LIGNOPHYTA

True roots +; lateral meristems: cork cambium producing cork abaxially, vascular cambium producing phloem abaxially and xylem adaxially.

EXTANT SEED PLANTS/SPERMATOPHYTA

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 derived from (some) sinapyl and particularly coniferyl alcohols, thus containing p-hydroxyphenyl and guaiacyl lignin units, (lignins derived from p-coumaryl alcohol, i.e. S [syringyl] lignin units); true roots present, apex multicellular, xylem exarch, and 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 and rays alone, tracheid/tracheid pits circular, bordered; mature sieve tube/cell lacking functioning nucleus, plastids with starch grains; phloem fibres +; stem cork cambium superficial, root cork cambium deep seated; leaves with single trace from sympodium ["nodes 1:1"]; stomata ?; leaf vascular bundles collateral; leaves megaphyllous [determinancy evolved first, then ad/abaxial symmetry], spiral, simple, lamina with vein density up to 5 mm/mm² [mean for all non-angiosperms 1.8]; axillary buds associated with at most some leaves; prophylls [including bracteoles] two, lateral; 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, developing after pollination, 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 duplications [three - [BP [A/N + C/O]] - copies], nrDNA with 5.8S and 5S rDNA in separate clusters; mitochondrial nad1 intron 2 and coxIIi3 intron and trans-spliced introns present.

MAGNOLIOPHYTA

Lignans, O-methyl flavonols, dihydroflavonols, triterpenoid oleanane, non-hydrolysable tannins, quercetin and/or kaempferol +, apigenin and/or luteolin scattered, [cyanogenesis in ANITA grade?], S [syringyl] lignin units common, positive Maüle reaction [syringyl:guaiacyl ratio more 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; shoot apex with tunica-corpus construction, tunica 2-layered; reaction wood ?, with gelatinous fibres; starch grains simple; primary cell wall mostly with pectic polysaccharides, poor in mannans; tracheid:tracheid [end wall] plates with scalariform pitting, wood parenchyma +; sieve tubes enucleate, sieve plate with pores (0.1-)0.5-10< µm across, cytoplasm with P-proteins, cytoplasm not occluding pores of sieve plate, companion cells from same mother cell that gave rise to the sieve tube; sugar transport in phloem passive; nodes unilacunar [1:?]; stomata with ends of guard cells level with pore, paracytic, outer stomatal ledges producing vestibule; leaves petiolate, lamina [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; most/all leaves with axillary buds; flowers perfect, pedicellate, polysymmetric, parts spiral [esp. the A], free, numbers unstable, development in general centripetal; P not sharply differentiated, with a single trace, outer members not enclosing the rest of the bud, often smaller than inner members; A many, filament not sharply distinguished from anther, stout, broad, with a single trace, anther introrse, tetrasporangiate, sporangia in two groups of two [dithecal], ± embedded in the filament, 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, 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, stylulus short, hollow, cavity not lined by distinct epidermal layer, stigma ± decurrent, dry [not secretory]; ovules few [?1]/carpel, marginal, anatropous, bitegmic, micropyle endostomal, outer integument 2-3 cells across, often largely subdermal in origin, inner integument 2-3 cells across, often dermal in origin, parietal tissue 1-3 cells across [crassinucellate], nucellar cap?; megasporocyte single, hypodermal, megaspore tetrad linear, functional megaspore chalazal, lacking sporopollenin and cuticle; female gametophyte four-celled [one module, nucleus of egg cell sister to one of the polar nuclei]; P deciduous in fruit; seed exotestal; pollen binucleate at dispersal, trinucleate eventually, germinating in less than 3 hours, pollination siphonogamous, tube elongated, growing at 80-600 µm/hour, with pectic outer wall, callose inner wall and callose plugs, growing between cells, penetration of ovules via micropyle [porogamous] within ca 18 hours, distance to first ovule 1.1.-2.1 mm, tube moves between nucellar cells; double fertilisation +, endosperm diploid, cellular [micropylar and chalazal domains develop diffently, first division oblique, micropylar end initially with a single large cell, divisions uniseriate, chalazal cell smaller, divisions in several planes], copious, oily and/or proteinaceous, embryo cellular ab initio, minute; germination hypogeal, seedlings/young plants sympodial; Arabidopsis-type telomeres [(TTTAGGG)_n]; whole genome duplication, ndhB gene 21 codons enlarged at the 5' end, 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 three copies of the PHY gene, [PHYB [PHYA + PHYC]].

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. This is because some taxa basal to the [magnoliid + monocot + eudicot] group have been surprisingly little studied, there is considerable homoplasy as well as variation within and between families of the ANITA grade 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 as details of sugar transport in the phloem, their placement on the tree is frankly speculative. Finally, for features such as parietal tissue/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), I am unsure where on the tree a thicker nucellus and a stylar epidermal layer are acquired.

[NYMPHAEALES [AUSTROBAILEYALES [[CHLORANTHALES + MAGNOLIIDS] [MONOCOTS [CERATOPHYLLALES + EUDICOTS]]]]]: vessels +, elements with elongated scalariform perforation plates; wood fibres +; axial parenchyma diffuse or diffuse-in-aggregates; tectum reticulate-perforate [here?]; ?genome duplication; "DEAER" motif in AP3 and PI genes lost, gaps in these genes.

[CUCURBITALES + FAGALES]: ovary inferior; fruit 1-seeded, indehiscent.

Evolution. Divergence & Distribution. Imperfect flowers pervade the two orders, but flower type varies considerably in Anisophylleaceae, with perfect flowers occurring in Combretocarpus, so there may have been reversal of this character, or imperfect flowers have evolved more than once. "Embryo with large cotyledons" may be another synapomorphy (Zhang et al. 2006), also a three-carpellate gynoecium. However, ovary position and fruit characters in particular reverse spectacularly in this clade (see also Matthews & Endress 2004; Zhang et al. 2006), although Endress (2011a) though that an inferior ovary might be a key innovation somewhere around here.

CUCURBITALES Berchtold & J. Presl  Main Tree, Synapomorphies.

(Frankia infection via intercellular penetration); ellagic acid ?; storied fusiform cambial initials; perforation plates not or minimally bordered; tension wood?; rays wide, multiseriate; cuticle wax crystalloids 0; leaves spiral, lamina with 2ndary veins palmate; K or P valvate, stomata on K/P raised, the two whorls rather similar in texture; (ovary with a roof, styluli +, often marginal); ovule with bistomal micropyle. - 7 families, 129 genera, 2295 species.

Evolution. Divergence & Distribution. Wikström et al. (2001: relationships are [Fabales [Rosales [Cucurbitales + Fagales]]]) date the origin of stem Cucurbitales to ca 84 million years before present (constrained), diversification beginning (68-)66(-64) million years before present. The age of crown group Cucurbitales was estimated as (85-)80, 78(-73) million years (two penalized likelihood dates), the stem group age being (107)103(-100) or (82-)88(-84) million years; Bayesian relaxed clock estimates were slightly older, to 90 or 109 million years respectively (Wang et al. 2009: note that relationships are [Rosales [Fabales [Cucurbitales + Fagales]]]), while Magallón and Castillo (2009: relationships are [Fabales + Rosales] [Cucurbitales + Fagales]) gave ages of ca 102.3 and 102.6 million years for relaxed and constrained penalized likelihood estimates for stem group Cucurbitales, and an age of ca 86.7 million years (both relaxed and constrained estimates) for the crown group.

Floral Biology & Seed Dispersal. Zhang and Renner (2003) suggested that the flowers are usually imperfect, but perfect flowers are known from Anisophylleaceae. Indeed, breeding systems vary considerably in Cucurbitales, and Schaefer and Renner (2010) have recently suggested that even within Momordica (Cucurbitaceae) there have been perhaps seven reversals from dioecy to monoecy - in a clade that is very approximately 35 million years old.

Plant-Animal Interactions. Butterfly caterpillars may be relatively uncommon on members of the order.

Chemistry, Morphology, etc. Cuticle waxes are usually not well developed. Possession of libriform fibres and slightly oblique end walls of vessel elements may be synapomorphies (Wagstaff & Dawson 2000, see also data in Nandi et al. 1998), as may banded wood parenchyma (Baas et al. 2000), homogeneous rays, and the possession of bitter compounds. Stipules are not placed as a synapomorphy of the order, although given the data presented by Matthews and Endress (2004) that might seem an option; not only do the nature of the stipules in Anisophylleaceae and also Corynocarpaceae need clarification, but the presence of stipules may well be a synapomorphy at a much higher level (see rosids et al. above). Polarity of leaf venation is unclear, since Combretocarpus is sister to other Anisophylleaceae and Corynocarpaceae are sister to Coriaraceae; the first member of both pairs has palmate venation, and these are the two clades basal to the rest of the order.

More or less lacinate petals (and staminodes) are common in the order (Endress & Matthews 2006b). The thickness of the outer integument varies considerably. Matthews and Endress (2004, summarized in 2006b) provide an excellent survey of floral morphology of the group.

Phylogeny. For the circumscription of the clade, rather unexpected as regards families like Coriariaceae and Anisophylleaceae in particular, see e.g. Setoguchi et al. (1999) and Schwarzbach and Ricklefs (2000). Coriariaceae and Corynocarpaceae were sister taxa, and their wood anatomy is very similar, consistent with such a position (Carlquist & Miller 2001). However, other relationships remained poorly understood (Brouillet 2001 for some comments), although Zhang and Renner (2003a), using a variety of both chloroplast and nuclear genes, suggested that Anisophylleaceae were sister to the rest of the clade. Recently, Zhang et al. (2006) in a nine-gene study (all three compartments) confirm these relationships, and also placed Cucurbitaceae as sister to [Tetramelaceae + Datiscaceae + Begoniaceae], although relationships within this latter group were still not entirely clear, the clade [Datiscaceae + Begoniaceae] having at best only moderate support (see also Schaefer et al. 2009; Schaefer & Renner 2011; cf. Soltis et al. 2007a). For relationships of Tetramelaceae and Datiscaceae, see Swensen et al. (1994, 1998). The tree here follows that of Zhang et al. (2006) and Schaefer and Renner (2011); the former also discuss aspects of morphological evolution and evaluate the extensive variation in breeding system in the clade. See Matthews and Endress (2004, summary in 2006b) for the floral evolution of the group.

The relationships of the holoparasitic Apodanthaceae were for some time unclear (e.g. they were unplaced in A.P.G. 2009). Nickrent et al. (2004) suggested relationships either within Malvales (especially the three-gene analyses and that of nuclear SSU rDNA), or in or near Cucurbitales (analysis of matR). Barkman et al. (2007: support weak, but rather comprehensive analysis) also suggested the latter position; the mitochondral genes cox1 and matR showed massive divergence, but not the atp1 gene (Barkman et al. 2007). Additional molecular analyses (D. Nickrent, pers. comm.; esp. Filipowicz & Renner 2010) support the position of Apodanthaceae in Cucurbitales, and this is consistent with their dioecy, extrose anthers, inferior ovary and parietal placentation, all features common in Cucurbitales (see also Filipowicz & Renner 2010), but all these features are generally common in parasitic plants (Renner & Ricklefs 1995). There are also a number of codon subsitutions in common between Apodanthaceae and Cucurbitales (Barkman et al. 2007; Filipowicz & Renner 2010). The exact position of the family in Cucurbitales remains unclear, the relationships suggested with the morphologically rather different (but apomorphically so) Corynocarpaceae and Coriariaceae being only weakly supported, and Apodanthaceae are on a very long branch (Filipowicz & Renner 2010).

Previous Relationships. Coriariaceae were previously often placed with other families with separate carpels and so thought to be "primitive". Although Rhizophoraceae and Anisophylleaceae were often associated in the twentieth century, e.g. being placed in separate but adjacent orders, as in Takhtajan (1997), or even in the same family, others thought that they were not so close. Thus although both were placed in Rosidae by Cronquist (1981), they were not adjacent. Indeed, morphological differences between the two are marked (e.g. Juncosa and Tomlinson 1988; Tomlinson 1988), and Rhizophoraceae are now securely placed in Malpighiales where they are sister to Erythroxylaceae, with which they have much in common. Similarities in floral morphology between Anisophyllea and Ceratopetalum (Cunoniaceae - Oxalidales: see Matthews et al. 2001; Endress & Matthews 2006b), although striking, are unlikely to be evidence of immediate close relationships of the two, even although the fossil Platydiscus peltatus seems to suggest similar relationships (Schönenberger et al. 2001a; see also Schönenberger & von Balthazar 2006). Apodanthaceae were often previously included in Rafflesiaceae s.l.

Includes Anisophylleaceae, Apodanthaceae, Begoniaceae, Coriariaceae, Corynocarpaceae, Cucurbitaceae, Datiscaceae, Tetramelaceae.

Synonymy: Anisophylleales Reveal & Doweld, Begoniales Link, Coriariales Lindley, Corynocarpales Takhtajan, Datiscales Dumortier

ANISOPHYLLEACEAE Ridley   Back to Cucurbitales

Trees and shrubs; plants Al-accumulators; cork?; cambium storying?; nodes 1:1; cuticle waxes as platelets; stomata usu. paracytic; branching from current growth, rythmic; lamina margins entire, (stipules 2-4, minute, very base of the petiole [= colleters?]); inflorescence paniculate, racemose or spicate; plant monoecious; flowers small; K epidermis with mucilaginous inner walls, postgenitally coherent, C open (0), lobed or laciniate (entire), bundle number?, ± enclosing groups of A; A 2x K, obdiplostemonous, incurved in bud; nectary of separate lobes; G (stylulus hollow), (roof over ovary, styuli marginal), compitum 0, stigma expanded or punctate; ovules often unitegmic, epitropous, outer integument 7-9 cells across, parietal tissue 1(?+) cells across, nucellar cap +; fruit drupe or samara, (K accrescent); embryo fusiform, largely hypocotylar; n = 7, 8; germination hypogeal.

4[list]/34: Anisophyllea (30). Pantropical (map: see Clement et al. 2004). [Photo - Anisophyllea Fruit]

1. Combretocarpus

Leaves with 2ndary veins pinnate; flowers perfect, 3(-4) merous; nectary lobes intrastaminal only; ovule 1/carpel; embryo sac bisporic, 8-nucleate [Allium-type]; cotyledons small.

2. The Rest.

cuticle waxes beaker-like - Polygonanthus; (leaves anisophyllous, appearing distichous, base asymmetrical - Anisophyllea); flowers imperfect (perfect), 4(-5) merous; nectary lobes both inter- and intrastaminal; ovules 2/carpel, testa multiplicative, (vascularized), 10-30 cells thick, (mesotesta also lignified); cotyledons indistinct.

• Anisophylleaceae are a rather undistinguished-looking family. The plant may be anisophyllous, the leaves being rather coriaceous, often yellowish drying, entire, and often more or less asymmetrical at the base. The flowers are small and the ovary is inferior.

Evolution. Divergence & Distribution. Crown-group diversification in Anisophylleaceae may have begun (107-)85(-67) million years ago (Zhang et al. 2007).

Chemistry, Morphology, etc. For further information, see Vincent and Tomlinson (1983: Anisophyllea architecture), Tobe and Raven (1987e, 1988a, c: floral morphology, embryology, fruit), Dahlgren (1988), Schönenberger et al. (2001a: fossils), Matthews et al. (2004: floral development), Zhang et al. (2007: phylogeny) and Schwarzbach and Tomlinson (2011: general).

Phylogeny. Combretocarpus is sister to the rest of the family (Zhang et al. 2007), somewhat in conflict with morphology-based relationships (e.g. Tobe & Raven 1987c, 1988c); this may have an important effect on the polarity of characters in the family or even order as a whole, so the genus is treated separately above.

Previous Relationships. Anisophylleaceae were often linked with or included in Rhizophoraceae (Malpighiales: see above). However, the inner integument is ca 2 cell layers thick, there are no laticifers, the petals are not aristate, and a sclerified exotegmen is absent; in these and many other characters Anisophylleaceae differ from Rhizophoraceae (see Juncosa & Tomlinson 1988a, b).

Synonymy: Polygonanthaceae Croizat

[[Corynocarpaceae + Coriariaceae] [Cucurbitaceae [Tetramelaceae [Datiscaceae + Begoniaceae]]]]: uniseriate rays 0; filaments shorter than anthers in bud, anthers basifixed; disc nectaries 0; styles separate

[Corynocarpaceae + Coriariaceae]: ellagic acid +; wood with broad rays; sieve tube plastids lacking both starch and protein inclusions; stomata paracytic; lamina margins entire; flowers small, K quincuncial, C thick, base broad; G superior, ascidiate; ovule 1/carpel, vascular bundle extending into the outer integument; cotyledons very large.

Chemistry, Morphology, etc. For sieve tube plastids, see Behnke (1981c).

CORYNOCARPACEAE Engler, nom. cons.   Back to Cucurbitales

Trees; young stem with separate bundles; petiole bundles forming line; lamina vernation conduplicate, 2ndary veins pinnate, stipule single, intrapetiolar; inflorescence paniculate; calyx and corolla distinct, C with a single bundle; stamens = opposite and basally adnate to C, incurved in bud, staminodes 5, fringed, petaloid, opposite sepals, with a basal, adaxial nectary; pollen heteropolar, dicolpate, psilate, infratectum granular; G [2], transverse?, only 1 fertile, stylulus short, usu. single, conduplicate, stigma capitate, dry; ovule with outer integument ca 11[?-30] layers thick; fruit a drupe, stylulus excentric; seed coat ?pachychalazal, initially thick, vascularized, becoming crushed; endosperm starchy; n = 22, 23.

1[list]/6. New Guinea to New Zealand, introduced on Hawaii (map: from van Steenis & van Balgooy 1966; George 1984). [Photo - Flower] [Photo - Fruit]

• Corynocarpaceae are best recognised by dissecting their flowers. These are quite distinctive, although small: the stamens are opposite the petals and alternate with fringed staminodes, and the ovary usually has but a single style. Vegetatively the family is rather less distinctive, having spirally-arranged leaves with entire margins and intrapetiolar stipules.

Chemistry, Morphology, etc. The cork develops from the cell layer beneath the epidermis. It is perhaps unclear whether the gynoecium is pseudomonomerous or unicarpellate. However, since some flowers have two styluli (e.g. Matthews & Endress 2004), the single, excentrically-placed structure at the apex of the gynoecium is called a stylulus and the gynoecium is pseudomonomerous {sic.]. The seeds are very poisonous, having bitter glucosides.

Some information is taken from Hemsley (1903: general), Nowicke and Skvarla (1983: pollen), Philipson (1987a: general), and Kubitzki (2011: general).

Previous Relationships. Corynocarpaceae are Celastralean in affinity according to Cronquist (1981), isolated, according to Takhtajan (1997), so here they are!

CORIARIACEAE Candolle, nom. cons.   Back to Cucurbitales

Usu. shrubs; roots with N-fixing Frankia; coriolic fatty acid [CH₃(CH₂)₄CH(OH)CH=CHCH=CH(CH₂)₇COOH] in seed, sesquiterpenes, myricetin +; perforation plates bordered; nodes 1:1; petiole bundle arcuate; buds usu. perulate; leaves opposite, lamina vernation ± flat, stipules small; plant polygamous or flowers perfect, inflorescences racemes, bracteoles 0; (flowers 6-merous), K quincuncial, C open, fleshy, often keeled adaxially; A 10, connective thin, septa between sporangia of theca not developed; tapetal cells 2-4-nucleate, pollen (2 colpate), starchy, (3-nucleate); G [5 (10)], opposite K, [(10)], stylulus slender, stigmatic all around, dry; ovule apotropous, micropyle endostomal, outer integument 3-4 cells across, inner integument 2-3 cells across, parietal tissue ca 8 cells across, nucellar cap ca 4 cells across; fruit achenes (nutlets), several, surrounded by accrescent C; exotesta of cuboid, "tanniniferous", thick-walled, lignified(?) cells, rest undistinguished; n = 10, 15.

1[list]/5. Very disjunct: circum S. Pacific to China and Himalayas, Mediterranean (map: from van Steenis & van Balgooy 1966; Good 1974). [Photo - Inflorescence] [Photo - Fruit]

• Coriariaceae can be recognised by their opposite, sessile, entire leaves with veins coming from the base and at most minute stipules; many branches look like compound leaves, apparently being of limited growth. The inflorescences are racemes and the flowers have a poorly differentiated, biseriate perianth and carpels that are connate at most only basally; the fruits are achenes that are surrounded by the accrescent corolla and have long, subbasal styles that remain attached.

Evolution. Divergence & Distribution. Fossils of Coriaria are known from about 33 million years before present (Saporta 1965).

Chemistry, Morphology, etc. Vessels are in multiples, there are true tracheids, and the wood parenchyma is (confluent) vasicentric. A compitum appears to be developed (Matthews & Endress 2004). Although there is but a single seed/carpel, several seeds are produced by each flower.

Information on nodal anatomy is taken from Sinnott (1914), on embryology from Sharma (1968a), and on wood anatomy from Yoda and Suzuki (1992).

For gynoecial development, see Guédès (1971), and for general information, see Kubitzki et al. (2011).

Phylogeny. For phylogenetic relationships within the family, see Yokoyama et al. (2000); the Eurasian clade is sister to the rest.

Previous relationships. Coriariaceae were placed in Ranunculales by Cronquist (1981) and as a monotypic Coriariales in Rosidae (Takhtajan 1997), largely because of their apparently separate carpels.

[Cucurbitaceae [Tetramelaceae [Datiscaceae + Begoniaceae]]]: perennial herbs; cucurbitacins [triterpenes] +, myricetin, ellagic acid 0; young stem with separate bundles; leaves with teeth, medial vein ending in a pad of packed translucent cells, lateral also entering [in Begonia lateral is dominant], stipules 0; flowers imperfect; G opposite sepals or median member adaxial, placentation parietal, a roof over the ovary, so styles marginal, stigmas large, elongated, bilobed; many ovules/carpel; seeds many.

Chemistry, Morphology, etc. For information on leaf teeth, see Hickey and Wolfe (1975). The development of a roof over the ovary formed from tissue adaxial to the stylulus (Matthews & Endress 2004) is obvious when well developed; the styluli are then widely separate and borne towards the margin of the ovary. Matthews and Endress (2006) note details of ovule morphology that this group has in common.

CUCURBITACEAE Jussieu, nom. cons.   Back to Cucurbitales

Lianes or perennial to annual vines (cauduciform [hypocotylar] succulents); (silicon concentration high); tendrils cauline, ± lateral, both branches and stem coiling; alkaloids, bitter tetra- and pentacyclic triterpenoids, punicic acid [C₁₈H₃₀O₂], citrullin [non-protein amino acid - alpha-amino-delta-ureidopentanoic acid], saponins +, little oxalate accumulation, tannins 0; root cork superficial; stem cork variable in origin; cambium storying?; extra-fascicular phloem +; vascular bundles initially separate, in two rings [outer opposite the angles of the stem]; rays multiseriate; outer collenchyma and sclerenchymatous sheath in cortex; petiole bundle arcuate, or ring of arcuate bundles; no pericyclic sheath; (cuticle waxes as platelets); indumentum rough hairy/prickly, walls calcified, cystoliths + (0), hairs often glandular; leaves often with extrafloral nectaries, (lamina margins entire); plants dioecious (monoecious), inflorescences axillary; flowers ebracteolate or not, (3-)5(-7)-merous; hypanthium + (0; tube formed by adnation of K and C), short, K often connate, open, (0), C (induplicate-)valvate, connate; staminate flowers: nectary type?; A extrorse, variously connate [and forming a central column] or free [and well apart on the hypanthium], anthers monothecal; pollen grains prolate, to 40 µm long, (micro)striate, starchy; pistillode 0; carpellate flowers: staminodes +; G 1 [(2) 3(-5)], inferior, (median member abaxial), nectary on the ovary, hypanthium base as nectariferous trichomes, placentae intrusive, stigmas dry or wet, (channelled; not bilobed); ovules (1-few/carpel), pendulous, (micropyle endostomal; naked), outer integument 4-8 cells across, vascularized, nucellar cap +, nucellar beak +; fruit a capsule (?type); seeds often flattened, pitted, testa multiplicative, complex, tegmen ± persistent, outer cells ± tracheidal; endosperm 0, chalazal haustorium + (0), cotyledons large, flat; germination epigeal (hypogeal - Momordica, etc.), seedlings with a peg [cortical outgrowth at the root-shoot transition].

97[list]/960 - two groups below. Largely tropical and subtropical, especially drier parts of Africa (map: from Heywood 1978 [N. part of range]; Saade 1998; Florabase 2006). [Photos - Collection, Staminate flower, Carpellate flower, Fruit.]

1. "Fevilleoideae" Burnett - note - this will be disposed of soon, but I am in the process of integrating the characters below with the first few tribes as recognised by Schaefer and Renner (2011).

Axillary bud also present; (tendril unbranched); styluli +, marginal; fruit obconical, opening apically; seed winged or not; epidermis unremarkable, hypodermal tissue ± thick-walled and lignified, inner sclerenchymatous layer several cells across, brachysclereidal, thickened, arenchyma with banded thickenings, inner layer +.

1. Gomphogyneae Bentham & J. D. Hooker

(Tendrils with adhesive pads); nectary of multicellular hairs on C; A 5, 1-thecal, 3, 2- or both 2- and 1-thecal, inserted at base of C and connate to middle of C; (pollen perforate-rugulate - Alsomitra); (fruit a berry); n = 11, 13.

6/56: Hemsleya (30). China and South East Asia to Australia and Fiji.

2. Triceratieae A. Richard

?Nectary; A also 1-3, variously 1 or 2-thecal, inserted at base of C; (pollen larger, reticulate - Gerrardanthus); female flowers: staminodia 5, (styles central); fruit opening variously, or 1(-3)seeeded samara or baccate; n = ?

5/24. Mostly tropical, New World (E. and S. Africa and Madagascar - Cyclanthopseris).

Synonymy: Nhandirobaceae Lestibudois

3. Zanonieae Bentham & J. D. Hooker

(Stalk of tendril coiled); (flowers weakly monosymmetric); nectary of multicellular hairs on C; A 5 (4), 1-thecal, inserted at base of C to middle of tube; (pollen reticulate); n = ?

4/12. Pantropical.

Synonymy: Zanoniaceae Dumortier

4. Actinostemma Griffith

Nectary of multicellular hairs on C; A inserted on base of tube, 5, or 2 pairs + 1, 1-thecal; style single; ovule pendulous; fruit opening ?, seeds winged or not; n = 8.

1/3. Central and East Asia.

5. The rest

Often annuals; additional non-protein amino acids +; tendrils complex, branches alone coiled [not Thladianthineae]; young stem with bicollateral vascular bundles; extrafascicular phloem system linked with adaxial phloem of bundles; staminate inflorescence + carpellate flower + bud + tendril making up axillary complex; plant monoecious; nectary parenchymatous, with stomata, (of multicellular hairs - Sicyoeae); staminate flowers: hypanthium well developed (short); A 5 [Luffa] or fewer, ?inserted, anthers often much bent and coiled, (locellate); pollen grains ± spherical, 40-70(-200: Cucurbiteae) µm long, ± spherical, reticulate, (colpate and [panto]porate; operculate; echinate); pistillate flowers: style single, stylar canal +, filled with secretion, shortly branched or not; ovules horizontal (to erect), (micropyle endostomal), outer integument 6-10 cells across, inner integument (1-)2-5(-ca 6 - Bryonia, Sechium) cells across, parietal tissue 5-11 cells across, nucellar beak; antipodals degenerate; fruit ± fleshy, (irregularly dehiscent); seed not winged; exotesta enlarged, ± palisade or cubic, mucilaginous, hypodermal tissue various, sclereid layer sharply distinguished, single cell across, much enlarged, elongated or not, walls much thickened, aerenchyma usu. unthickened, innermost layer chlorenchymatous; n = 9, 11, 12, 14 (etc.).

82/880: Trichosanthes (100), Sicyos (75), Momordica (60), Zehneria (60), Cucumis (55), Cayaponia (55), Cyclanthera (40), Gurania (37). Tropical to warm temperate.

Synonymy: Bryoniaceae G. Meyer, Cyclantheraceae Lilja

• Cucurbitaceae are easily recognised. They are vines or lianes with often rather roughly hairy exstipulate leaves that have palmate venation. In the leaf axil there is a usually lateral and branched tendril, a vegetative bud, a flower, or a yet more complex organisation. The plants are dioecious or monoecious, and the flowers have a hypanthium, an often complex androecium made up of twisted, monothecal anthers, and an inferior ovary with parietal placentation. The corolla is often more or less yellow, sometimes red. The seeds are more or less flattened.

Evolution. Divergence & Distribution. Schaefer et al. (2009) suggest that stem Cucurbitaceae are some (69-)63(-61) million years old, i.e. late Cretaceous, with the current world-wide range of the family being in large part the result of extensive dispersal, Madagascar being colonized an estimated thirteen times and Australia twelve times - and the latter currently has only twelve genera and thirty species. Interestingly, the woody Socotran endemic Dendrosicyos is dated to (30-)22(-14) million years, although Socotra itself is only about ten million years old, which suggests that the clade now represented by just the single species - woody, but secondarily so, of course - was once on the mainland and has since become extinct there (Schaefer et al. 2009). For more on dispersal in the family, see Duchen and Renner (2010).

Plant-Animal Interactions. Low concentrations of the very bitter cucurbitacins, tetracyclic sterol-like triterpenes that are among the most bitter substances known to humans, elicit a compulsive feeding response from luperine beetles, including rootworm leaf beetles (Chrysomelidae: Galerucinae: Luperini: see Metcalf et al. 1980; Jolivet & Hawkeswood 1995); Gillespie et al. (2008) outline their phylogeny. 80% of the host plant records of the some 4,000 species of the group are from this family, and many species are pharmacophagous. That is, adults visit the flowers, feeding on pollen, and sometimes other parts of the plant, and they sequester these bitter cucurbitacins (Eben 1999; Tallamy et al. 2005). The larvae of some galerucines do feed on Cucurbitaceae, and this ability may have evolved independently in Old and New World members, Aulacophorina and Diabroticina respectively (Gillespie et al. 2003). Beetles are attracted by volatiles coming both from flowers and other parts of the plant (Andrews et al. 2007 for references). The larvae sometimes cut leaf veins (they "trench" the leaves), so locally interrupting the translocation of cucurbitacins to the leaf tissue and so apparently allowing the insect to eat it (Dussourd & Eisner 1987). However, given that at least some of these beetles will eat cucurbitacin crystals, physical avoidance of the copious sap produced by Cucurbitaceae is a more likely explanation of this feeding behaviour. Indeed, the concentration of cucurbitacins inside and outside the trench is about the same, but the sap is very rich in P-protein and eventually gels and so would probably thoroughly gum up the mouth parts, etc., of the beetles if they ate untrenched leaves (McCloud et al. 1995).

The glandular hairs of Cucurbitaceae are also involved in the defence of the plant against herbivores, quickly-solidifying secretions being produced when the trichomes are touched by insects (Kellogg et al. 2002). Discharge of the sap may be explosive, Zimmermann (1922) mentioning what he called "Explosionshaare" in the family.

Larvae of a remarkable number of different Blepharoneura species (Diptera - Tephritidae [fruit flies]) are being discovered in flowers and fruits of neotropical Cucurbitoideae like Gurania. Some species are specific to staminate flowers, others to carpellate, and all told some 52 species of flies were found on 24 species of cucurbits (Condon et al. 2008; Steele 2010 for Psiguria).

Butterfly caterpillars are not often found on members of this family (Ehrlich & Raven 1964).

Ecology & Physiology. The phloem transport system in Cucurbitaceae is very complex. From early observations made by Fischer (summarized in Fischer 1884), bicollateral vascular bundles appear to be an apomorphy of only part of the family (see above). In taxa with bicollateral vascular bundles there are also extrafascicular phloem strands that are found in the cortex outside the sclerenchymatous ring, although they are barely developed beyond the ring in taxa like Thladiantha and Momordica (Fischer 1884). These extrafascicular strands are reported to link with the adaxial phloem cells of the vascular bundles (Schmitz et al. 1987), but only at the nodes (Fischer 1883), and in taxa like Cucurbita phloem strands permeate the cortex, even occuring in the collenchyma; commissural strands are common (Fischer 1884). The sieve tubes of this extrafascicular phloem system differ in morphology from those of the fascicular phloem, although they may be similar to cells in the peripheral part of the latter (Crafts 1932). When the plant is damaged, the copious phloem exudate comes largely from this extrafascicular system, flow from the bundle phloem stopping almost immediately, and the composition of this exudate is very different from that of the fascicular phloem (Zhang et al. 2010). The latter is rich in sugars and unidentified proteins, etc., the former contains P-proteins, amino acids, and various secondary metabolites, but little sugar (Zhang et al. 2010). Interestingly, aphids feed on the abaxial phloem, and autoradiograms of minor veins suggest that abaxial phloem cells here are the sole conduits for carbon export and import, indeed, in the very finest veins there is no adaxial phloem system at all, i.e. the bundles there are collateral (Botha & Evert 1978; Schmitz et al. 1987; Hebeler 2000 and references). More needs to be done to understand the complex vascular anatomy of the family by improving sampling, expecially in the basal clades, to work out just how the extrafascicular system is involved in plant defence, as it probably is (Turgeon & Oparka 2010), and how it links with the adaxial intrafascicular system. Note that much of the work that has been done on Cucurbitaceae phloem and phloem transport - because of the copious exudate, this has been a much-studied system - actually has been carried out on the extrafascicular system (Zhang et al. 2010)! Blyth (1958) integrate.

Floral Biology & Seed Dispersal. Breeding systems in Cucuritaceae can be very labile (Schaefer & Renner 2010). There are a number of interesting pollinator-plant interactions in the family. Heliconius butterflies, whose caterpillars eat Passiflora vines (see Passifloraceae), are also closely associated with Psiguria (Cucurbitaceae), which they pollinate while at the same time obtaining nutrients from the pollen, probably by enzymatic activity of the saliva; the butterflies also visit some species of the closely related Gurania (e.g. Gilbert 1972, 1975; Boggs et al. 1981; Spencer 1988; Eberhard et al. 2009: see Steele 2010 for a summary and Steele et al. 2010 for a phylogeny of Psiguria). Interestingly, both Psiguria and Gurania have pollen grains in tetrads, alone in the family (Steele 2010). Psiguria in particular has very long-lived inflorescences, and in the staminate phase flowers are produced continuously for months or more. Other Cucurbitaceae, notably Momordica and Thladiantha, are oil flowers, the ca 19 extant species of Ctenoplectrini bees being associated with Cucurbitaceae and collecting material from the oil-secreting hairs as well as pollen and nectar (Buchmann 1987; Vogel 1990 for details). Ctenoplectrini, an Old World group, probably sister to the Eucerini, may have diverged in the early Eocene ca 50 million years ago (Schaefer & Renner 2008b). Ctenoplectra bees take pollen, oil and nectar from the flowers of Momordica, although female flowers in some species may lack rewards (Schaefer & Renner 2010).

Squash and gourd bees, some 20 species of the genera Peponapis and Xenoglossa, pollinate only species of Cucurbita. They feed early in the day, even flying in the dark pre-dawn hours (Hurd et al. 1971). They are attracted by particular floral volatiles, repelled by others which attract the cucumber beetles (yet others seem to attract both herbivore and pollinator: Andrews et al. 2007). The bees, restricted to the Americas north of northern Peru, show some species-specific variation in pollen collecting devices (Hurd & Linsley 1964: obviously Cucurbita can be pollinated by a variety of other bees since it is cultivated pretty much world wide). Within Cayaponia there have been shifts from bat to bee pollination, rather unusual in flowering plants (Duchen & Renner 2010). Finally, a number of taxa have elaborately fringed margins of their corolla lobes, while Momordica anigosantha has quite strongly monosymmetric staminate flowers, both in form and especially colour patterning, yet the female flowers are less remarkable (Zimmermann 1922).

Vegetative Variation. Cucurbitaceae are very largely a group of climbing plants, whether lianes or vines, and are a prominent component of this vegetation element in the New World (Gentry 1991). The tendrils of Cucurbitaceae are branched, and represent a branch complex. The length of the unbranched part of the tendril varies considerably, and the tendril may be ad- or abaxially curved in bud (Zimmermann 1922). Often there is a sublateral tendril + bud + slightly lateral flower associated with each leaf, or a tendril + vegetative bud + carpellate flower + staminate inflorescence, all more or less collaterally arranged, or other variants. Eichler (1875) and Goebel (1932) suggested that the tendril branches were prophylls, and in Bryonia dioica paired tendrils occur on the pedicel of an axillary flower (see also Zitnak et al. 2010). Non-flowering Zanonioideae have tendrils more or less lateral to vegetative axillary buds. When flowering finally begins, tendrils are replaced by flowers, which are more or less adaxial to the axillary bud. This bud produces an inflorescence branch that has an internode below the prophyllar leaf that subtends the first flower. Most Cucurbitoideae lack an initial prolonged vegetative period, and in one interpretation the inflorescence branch lacks a basal internode, so the first flower, often carpellate, arises in the leaf axil of the main branch and is subtended by a prophyll; the staminate inflorescence represents the development of this prophyllar bud (Lassnig 1997, for details of branching; axillary structures may be collateral in the vegetative part of the plant, sometimes superposed in the reproductive part). However, Gerrath et al. (2008) found that in Echinocystis lobata tendril, axillary bud, carpellate flower, and staminate inflorescence were all more or less independent in origin, although the latter two did arise from a common primordium (see also Zitnak et al. 2010). Joliffieae may be critical in understanding the evolution of the branch-tendril complex. Acanthosicyos has paired thorns at the nodes; I do not know anything about their development.

Economic Importance. Cucurbitaceae were particularly important in early agriculture in the Americas, being one of the triumvirate of squash, corn and beans. For discussion of various aspects of the history of cultivation of Lagenaria and Cucurbita in particular, see Teppner (2004). For the domestication of squash (Cucurbita spp., inc. C. moschata and C. agyrosperma) which began ca 10,000 years ago, see Dillehay et al. (2007), Piperno et al. (2009) and Ranere et al. (2009); for phytoliths of the family, see Piperno (2006). Sebastian et al. (2010) suggest that the relatives of cucumber and melon (Cucumis) are Asian-Australian.

Chemistry, Morphology, etc. Cucurbitaceae produce phytoalexins only with difficulty (Harborne 1999). Raffinose is the main transport carbohydrate (Turgeon & Ayre 2005). Distinctive long-chain fatty acids occur in the seed oils; eleostearic acid, an isomer of punicic acid, is restricted to Joliffieae (Hopkins 1990). There are crystalloid inclusions in the protein bodies found in embryos of this family, perhaps unusual for flowering plants (Lott 1981).

A number of African Cucurbitaceae have swollen stem bases. Vascularisation of the leaf is complex, e.g. the leaf being supplied by one of the outer ring of bundles in its entirety and by branches from two other bundles of the outer ring, the bud being supplied by the inner ring (Sensarma 1955). Indeed, as the literature summarized by Sensarma (1955) suggests, there has been much speculation about the nature of the cauline vascular system and, based on how leaves are supplied from the cauline system, whether from the inner or inner + outer rings of bundles, and, related to this, on the presumed cauline versus foliar nature of the tendrils.

The petals of the woody, succulent-leaved vine, Xerosicyos, are free; those of Echinocystis and Lagenaria at least have several traces. Most, but not all ex-Cucurbitoideae ("the rest" above) have a disc-like nectary that may even be covered by a flap of tissue, while the nectary hairs of ex-Fevilleoideae are less localized and are borne on the petals (Vogel 1981b; 1997). Note that the flowers in taxa in which the androecium has two pairs of stamens and a single stamen, four stamens and a staminode (Gerrardanthus), etc., are strictly speaking monosymmetric. When the stamens are connate 2 + 2 + 1, the vascular supply shows evidence of this, although there are differences over the interpretation of the apparently bithecal stamens (e.g. de Wilde & Duyfjes 1999); Schaefer and Renner (2011) suggest that the plesiomorphic condition of the family is to have five bithecal stamens. Several tubes may arise from the one pollen grain (the pollen is polysiphonous), and some Cucurbiteae have very large almost spherical grains up to 200 µm or so long and across. The carpellate flower may have two rings of rudimentary anthers, while in the staminate flowers a ring of processes may alternate with the stamens. The chalazal haustorium of Sechium [= Sicyos] edule, at up to 19,000 µm long, is apparently the longest in the family, although others are also quite long; only Santalales have longer embryo sacs (Mikesell 1990; Johri et al. 1992). Seedlings commonly have a peg, a cortical outgrowth towards the bottom of the seedling axis at the root-shoot transition (e.g. Klebs 1884 for a list of taxa); this is not found in species in which germination is hypogeal (Zimmermann 1922).

In Cucumis, at least, mitochondria (but not chloroplasts) are transmitted paternally (Havey et al. 1998). The mitochondrial genome is very variable in size, from ca 379,000-2,900,000 bp long, the shorter sequences, at least, having expanded by the acquisition of chloroplast sequences and the accumulation of numerous short repeats (Alverson et al. 2010).

For general information, see Jeffrey (1980), Bates et al. (1990), Jeffrey and de Wilde (2006) and especially Schaefer and Renner (2011a), for non-protein amino acids, see Fowden (1990), for cork cambium, see Dittmer and Roser (1963), for wood anatomy and secondary thickening, the latter odd, see Carlquist (1992c) and Patil et al. (2011), for seed coat anatomy, which is complex, Kratzer (1918), B. Singh (1972), D. Singh and Dathan (1973, 1974, 1998, 2001) and Teppner (2004), for embryology, etc., Kirkwood (1905), Warming (1913), Chopra (1955), Johri and Roy Chowdhury (1957), and Singh (1970), for floral morphology, etc., see Leins and Galle (1971) and Leins and Erbar (2010), for pollen, see Van der Ham et al. (2010: reticulate pollen derived?), and for general information, see .

Phylogeny. Renner et al. (2002) suggested that Cucurbitoideae were probably monophyletic, with Thladiantha possibly sister to the rest; Fevilleoideae (Zanonioideae) formed an unresolved basal polytomy. This was largely confirmed by Kocyan et al. (2007), although Indofevillea was sister to other Cucurbitoideae; monophyly of Fevilleoideae was not well suppported, Alsomitra sometimes appearing as sister to Cucurbitoideae. Schaeffer and Renner (2008) also found Fevilleoideae to be paraphyletic and did not recognise the subfamily. Indeed, Schaefer et al. (2009) found a grade of four clades basal to a well-supported Cucurbitoideae, in which Indofevilleeae were well supported as being sister to the rest. This basal grade included Gomphogyneae, with only moderate support, Alsomitra being sister to the rest, a strongly supported Fevilleeae, Zanonieae, with 76% ML bootstrap, and a strongly supported Actinostemmateae; relationships between these clades had no support (see also Schaefer & Renner 2011). Fevilleoideae are not recognised here, but establishment of the phylogenetic relationships within the basal polytomy is important since it will considerably affect optimisation of characters on the tree.

Within Cucurbitoideae, phylogenetic studies suggest that Jeffrey's tribes (Jeffrey 2005) are largely monophyletic, although his subtribes are not (Kocyan et al. 2007). Jobst et al. (1998: ITS) found Benincaseae (Cucurbitoideae) to be polyphyletic; Chung et al. (2003) and Schaefer et al. (2008, esp. 2009: Alsomitra in Zanonieae) also looked at relationships within Cucurbitoideae. For smaller-scale studies within Cucurbitoideae, see Ghebretinsae et al. (2007), Wilde and Duyfjes (2006), Renner et al. (2007a), Schaefer et al. (2009), Schaefer and Renner (2010), Duchen and Renner (2010) and Sebastian et al. (2010: Cucumis).

Classification. For the suprageneric classification I follow Schaefer and Renner (2008a, esp. 2011b); in the latter there is provide a more comprehensive tribal classification which should be consulted for details (cf. Jeffrey 2005). There are many small genera in Cucurbitaceae, and generic limits need attention, thus Ghebretinsae et al. (2007) had to adjust the limits of Cucumis; see also Wilde and Duyfjes (2006), Renner et al. (2007a) and Schaefer et al. (2009).

Previous Relationships. Cucurbitaceae have usually been associated with other families that have parietal placentation, whether placed all together in Violales (Cronquist 1981) or in a group of small orders placed next to each other in Dilleniidae (Takhtajan 1997).

Tetramelaceae [Datiscaceae + Begoniaceae]]: pollen spherical, stigmas elongated; fruit a septicidal capsule [dehiscing apically]; seed with operculum; exotestal cells honeycomb [Clement et al. 2004], inner walls strongly thickened and lignified; cotyledons moderate in size.

Chemistry, Morphology, etc. Tebbitt (2005) suggests that the seeds of this group have a lid, but whether this is a synapomorphy or not is unclear. Seeds of Tetramelaceae are apparently unknown, and Boesewinkel (1984) found that the opercula of Datiscaceae and Begoniaceae were rather different. See Mauritzon (1936b) for some details of the ovules.

TETRAMELACEAE Airy Shaw   Back to Cucurbitales

Trees; tannin 0; (wood fluorescing); (nodes with 2 traces from the lateral gaps); hairs glandular or lepidote; (lamina margins entire); plant dioecious, inflorescence spicate; K 4-8, postgenitally coherent, staminate flowers: C 0, or 6-8 [Octomeles], stamens = and opposite petals, incurved; carpellate flowers: C 0; G [3-8], (disc on top), placentation axile, placentae bilobed [Octameles], stigmas undivided, decurrent to clavate; fruit also opening down the sides; seed coat?; n = ca 23.

2[list]/2. Indo-Malesia (map: from van Steenis 1953). [Photo - Tree]

Evolution. Divergence & Distribution. Tetrameles wood is known fossil from the Deccan Traps in India ca 70.6-65.5 million years old (Zhang et al. 2007).

Chemistry, Morphology, etc. Octomeles has sclereids; its capsular fruits split into two layers, the outer of which falls off. For general information, see Swensen and Kubitzki (2011, in Datiscaceae).

[Datiscaceae + Begoniaceae]: herbs; outer and inner integuments 2 cells across.

DATISCACEAE Dumortier, nom. cons.   Back to Cucurbitales

Roots with N-fixing Frankia; cambium not storied; medullary bundles +; tannin sacs +; nodes 1:3; leaves deeply divided to odd-pinnate, lamina vernation conduplicate, 2ndary veins ± pinnate; plant (andro)dioecious, inflorescence fasciculate; P 4-10; staminate flowers: P valvate; A 6-25, outer members opposite P, filaments very short, pistillode 0; carpellate flowers: staminode 0, G [3-8], opposite P; ovules with parietal tissue 3-5 cells across, cap 2-3 cells across; embryo sac bisporic, 8-nucleate; fruit septicidal?; seeds with lid, exotegmic cells large, cuboid; endosperm slight; n = 11.

1/2. W. North America, Crete to India (map: from Liston et al. 1989; Clement et al. 2004). [Photo - Flower, Flowers.]

Chemistry, Morphology, etc. The lid on the seeds of Datisca is not surrounded by a ring of collar cells (Boesewinkel 1984: cf. Begoniaceae). The stamens show no particular relationship to the calyx (Takhtajan 1997). Much information is taken from Davidson (1973, 1976); see Leins and Bonnery-Brachtendorf (1977) for floral development and Swensen and Kubitzki (2011) for general information.

BEGONIACEAE C. Agardh, nom. cons.   Back to Cucurbitales

Fleshy herbs (scandent; shrubby), often tuberous or rhizomatous; tanniniferous, soluble oxalate accumulation; cork subepidermal; cortical (and medullary) bundles +; vessel elements also with scalariform perforation plates; nodes swollen; petiole bundles annular (central bundles +); no pericyclic sheath; sclereids and uncalcified cystoliths +; stomata anisocytic or with accessory cells in two rings [helicocytic]; hairs diverse, often prominent, flattened, pearl glands + [hairs spherical, multicellular, sessile]; leaves two-ranked (spiral; opposite), (compound), lamina vernation laterally or vertically conduplicate (supervolute-curved [prophylls]), asymmetrical, (margins entire), stipules +, large, cauline-extrapetiolar; plants mon(di)oecious, inflorescence cymose, staminate flowers first produced (racemose, pistillate flowers first produced); K petaloid; staminate flowers: A 3-many, centrifugal (connate), basifixed, (porose), connective enlarged; pollen colpate; carpellate flowers: placentation often axile, placentae large, bilobed, stigmas often twisted; ovules with parietal tissue ca 2 cells across, micropyle zig-zag, endothelium +; seeds minute, with lid and surrounding collar cells.

2[list]/1401: Begonia (1500: artificial hybridisation within Begonia has been extensive). Largely tropical (map: from Tebbitt 2005). [Photo - Flower, Fruit] [Photos - Collection]

1. Hillebrandia

Plant ± tuberous; T 10, 2-whorled, inner [= "C"] very small; A often with branched vasculature; G [5], only partly inferior, placentation axile at base and parietal at top; n = ?

1/1: Hillebrandia sandwicensis. Hawaii.

2. Begonia

Plant rhizomatous (tuberous); (carpellate flowers first produced - Symbegonia group); staminate flowers: T in 2s, 2(3)4(-8); carpellate flowers: P (2-)5(-9); G [(1)2-3(-6)], placentation axile to parietal (placentae not bilobed), styles central; capsule dehiscing loculicidally (and septicidally) down sides, often asymmetrically winged, (fruit a berry); n = 8-21+.

1/1500. Largely tropical, but not Hawaii or the Antipodes.

• Begoniaceae may be recognised by their succulent, herbaceous habit and often two-ranked leaves the blades of which have asymmetrical bases and well developed teeth. The hairs are often large, flattened, or almost setular. The plants are monoecious, the staminate and carpellate flowers often have different numbers of tepals, both stigmas and anthers are bright yellow, and the fruit is often winged.

Evolution. Divergence & Distribution. Hillebrandia, from Hawaii, is sister to Begonia as a whole (Clement et al. 2001; Swensen et al. 2001), and its age has been estimated at 58.5-45 million years (Clement et al. 2004, errata 2005). This causes some biogeographical problems, since either Hillebrandia arrived from some continental area where it is now extinct, or it has been island hopping for 50 million years or more (for the ages of the island chain, see Sharp & Clague 2006 and references). The age for crown group Begonia may be 37.3-23.2 million years (Clement et al. 2004), although other estimates put diversification of the genus as occurring some time from the Eocene to early Oligocene 45-25 million years ago during a period of global cooling (Goodall-Copestake et al. 2009, comprehensive) or considerably later at (31-)24(-18.2) million years ago (95% HPD: Thomas et al. 2011b, focus on Malesia); Hillebrandia may be younger than thought (Renner 2005).

Begonia may have originated in Africa, and sister to the South American and Southeast Asian clades that represent the rest of the genus (whatever the reconstruction) are seasonally adapted species with perennating organs (Goodall-Copesteak et al. 2010). Thomas et al. (2011b) found several invasions of Malesia by Begonia and subsequently generally west-to-east movement; the main clade of Malesian species was made up of members of four sections, including the speciose section Petermannia, with at least 270 species; the age of that whole clade was (16.5-)11.5(-6.6) million years (95% HPD). Sections in Begonia are generally limited to single continents, but the very recently described B. afromigrata, a species known from Thailand and Laos and in a section otherwise known only from Africa, is an excaption (de Wilde et al. 2011).

Hughes and Hollingsworth (2008) suggested that the dearth of widespread species in Begonia is due in part to the low levels of gene flow (found in the few studies on members of the genus that have been carried out) and hence for the propensity of divergence in allopatry. De Wilde et al. (2011) drew attention to the wider distributions of those species of Begonia that had fleshy fruits compared to the narrower distributions of species with dry fruits - despite the fact that these latter species had minute seeds and perhaps might be supposed to disperse easily.

Animal-Plant Interactions. Butterfly caterpillars are not often found on Begoniaceae (Ehrlich & Raven 1964).

Floral Biology & Seed Dispersal. Staminate flowers of Begoniaceae produce pollen, carpellate flowers usually have no reward, but have bright yellow and anther-like stigmas; deceit polination is probably involved (Schemske et al. 1996; de Wilde 2001). There are a few ornithophilous species with nectaries at the base of the styles in carpellate flowers only, others have no reward at all; various levels of deceit/mimicry are again involved (Vogel 1998b; Renner 2006).

Tebbit et al. (2006) looked at the evolution of dispersal mechanisms in the speciose Southeast Asian Begonia; taxa with animal or rain-ballist dispersed predominate in a single clade. De Wilde (2011) discusses possible dispersal mechanisms in detail, and de Wilde et al. (2011) focussed on seed dispersal of fleshy-fruited members of the genus.

Chemistry, Morphology, etc. The basic vegetative morphology of Begonia is interesting. There are various intermediates between trichomes, leaf teeth, and leaf-like appendages on the leaf; some taxa also have epiphyllous inflorescences (Dickinson 1978 for references). The leaf teeth are supplied by several veins.

Begoniaceae are unusual among monoecious taxa with cymose inflorescences in that the carpellate flowers are produced only later and the first flowers produced in the inflorescence are staminate, although the derived Symbegonia group has racemose inflorescences and shows the reverse arrangement (de Wilde 2011).

There are five small orange inner perianth parts (= "petals") in Hillebrandia that are very different from the large white outer perianth members (= "sepals"); the stamens are also orange... (Gauthier & Arros 1963). It has been suggested that the perianth of Begonia is to be compared with the sepals of Hillebrandia (see Gauthier 1959), and also that the petals of Hillebrandia are staminodial (Ronse Decraene & Smets 1990a, comparison with Papaveraceae), which they are in colour but not in position. The plesiomorphic tepal number of Begonia may be four in staminate flowers (a single whorl, cf. Garcinia, or two bimerous whorls?) and five in carpellate flowers (Forrest et al. 2005). De Wilde (2011 and references) noted the variation in vascular supply to members of the perianth, while the vascular bundle to the stamen may be branched (Gauthier 1963). The stigmas are described as being antisepalous (Davidson 1973); any style is at most short.

For floral development, see Charpentier et al. (1989), for ovules, see Boesewinkel and de Lange (1983), for placentation, de Wilde and Arends (1989 and references), for seed morphology, see de Lange and Bopuman (1999 and refernces), and for a good general account, see de Wilde (2011).

Phylogeny. For phylogenetic relationships within Begonia, see Plana et al. (2004), Forrest and Hollingsworth (2003) and Forrest et al. (2005). Thomas et al. (2011a), focusing on Asian Begonia, emphasized that several sections were para- or poplyphyletic, while Thomas et al. (2011b) showed that there had been several invasions of Malesia, including one now represented by memebsr of four sections.

Hillebrandia has a number of perhaps plesiomorphic features, and some of the features we think of as being characteristic of Begoniaceae as a whole (style position, fruit dehiscence) may in fact be apomorphies for Begonia alone.

Classification. For a sectional classification, etc., of Begonia, see Doorenbos et al. (1998), and for the species of the genus, see Smith et al. (1986), Golding and Wasshausen (2002) and Tebbitt (2005: more horticultural).

Previous Relationships. Like Cucurbitaceae, Begoniaceae have usually been associated with the other families that have parietal placentation, whether placed in Violales (Cronquist 1981) or in a group of small orders placed next to each other in Dilleniidae (Takhtajan 1997).

APODANTHACEAE Takhtajan   Back to Cucurbitales

Endophytic parasites; plant monoecious or dioecious; nodes?; stomata anomocytic; flowers fairly small; P 2 + 4 + 4 or 3 + 6 + 6, adaxial tuft of hairs on inner whorl; nectary +, at base of style/gynostemium; staminate flowers: gynostemium +; A synandrial, ca 15?, no vascular bundles evident, pollen sacs in 2 or 3 rings, extrorse, endothecium 0; pollen tricolpate, (apertures 0 - Berlinianche), psilate; pistillode +, vesicular hairs on margin (all over); carpellate flowers: staminodes 0; G [4 (5)], ± inferior, carpels opposite inner P, style short, very stout, hollow, stigma ± hemisperical; placentation parietal; ovules many/carpel, lacking vascular supply, tenuinucellate, micropyle bi/endostomal, or naked, outer integument 1 cells across, inner integument 1-2 cells across; antipodals persist?; fruit baccate; testa thin-walled, mucilaginous, exotegmen massively lignified; endosperm +, embryo undifferentiated; n = ± 12, 16, 30.

3/23(+): Pilostyles (20)[Photo - Flower]. New World from California and Florida southwards, Mediterranean and S. W. Asia, S. W. Australia and E. Africa (map: from George 1984; Novoa 2005; the Parasitic Plants Website 2004). Also Apodanthes, Berlinianche.

• All genera of Apodanthaceae occur as groups of rather small and sessile flowers growing along the stems of their hosts. The perianth of the flowers is free and more or less spreading, the stigma is conspicuous and mound-like in the middle, and the ovary is inferior.

Evolution. Recorded hosts include Fabaceae (perhaps most common), Salicaceae, Burseraceae, and Meliaceae.

Genes & Genomes. For the much increased rate of variation in synonymous substitution in some mitochondrial genes, see Mower et al. (2007 and references).

Chemistry, Morphology, etc. For the effect of Pilostyles on the wood structure of its host, Mimosa, see do Amaral and Ceccantini (2011).

There are cushions of hairs at the bases of the inner perianth parts (cf. Malvaceae?). Interpreting the meristicity of the flower is not easy, and the androecium in particular is difficult to understand (Blarer et al. 2004).

For information, see Harms (1935a: general), Kuijt (1969: general), Rutherford (1970: esp. anatomy, cytology), Takhtajan et al. (1985: pollen), Visser (1981), Blarer et al. (2004: floral morphology). For vesicular cells, see Blarer et al. (2002, 2004) and for a detailed study of Pilostyles ingae, see Endriss (1902). Also the Parasitic Plants website (Nickrent 1998 onwards) and also Heide-Jørgensen (2008) - both general - can be consulted with profit.

Previous Relationships. Apodanthaceae, like other holoparasitic angiosperms, were often previously included in Rafflesiaceae s.l., Rafflesiaceae s.s. here is included in Malpighiales. Recently, relationships with Malvales have been suggested, perhaps because some Malvaceae in particular lack normal anther thecal structure, the androecium may be fused, etc. (e.g. Blarer et al. 2004; Endress & Matthews 2006a; Schönenberger & von Balthazar 2006).

I am gratefull to S. Renner for comments.