Bombacoideae (Bombacacidites) is known from deposits 69-65 million years old (Krutzsch 1989; Pfeil et al. 2002 for references). For the early Tertiary fossil history of Graigia (Tilioideae), see Manchester et al. (2009); Reevesia (Helicteroideae), now East Asian, is known from throughout the North Hemisphere in the early Tertiary (Ferguson et al. 1997).
Bombacoideae are an important component of neotropical seasonally dry tropical forest (Pennington et al. 2009).
Caterpillars of the nymphalid Acraea are quite commonly found on members of Malvaceae, as are members of Lycaeninae (Fielder 1995). Acanthoscelides, whose larvae eat seeds, are bruchids (Chrysomeloidea-Bruchidae/Bruchinae) that have diversified on Malvaceae s.l., esp. on Malvoideae, more than on other groups outside their primary hosts, members of Fabaceae (Kergoat et al. 2005). Seed-eating bugs of the Hemiptera-Lygaeidae-Oxycareninae are also concentrated on Malvoideae (Slater 1976).
In most Malvaceae studied the nectary is made up of multicellular glandular hairs covering glandular tissue beneath; the same morphology even occurs in the foliar extrafloral nectaries in Triumfetta (Leitão et al. 2005). The nectaries that provide rewards for pollinators are commonly found on the inside of the calyx and the corolla is fused at most basally, the petal lobes soon separating and leaving a space through which the pollinator can get at the nectar. These two features are probably connected, because access to the nectar must be permitted, and if the corolla were completely connate there would be no easy way for the pollinator to reach the nectar while simultaneously pollinating the flower (for nectary morphology, etc., see Sawidis et al. 1989 and references; Vogel (2000; Leitão et al. 2005). However, overall there is considerable variation in nectary - and staminodium - position and type in the family. Nectaries are borne on the petals in e.g. Grewia and Luehea and on the androgynophore and leaf in e.g. Triumfetta (all Grewioideae: Leitão et al. 2005), while nectar in some Byttnerioideae, at least, is secreted through nectarostomata. Byttnerioideae often have remarkably complex if sometimes quite tiny "basket" flowers that are pollinated by small flies and the like (see Westerkamp et al. 2006). The petal often has a concave lower portion more or less enclosing the rest of the flower, the limb of the petal may dangle and twist in the wind, as in Abroma, and the prominent staminodia are opposite the sepals. For pollination in Bombacoideae, see Janka et al. (2008) and references. The flowers of Helicteres isora (Helicteroideae) and some species of Pavonia (Malvoideae) can be monosymmetric, but I do not know the plane of symmetry.
The pattern of evolution of dioecy in Dombeya, certainly paraphyletic, in the Mascarenes and its dispersal between Madagascar, the Mascarenes, and Africa is complex (Le Péchon et al. 2009). Diversification and molecular evolution in Hibisceae pick up almost together (Baum et al. 2002, 2004), while Andreasen and Baldwin (2001) noted that the rate of molecular evolution of 18S–26S nuclear ribosomal DNA in annual Sidalcea was faster than that in the perennials.
Economic Importance. For the domestication of cotton (Gossypium barbadense), see Dillehay et al. (2007).
Chemistry, Morphology, etc. Tile cells are best observed in radial section; they are of two or three main types (Manchester & Miller 1978; Carlquist 1988b; Tang et al. 2005b). Any correlation of tile cell "type" with phylogeny awaits a more completely resolved tree, thus the Durio type occurs in both Malvoideae and Byttnerioideae. Vestured pitting is reported, but probably incorrectly, from Schoutenia and Ochroma (Jansen et al. 2000a). For a discussion of the various kinds of extrafloral nectaries in Triumfetta, see Letãio et al. (2005). The inflorescence in most Malvaceae is made up of "bicolor units" - a terminal flower with three bracts, two of which may subtend cymose part inflorescences with normal bracteole number and arrangement and the other subtends nothing. The epicalyx seems to be made up of these three bracts, and so a flower with an epicalyx represents a highly reduced "bicolor unit" (Bayer 1999); it has evolved several times in the family. Although the inflorescence of Sterculioideae seems to be very differently constructed from that of other Malvaceae, it, too, may be made up of much modified bicolor units (Bayer 1999). Nototriche (Malvoideae) has epiphyllous inflorescences.
For androecial development in Malvaceae s.l. compared to that in other Malvales, see Nandi (1998b) and von Balthazar et al. (2006). Von Balthazar et al. (2006) suggest an interpretation of the androecium of [Malvoideae + Bombacoideae] - and extend their findings to determine the basic androecial structure for Malvaceae as a whole. They propose that the basic androecial structure in Malvaceae is obdiplostemonous, with stamens developing in one or both whorls; anther dehiscence is extrorse. In [Malvoideae + Bombacoideae] each androecial unit consists of an antesepalous primordium with its own vascular supply and which remains sterile (usually). It is flanked on either side by a single primordium, each derived from a separate antepetalous primordium and that each give rise to a sessile, elongated theca. Each theca is supplied by a branch from an antepetalous vascular bundle. The androecial unit thus consists of [half anther + sterile primordium + half anther]. The androecia with numerous stalked, unithecate, staminal units so common in this clade are independently derived in Bombacoideae and Malvoideae (von Balthazar et al. 2006). Further details are given by Janka et al. (2008), focussing on Adansonia and relatives. Ceiba pnetandra has only five alntisepalous stamens, supplied by branches of the oppositipetalous traces. For additional details of androecial development in Malvoideae, see Janka (2003) and von Balthazar et al. (2004). For floral morphology and development in Dombeyoideae, see Tang (1998) and Tang et al. (2006). In Grewioideae the stamens may arise from ten or five oppositisepalous primordia, or from ringwall primordia and the vascular supply to the stamens is variable in origin (Brunken & Bayer 2005); for additional details of floral development, see Brunken (2003). Van Heel (1966) and Schönenberger and von Balthazar (2006) also discuss androecial development.
Although starchy pollen is common in Malvaceae in the old sense, it is not in the old Sterculiaceae; details of variation in the clades recognised here are unclear.
Not only are the carpels of Sterculioideae secondarily free, but in Firmiana they open early in development, exposing the developing seeds on the carpel margins; the ripe fruits with seeds attached are dispersed by wind. However, there is a compitum even in these apocarpous Sterculioideae; it develops after post-genital connation of the apical parts of the styles (Jenny 1988). The carpels are usually opposite the corolla, although not infrequently (e.g. Hibiscus, Fremontodendron, Sterculia) they are opposite the calyx; when there are three carpels, the median member may be either ad- or abaxial (reports of carpel orientation in individual taxa may conflict - e.g. Eichler 1878; Ronse Decraene 1992). Leptonychia has parietal placentation, short fibers in the exotegmen, but starchy endosperm.
Several taxa have palmate leaves. In Brachychiton and Adansonia there is comparable variation within a flush - the first leaf/leaves have a very short petiole and long, narrow ?phyllode, while later leaves are palmate; any intermediates have winged petioles and a few leaflets.
For nectary morphology, etc., see Sawidis et al. (1989 and references), for inflorescence structure, Bayer (1994, 1999), for nectaries, Vogel (2000), for wood anatomy, Chattaway (1933b, 1937), Webber (1934), Manchester and Miller (1978), Carlquist (1988b) and Tang et al. (2005a, b), for pollen variation in Grewioideae, Tilioideae and Brownlowioideae, see Perveen et al. (2004), and for the embryology of Eriolaena in the context of embryological variation in Dombeyoideae as a whole, see Tang et al. (2009). For more general information, see the Malvaceae Pages website (Hinsley 2002), Cheek (2007), and especially Bayer and Kubitzki (2002).
Phylogeny. Apart from Malvaceae, here in Malvoideae, all the other families in the old Malvales are highly para- or polyphyletic. For information on relationships in the extended family, see Alverson et al. (1998, 1999) Bayer et al. (1999), and Nyffeler et al. (2005). [Grewioideae + Byttnerioideae] are probably sister to the rest of the family (see also Soltis et al. 2007a), while Sterculioideae are perhaps sister to the well supported [Malvoideae + Bombacoideae]; see the latter for relationships within it. Other relationships between the subfamilies are unclear. However, Dombeyoideae and Tilioideae are sometimes weakly associated (Alverson et al. 1999), but the former may rather be sister to all other taxa in the major polychotomy (Nyffeler et al. 2005).
For relationships in Dombeyoideae, Dombeya certainly being paraphyletic, see Le Péchon et al. (2009) and Won (2009: to include Corchoropsis). Within Helicteroideae, Helictereae (ex Sterculiaceae) are sister to Durioneae (ex Bombacaceae), the latter being from Sri Lanka, Burma to West Malesia and having lepidote indumentum and an initially connateepicalyx. The anthers of many Durioneae are polylocular - see also Nyffeler and Baum (2000).
Ochroma may be sister to other members of Bombacoideae; its filaments are connate into a tube. However, it and Patinoa formed a clade with no obvious immediate link with Bombacoideae in some analyses (Alverson et al. 1999; see also Baum et al. 2002, 2004) and are best excluded from it. Sister to the rest - or almost so - in this whole [Malvoideae + Bombacoideae] area may be things like Fremontodendron, etc. (ex Sterculiaceae - C 0), and Quararibea, etc. (ex Bombacaceae), but with only weak support (Alverson et al. 1999; cf. Bayer et al. 1999; Baum et al. 2002). Quararibea, etc., are best placed in Malvoideae, while [Ochroma + Patinoa] and Septotheca are unplaced (Baum et al. 2004). Indeed, Baum et al. (2004) suggest that [Fremontodendron + Chiranthodendron] may be sister to the rest of the [Malvoideae + Bombacoideae] area since they lack a 6 bp deletion in a conserved region of the matK gene found in all other members of this clade, although there is little other evidence for this position. Taxa at these nodes would have been neotropical (Nyffeler et al. 2005).
For groupings within Malvoideae, see La Duke and Doebley (1995: restriction site analysis) and Judd et al. (2002). Within Malveae, Tate et al. (2005) found that presence or absence of an epicalyx correlated very well with two major clades recognizable on analysis of ITS sequence data, but a subsequent study using this gene and four others found that Malva, at least, was polyphyletic (García et al. 2009). There generic limits had in the past been based mainly on the number of parts of the epicalyx and their fusion, but fruit characters seemed to be more useful features to use when characterising clades. For relationships within Hibisceae, see Pfeil et al. (2002), Pfeil and Crisp (2005) and Koopman and Baum (2008: Malagasy taxa); generic limits around Hibiscus are especially difficult - s. str. or s. lato?, but Hibiscus should probably include Pavonia, etc.
Classification. For a discussion of groupings in the extended family, Robert Brown's comments over 150 years ago (Brown 1814) on family limits in the Malvales (= Malvaceae here) are a good starting point. Malvaceae + Bombacaceae + Sterculiaceae + Tiliaceae make a readily recognized and well circumscribed group, yet the clades within it are mostly difficult to distinguish, even with flowers, so combination seems sensible (Judd & Manchester 1997; Alverson et al. 1999; Bayer et al. 1999); Cheek (2007), however, opts for wholesale dismemberment into ten families.
Synonymy: Plagianthaceae J. Agardh, Philippodendraceae Endlicher, Triplobaceae Rafinesque