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.

ASTERIDS - Sympetalae redux? - ASTERANAE Takhtajan   Back to Main Tree

Nicotinic acid metabolised to its arabinosides, caffeic acid derivatives +, (iridoids +); tension wood decidedly uncommon; C sympetalous, if evident only early in development, anthers dorsifixed?, (nectary gynoecial), ovules unitegmic, integument thick, endothelium +, nucellar epidermis does not persist, style +, long; endosperm cellular.

Evolution. The age of the stem group asterids may be ca 128 million years before present, mid Early Cretaceous, the Cornales and Ericales diverging soon afterwards, and the other asterid orders all diverging over 100 million years before present (K. Bremer et al. 2004); Wikström et al. (2003) suggest a crown group age of 117-107 million years before present, while Anderson et al. (2005: asterids other than Cornales and Ericales not sampled) suggest figures of ca 112 million years before present for the stem group, ca 109 million years before present for the crown group. Soltis et al. (2008: a variety of estimates) suggest an age of divergence of Cornales from the rest of 130-115(-86) million years ago. Magallón and Castillo (2009) offer estimates of ca 106.1 and 106.6 million years for relaxed and constrained penalized likelihood datings respectively for the divergence of Cornales from other asterids, the stem group of asterids was dated to 110.7 and 111.3 million years (relaxed and constrained again), while Moore et al. (2010: 95% highest posterior density) suggest much younger ages of (89-)84(-80) million years for diversification within this clade.

Iridoids, common in asterids, have been implicated in herbivore preferences, detering some and attracting others (e.g. see discussion under Plantaginaceae, Scrophulariaceae, etc.: Bowers 1980, 1988); iridoids have a bitter taste and are emetics for vertebrates, at least, and they are also sometimes sequestered by the insect eating the plant and used in its defence against predators (Nishida 2002 for a summary) - they may also be synthesized de novo by the insect (Burse et al. 2009 - Chrysomelina). Thus Uraniidae (moths) are found on Dipsacales, Lamiales, Gentianales - and also Daphniphyllaceae (an iridoid-containing member of Saxifragales - see Lees & Smith 1991), while larvae of Nymphalidae-Melitaeini butterflies are also almost restricted to asterids, although in this case they also quite common on Asteraceae and Acanthaceae, which, although asterids, lack iridoids; Melitaeini distinguish between plants with route I secoiridoids, which they eat, and route II decarboxylated iridoids (iridoid glycosides), which they rarely eat (Wahlberg 2001). Iridoids may also be involved in plant-plant relationships, the iridoid aglucone, formed by removing the sugar moiety, being toxic and perhaps accentuating the effect of parasite Orobanchaceae on their hosts (Rank et al. 2004), while iridoids from roots of Verbascum (Scrophulariaceae) may depress germination of competitors (Pardo et al. 2004).

Ericales and Cornales in particular show much variation in the degree of sympetaly, stamen number and development, adnation of stamens to corolla, and in ovule morphology and anatomy; some of this variation is like that found in rosids, Dilleniales, etc., and is unlike that in the asterids I + II. They may also have ellagic acid, which has a rather similar distribution (but cf. iridoids). For further discussion of this variation, see the asterids I + II.

Chemistry, Morphology, etc. Albach et al. (2001a) discussed iridoid distribution, etc., in the asterids, as do Soltis et al. (2005b). Mølgaard and Ravn (1988) and Rønsted et al. (2002) outlined the systematic utility of caffeic acid derivatives; chlorogenic acid, an ester of caffeic and quinic acid, is especially common in asterids, but also occurs elsewhere (see also also Lamiales and Boraginaceae in particular for other derivatives). Characteristic of the whole clade - although with numerous exceptions (derived), is the Baileyan wood anatomical syndrome of predominantly solitary vessels, scalariform perforation plates, mainly opposite vessel pitting, very long vessel elements and fibers at least 800 and 2190 µm long respectively, non-septate fibers with distinctly bordered pits, and diffuse to diffuse-in-aggregates and scanty paratracheal axial parenchyma (Lens et al. 2008).

Compound leaves are relatively uncommon in asterids, and when they occur the leaflets are often not articulated and/or distinct, however, elements of development are largely identical in very different-looking compound leaves (Bharathan et al. 2002; Blein et al. 2008). Taxa with stipules are also fairly uncommon, as are those with apetalous flowers and arillate seeds. Monosymmetry may have arisen some fifteen times in the asterids, with several reversals in Lamiales and Dipsacales (Jabbour et al. 2008: see also Donoghue et al. 1998; Ree & Donoghue 1999). Such flowers may have one, or rarely two, spurs (Jabbour et al. 2008). Lee et al. (2004) suggest that the CRABS CLAW gene is expressed in the rather different nectaries in the rosids (receptacular nectary) and asterids (gynoecial nectary) that they sampled; Bernadello (2007) surveys nectary variation in asterids.

K. Bremer et al. (2001) suggest some morphological synapomorphies for groupings. In general, where many characters are to be placed on the tree depends on resolution of relationships within Ericales and Cornales, and even then the pattern of gain-loss of some of these features is liable to be complex. Some characters common in asterids, including those of wood anatomy, probably have functional and logical linkages that also must be taken into account. Thus the presence of a tenuinucellate nucellus is linked with that of unitegmic ovules, the development of an endothelium (Kapil & Tiwari 1978), and a simple exotestal seed type (Netolitzky 1926); that of sympetalous monosymmetric flowers with epipetalous stamens, etc.

Sympetalae of older studies were defined largely by their sympetalous corolla, but some families here included in the asterids seem to be polypetalous. However, developmental studies like those of Erbar (1991) suggest that they have a ring primordium very early on (see, for example, Reidt & Leins 1994), i.e., they show early sympetaly (it is somewhat paradoxical that early corolla tube formation should often be linked with a corolla that is polypetalous at maturity!). The position of early initiation of the corolla tube on the tree is quite uncertain. Apiales, Asterales, and Dipsacales have many members with such initiation, as do both Oleaceae and Rubiaceae, "basal" or almost so in their orders in asterid I group, and so do some Cornales. Sampling leaves a great deal to be desired, but the condition of early initiation could conceivably be a synapomorphy for the asterids (see Erbar & Leins 1996b; Leins & Erbar 2003b for details). However, there may be an association of early corolla tube formation with an inferior ovary (Ronse Decraene and Smets 2000: families like Oleaceae with superior ovaries and apparently early corolla tube formation need more study from this point of view), in which case the character will need re-evaluation. Only in Ericales and other asterids does the mature flower have a corolla tube, hence the tentative assignment of posession of a corolla tube as an apomorphy for [Ericales + other asterids]; taxa with a pronounced corolla tube quite often have late corolla tube initiation, the petal primordia initially being free.

Note that a tube-forming hypanthium is rare in asterids, much less so in rosids. The integument, when single, is often dermal in origin, as is the inner integument of other angiosperms, while the outer integument is largely subdermal (but cf. monocots: see Bouman 1984; de Toni & Mariath 2009). Although this might suggest that the single integument of asterids corresponds to the inner integument of many bitegmic angiosperms, sampling need to be improved, and given that a number of Cornales and Ericales in particular have bitegmic ovules, the evolutionary story is unlikely to be simple.

Phylogeny. See the Dilleniales page for discussion on the relationships of the asterids; Caryophyllales and/or Santalales may be their sister group.

The monophyly of the asterids is well established (e.g. Olmstead et al. 1992, 1993, 2000; P. Soltis 1999); Albach et al. (1998) suggest the four main groupings recognised here. Relationships in phylogenies proposed by K. Bremer et al. (2001: analysis of 2 genes + morphology) and Albach et al. (2001b: analysis of four genes) are largely congruent. Differences are almost entirely in taxa not assigned to orders by A.P.G. (1998), although many of these may be assignable if the relationships suggested in the still provisional Bayesian analyses of Lundberg (2001b, d) hold. B. Bremer et al. (2002) provide a recent comprehensive phylogeny of the clade, although with minimal sampling within families, using three coding and three non-coding chloroplast markers. Both B. Bremer et al. (2003) and Olmstead (2000) suggest that there is strong support for Cornales being the sister to all other asterids; see also Albach et al. (2001) and Soltis et al. (2003). However, Hilu et al. (2003) reverse the positions of Cornales and Ericales, but the matK gene alone was sequenced; note that in their study, Caiophora (Loasaceae) appears in Asterales, far separate from the other members of the family in the analysis - perhaps a case of mistaken identity?

Previous Relationships. The distinction between the other angiosperms and the asterids partly corresponds to the distinction between the crassinucellate and tenuinucellate groups of Young and Watson (1970, based on phenetic analyses). There are also substantial differences, for example, Young and Watson included Apiaceae-Araliaceae in their crassinucellate group. Philipson (1974) further emphasized the distinction between the crassinucellate and tenuinucellate groups of Young and Watson, linking the two via Celastraceae, Grossulariaceae and Brexiaceae (here Celastrales, Saxifragales, and Crossosomatales, all rosids and not immediately related to asterids); Theales, Primulales and Ebenales together made up a separate lineage (here part of Ericales). Later Philipson (1977) resurrected van Tieghem's (1901) names Unitegminae and Bitegminae for these two groups; integument number and nucellus condition are correlated. General morphology indeed suggests such relationships, as Hufford (1992a) found even with phylogenetic analyses - Theaceae, Paracryphiaceae, Apiaceae and Araliaceae were members of Rosidae (but Pittosporaceae were sister to Polemoniaceae).

CORNALES Dumortier   Main Tree, Synapomorphies.

Iridoids diverse, ellagic acid +, flavones 0; vessel elements with scalariform perforation plates; nodes 3:3; inflorescence cymose; (flowers 4-merous), C valvate, apparently free, tube formation early; A basifixed, G inferior, crowned with disc-like nectary, 1-2 apical ovules/carpel, ventral carpellary bundles in the carpel wall [transseptal bundles, i.e. vascular bundles to ovules go over the top of the septum and then down; there are no bundles running up the central axis of the gynoecium]; fruit drupaceous, with apical germination valve(s) in the stone, K persistent. - 6 families, 51 genera, 590 species.

Evolution. Fruits of this small clade are recognizable by their distinctive anatomy and are well represented in the fossil record (Manchester et al. 2007) and are datable to the Cretaceous-Maastrichtian, ca 70 million years before present (Nyssa) and Coniacian, ca 87 million years before present (Takahashi et al. 2002: Hironoia). Anderson et al. (2005) suggest figures of ca 109 million years before present for the stem group, 101-97 million years before present for the crown group; Janssens et al. (2009) date stem group Cornales to ca 128 million years ago and the crown group to 104±13.1 million years; and in Bremer et al. (2004) crown group divergence is estimated to have begun some 112 million years ago. Magallón and Castillo (2009) offer estimates of ca 106.1 and 106.6 million years for relaxed and constrained penalized likelihood datings respectively for the divergence of stem Cornales from other asterids, crown Cornales being dated to 101.4 and 101.7 million years (relaxed and constrained again).

Chemistry, Morphology, etc. The strands of apotracheal parenchyma are relatively long (at least 9 cells long) in Cornaceae s.l. (inc. Curtisiaceae) when compared with some of their putative relatives (Noshiro & Baas 1998). Spirally-thickened vessels holding the two halves of transversely-torn leaves together are quite common... Teeth of Nyssaceae and Hydrangeaceae have a clear apex with a foramen, higher order laterals are involved (Hickey & Wolfe 1975). The petals may be free, but corolla tube formation, when known, is early (e.g. Reidt & Leins 1994).

For more details, see Faure (1924), Ferguson (1977: pollen), Sato (1976: embryology), Grayer et al. (1999: saponins).

Phylogeny. Molecular studies (e.g. Xiang et al. 1993) suggest a break-up of the old, broadly circumscribed Cornaceae; the core is here. Relationships between genera in this core are unclear, but at least some aggregation of the families they represent is in order (e.g. Albach et al. 2001b; Xiang et al. 2002), indeed, relationships in Cornales as a whole are unclear. Although Cornus is sister to Mastixiaceae in some morphological trees (Murrell 1993), it is not nearly so close in rbcL trees (Xiang et al. 1993, 1997). For the relationships of Grubbiaceae and Hydrostachyaceae (placement of the latter is especially difficult, see below), see especially Hempel et al. (1995), Xiang (1999), Soltis et al. (1997, 2000, 2007a), Savolainen et al. (2000b), Fan and Xiang (2003) and Xiang et al. (2002); the tree here is based largely on the last two papers. There is support for a sister group relationship between Grubbiaceae and Curtisiaceae (e.g. Fan & Xiang 2001).

Indeed, the position of Hydrostachyaceae has long presented problems. The embryology of the family shows certain similarities with that of Crassulaceae, but neither relationships there or with Podostemaceae (see below) can be close. Members of sympetalous groups, especially Lamiales, show similarities to Hydrostachyaceae in ovary structure (apical septae) and ovule and endosperm development, but although the coenocytic micropylar haustorium is well developed, the chalazal endosperm cell, which remains undivided, is barely haustorial and the carpels are tranverse, rather than median as in most Lamiales (Jäger-Zürn 1965; see also Rauh & Jäger-Zürn 1966, 1967 [strongly supporting a relationship with Lamiales]; Leins & Erbar 1988, 1990). However, in some Orobanchaceae (e.g.) the chalazal haustorium is also very poorly developed (Tiagi 1963), as it is in Lamiales basal to Calceolariaceae. A position within Hydrangeaceae has also seemed to be quite likely (Xiang 1999; see also Hempel et al. 1995; Olmstead et al. 2000; Albach et al. 2001; Fan & Xiang 2001; Xiang et al. 2002), but note the very long branch; what about the mitochondrial coxII.i3 intron (Joly et al. 2001)? As Albach et al. (2001) note, few morphological characters support this position, but one could argue that this is perhaps to be expected of a highly-derived aquatic... However, Burleigh et al. (2009) recently found that in a five gene analysis there was strong support (97% ML bootstrap) for a position of Hydrostachys within Lamiales, largely because of the matK sequence added. Where in the Lamiales it might be placed was unclear, although it would be in a clade that excluded Oleaceae, at least. More comprehensive analayses are needed; Calceolaria and other clades below it in Lamiales other than Oleaceae were not sampled by Burleigh et al. (2009). All in all, however, it seems to me to be likely that Lamiales are where Hydrostachyaceae will end up; so in addition to being included here, Hydrostachyaceae are also provisionally placed immediately before Calceolariaceae in Lamiales. Note that the inclusion of Hydrostachys in analyses of Cornales considerably affects topologies and support values throughout the tree of the order.

Previous Relationships. Takhtajan (1997) included Hydrangeales in Cornidae-Cornanae, but Loasales-Loasanae were part of his Lamiidae. 11/15 of the genera of Cornaceae s.l. have been placed in monotypic families, or the family has been circumscribed very broadly, as by Mabberley (1997). Previous inhabitants of the old Cornaceae may be found here in Garryaceae (Garryales), Curtisiaceae (Solanales), Argophyllaceae (Asterales) and Griseliniaceae (Apiales).

This is the asterid IV group of some early phylogenetic studies.

Includes Cornaceae, Curtisiaceae, Grubbiaceae, Hydrangeaceae, Hydrostachyaceae, Loasaceae.

Synonymy: Alangiales de Candolle, Grubbiales Doweld, Hortensiales Grisebach, Hydrangeales Nakai, Hydrostachyales Reveal, Loasales Bessey, Nyssales Jussieu, Philadelphales Link - Cornanae Reveal, Loasanae Reveal - Cornidae Reveal

CORNACEAE Dumortier, nom. cons.   Back to Cornales

(Plants Al accumulators); route I secoiridoids, triterpenoid saponins +, tanniniferous; (mucilage +); (laticifers +), hairs T-shaped, unicellular; flowers small, K notably small, pollen with complex endaperture [a pore joining two lateral thinnings parallel to the colpus], (archesporium multicellular), style short; (endosperm also nuclear).

7: 107. Predominantly north temperate, rather scattered elsewhere. Two groups below.

1. Cornoideae Endlicher

Trees and shrubs (stoloniferous subshrubs); flavonols, also route II decarboxylated iridoids, isoquinoline alkaloids +; (latex +); (vessel elements with simple perforation plates); sclereids +; petiole bundle arcuate, or D-shaped or annular (with medullary bundle); branching various; hairs (stellate), walls often with crystals; leaves opposite, bases joined by a line, spiral or two-ranked, conduplicate(-flat) or curved (both -plicate) or involute, margins entire (lobed), 2ndary veins pinnate or subpalmate; flowers 4(-10)-merous, K connate or not, stamens = and opposite sepals (-4x, anthers long - Alangium), pollen with H-shaped endapertures, often starchy; G [1-4], (1 loculus), ovules apotropous, (tenuinucellate - red-fruited Cornus), (style long, with long arms), stigma truncate to capitate, dry; drupe 1-2-seeded, walls made up of sclereidal cells; testa of elongated cells, much compressed, (ca 6 cells thick, vascularized - Alangium); endosperm hemicellulosic, embryo green; n = 8-11.

2[list]/85: Cornus (65 spp). Scattered, not S. South America (map: see van Steenis & van Balgooy 1966; Aubréville 1974; George 1984; Meusel et al. 1978; Hultén & Fries 1986; Xiang & Thomas 2008). [Photos - Habit, [Photo: Cornus Inflorescence, Flower, Fruit.]

Synonymy: Alangiaceae Candolle, nom. cons.

2. Nyssoideae Arnott

Trees and shrubs; route I secoiridoids +; (resin +); petiole bundles arcuate or with adaxial plate; leaves spiral (opposite), conduplicate [Nyssa], margins serrate or entire; plants andromonoecious, dioecious, etc., or flowers perfect; inflorescences various (capitate); flowers 4-5-merous; C ± imbricate, or valvate and inflexed at apex, or P 0; A 4-26 [often diplostemonous], G [5-10], (loculus and ovule 1), (styles +); endocarp fibrous; fruits 1-5-seeded, stone walls of fibrous cells [?an apomorphy]; (seed U-shaped - Mastixia), testa multiplicative, exotesta lignified; embryo long or short; n = 11, 13 [both Nyssa], 21, 22.

5/22. Mainly East Asia, also Indo-Malesia and E. North America (map: see van Steenis & van Balgooy 1966; Matthew 1977). [Photo - Nyssa Flower, Fruit © H. Wilson.]

Synonymy: Davidiaceae H.-L. Li, Mastixiaceae Calestani, Nyssaceae Dumortier, nom. cons.

• Many Cornoideae are recognisable vegetatively by their T-shaped hairs and leaves often with entire margins and actinodromous venation, the lateral veins arching and proceeding towards the apex. The flowers are often small and aggregated into heads or other compact inflorescences; the flowers of one inflorescence open more or less together. The flowers have a small calyx, apparently free and valvate petals, an inferior ovary crowned by a disc and with 1-2 locules/carpel, and a 1-seeded drupaceous fruit.

Evolution. Xiang and Thomas (2008) suggested that stem-group Cornus was Late Cretaceous in age, ca 80 million years old, substantial diversification having occured by ca 66 million years ago. For the early Tertiary fossil history of what are now East Asian endemic genera of Cornaceae, see Manchester et al. (2009) and references. Fossils of fruits of Nyssoideae are widespread in the northern hemisphere in the early Tertiary, some being 3- or 4-carpellate (Eyde 1997, for details), Mastixia being especially abundant in Europe 65-70 million years before present.

For the indole alkaloid camptothecin, see Lorence and Nessler (2004). The enzyme that camptothecin targets is in the plant, but the plant is probably protected by changes in its amino acid sequence, one of which (serine in position 722) is the same as is found in camptothecin-resistant tumours in humans (Sirikantaramas et al. 2009)! In addition, note the sequestration of camptothecin in glandular hairs although not in the laticifers (Hagel et al. 2008).

Chemistry, Morphology, etc. Within Cornoideae, blue-fruited dogwoods have lost iridoids (Xiang et al. 1997). In nodes of Alangium the central vascular trace may immediately divide into three (nodes 3:5). Mabberley (1997) describes Alangiaceae as having spiral leaves; they are often two-ranked. The floral organ diversity genes B and C are expressed in the large, white inflorescence bracts of Cornus (Maturen et al. 2005; see also Zhang et al. 2008 - extensive PI-like gene duplication). In Alangium there is a very little vascular tissue in the center of the ovary, while there is considerable variation in embryo sac development in Cornus in particular (Johri et al. 1992 for references).

For further information, see Adams (1949: anatomy) and Eyde (e.g. 1968, 1988: flower and fruit in particular), Neubauer (1978: petiolar anatomy), Jensen et al. (1975a: iridoids), and Kubitzki (2004b: general).

In Nyssoideae, Davidia has flowers in capitula subtended by 2 large white bracts; it lacks a perianth and may have bitegmic ovules. Diplopanax has recently been placed in Mastixiaceae s. str. (Eyde & Quiyun 1990; cf. Xiang et al. 1997). It has five lobes on the disc opposite the corolla and a single-seeded fruit the embryo of which is C-shaped in transverse section (Ying et al. 1993), and it also contains petroselenic acid (Zhu et al. 1998). Although they did not find it in Fatsia (Araliaceae) or Aucuba (Garryaceae), petroselenic acid is found in a number of Apiales, and relationships between Cornaceae and some Apiales have been suggested in the past...

For cytology, see He et al. (2004), for general information, see Kubitzki (2004b: as Cornaceae), for Mastixia, see Matthew (1976); embryological details are unknown for it and Diplopanax.

Phylogeny. For a careful study of reconstructing ancestral areas and characters in Cornus, see Xiang and Thomas (2008); the results depended much on the methods used, etc. For relationships within Cornus, cf. Murrell (1993) and Xiang et al. (1993, 2006), and for relationships within Alangium, see Feng et al. (2009).

Hydrangeaceae + Loasaceae: similar route I secoiridoids and route II decarboxylated iridoids [C-8 iridoid glucosides +; C-9 iridoids, keeping the C-11, e.g. deutzioside, +], flavonols +, ellagic acid 0; cork cambium deep-seated; hairs tuberculate, walls calcified, with basal cell pedestals; leaves opposite, with glandular teeth (lobed); A (initiated as antesepalous triplets), (2x C-)many, placentation parietal, gynoecium with axial/central vascular bundles, many tenuinucellate ovules/carpel, stigma dry; fruit septicidal, (persistent placental strands +); exotestal cells variously elongated, inner walls thickened; micropylar and chalazal endosperm haustoria +; mitochondrial coxII.i3 intron 0.

Chemistry, Morphology, etc. For the distinctive iridoids of this family pair, see Frederiksen et al. (1999). The androecium of both families is very variable in development (Hufford 1990, 1998). It is possible that diplostemony is plesiomorphic, with polystemony derived. The antisepalous androecial triplets sometimes found here are also found in Rosaceae and Zygophyllaceae (Hufford 2001b, see also Ronse Decraene & Smets 1996a). The embryology of the group is poorly studied.

HYDRANGEACEAE Dumortier, nom. cons.   Back to Cornales

Shrubs, vines, or herbs; (plants Al accumulators); kaempferol, flavonols +, tanniniferous; (hairs stellate or branched); cork inner cortical or outer pericyclic; (vessel elements with simple perforation plates); true tracheids +; (stomata paracytic); leaves conduplicate or supervolute, bases joined by lines across the stem, (2ndary veins palmate); inflorescence cymose; flowers 4-5(-10)-merous, anthers with basal pit, nectary vascularised, G [(2-)3-5(-12)] to inferior, ovary ribbed, arrangement variable, placentation intrusive parietal, ovules apotropous, integument 5-7(-10 - Hydrangea) cells across, style or styles +, stigma linear to capitate; seed winged or not; endosperm moderate; n = 13-18.

17[list]/190 - divided into two subfamilies, and one subfamily into two tribes. Warm temperate, some species in tropics. [Photo - Flower] [Photos - Collection]

1. Jamesioideae Hufford

Nothing obvious! (Myricetin +); leaf buttresses prominent after leaves fall; K valvate, C free; A 10; G (3-)4-5, style branches separate or almost so; endosperm nuclear [Fendlera]; n = 16.

2 (Jamesia, Fendlera)/ca 5. W. North America (map: from Holmgren & Holmgren 1989).

2. Hydrangeoideae Burnett

Nodes also 5:5, 7(+):7(+); petiole bundles (arcuate [+ inverted bundles]) annular, often with medullary bundles; raphide sacs + (0); stomata variable; (hairs stellate); (filaments winged), cytokinesis simultaneous [Platycrater], pollen grains 2-celled, integument 3-5 cells across.

15/185. Warm temperate, esp. South East Asia and North America, S. to Chile and Malesia (map: from Hu 1955; Zaikonnikova 1966; McClintock 1957; van Balgooy 1984; Mai 1985; Hong 1993).

2a. Philadelpheae

(K connate), C imbricate; A initiation as five common primordia, embryo sac ± protruding from the nucellus, style single.

6/130: Philadelphus (65), Deutzia (60). Warm temperate, esp. South East Asia to the Philippines, SW North America, also Central America, Philadelphus coronarius in Europe.

Synonymy: Philadelphaceae Martynov

2b. Hydrangeeae

Myricetin + [Decumaria]; raphides +; inflorescence often with conspicuous marginal flowers; C valvate, styles separate; fruits loculicidal, (baccate).

9/65. Warm N. temperate, S. to Chile and Malesia.

• Hydrangeaceae are usually rather robust herbs or shrubs with opposite leaves that are joined by a line across the stem. The flowers have free petals that are valvate in bud and at least twice as many stamens as petals; the ovary is more or less inferior.

Chemistry, Morphology, etc. Species of Deutzia have stamens in a single whorl with strongly flattened filaments that may form a tube around the ovary and style. Philadelphus shows centrifugal androecial development. In Philadelphus, Dichroa and Deutzia the four carpels alternate with the sepals, or there are three carpels with the odd member adaxial; in Hydrangea the odd carpel is abaxial, while in Broussaisia the five carpels are opposite the sepals. In a number of taxa the embryo sac more or less protrudes into the micropyle or beyond (Maheshwari 1950; Hufford 2004). The presence of chalazal haustoria needs confirmation.For floral morphology of Hydrangeae, see Hufford (2001). The endocarp of Hydrangea consists of large cells with digitate-interlocking anticlinal walls, as in Curtisia, but not in Cornus (Manchester et al. 2007). The base of the endosperm is lignified; Fendlera has nuclear endosperm (Johri et al. 1992).

For variation in the position of the carpels when the gynoecium is bicarpellate, see Eichler (1878; also Eckert 1966), for vegetative anatomy, Watari (1939), Styer and Stern (1979 and references) and Gornall et al. (1998), for floral anatomy; see Bensel and Palser (1975c), for flavonoids, see Bohm et al. (1985), for seeds, Hufford (1995, 1997) and Nemirovich-Danchenko and Lobova (1998), for iridoids, Frederiksen et al. (1999), for androecial development, Gelius (1967) and Hufford (1998, 2001a), for some embryology, Ao (2008), and for general information, see Hufford (2004).

Phylogeny. For relationships within the family, see Hufford (1997b), Hufford et al. (2001: support for its monophyly is not overwhelming), and Soltis et al. (1995a).

Classification. For a classification of Hydrangeaceae, see Hufford et al. (2001).

Synonymy: Hortensiaceae Berchtold & J. S. Presl, Kirengeshomaceae Nakai

LOASACEAE Jussieu, nom. cons.   Back to Cornales

Often coarse herbs (shrubs); myricetin, tannins 0; cork inside pericycle; vessel elements with simple perforation plates; petiole bundles arcuate or annular, with wing bundles; trichomes glochidiate (stinging), often silicified; leaves (spiral; compound; margins lobed), 2ndary veins pinnate-palmate; flowers (4-)5(-7)-merous; K connate, C with three traces, ?valvate, C-A synorganisation, C-A plate formed, filaments terete, tectum striate; G [5], (± superior), opposite sepals, ovules epitropous, integument very thick, style hollow, lobed, stigma narrow or clavate; (fruit spirally twisted); (testa with hypodermal layer thickened); endosperm copious to none.

14[list]/265 - five clades below. Mostly American, but also Africa and the Marquesas Islands (map: from Heywood 1978).

1. Eucnide

(C connate), A (adnate to C), centripetal, connate basally; fruit a septicidal capsule; n = (?19-)21.

1/15. S.W. North America. [Photo - Eucnide Flower © J. Reveal]

2. Schismocarpus

A 10, filaments shorter than the anthers; G opposite petals, stigma capitate.

1/1: Schismocarpus pachypus. Mexico.

Loasoideae [Mentzelioideae + Gronovioideae]: G [3-5], when [3], odd member adaxial.

3. Loasoideae Gilg

K and C shed separately, petals cymbiform, clawed; A centripetal and centrifugal, stamens in 5 groups opposite petals, pollen ?not striate, antesepalous staminodes + [outer whorl connate, as scales, inner whorl separate, more elaborated]; n = 6.

Nasa (105). America, but also Africa (Kissenia) and the Marquesas Islands (Plakothira). [Photo - Flower, Flower, Flower, Fruit, Flower.]

Mentzelioideae + Gronovioideae: loss of C-A synorganisation.

4. Mentzelioideae Gilg

K and C quincuncial, shed as a unit, A centripetal, connate basally, (forked staminodes +); n = 7.

Mentzelia (60). [Photo - Flower © S. Wolf.]. America.

5. Gronovoioideae M. Roemer

(Hypanthium +), C valvate, petals with a single vascular trace; A 5, opposite sepals (2, three staminodes), anthers bifacial; G 3, with 1 apical crassinucellate [Petalonyx, Gronovia] ovule, funicular obturator +; fruit a cypsela; testa none; endosperm haustoria 0.

[Photo - Gronovia Flower.] America.

Synonymy: Cevalliaceae Grisebach, Gronoviaceae Endlicher

• Loasaceae are usually herbs that have barbed or glochidiate silicified and sometimes stinging hairs; at least the first pair of leaves is opposite. The flowers, which usually have separate, spreading petals, numerous, radiating stamens with long filaments, and an inferior ovary, are distinctive.

Evolution. Schenk and Hufford (2008) suggest dates for some of the splits of the main clades in the family.

There are some remarkable floral morphologies in the family. Weigend and Gottschling (2006) discuss pollination in Nasa; there are revolver flowers in the genus (see also Ackermann & Weigend 2006).

Chemistry, Morphology, etc. There is variation in the composition of fatty acids in the seeds, but its systematic significance is unclear. Of the taxa studied by Weigend et al. (2004b), Nasa (Loasoideae) was most distinct; it is well embedded in the family. In some species of Petalonyx (Gronovoideae) there is postgenital fusion of the corolla, this forces the stamens outside the corolla. For the complexities of androecial initiation, see Hufford (1990); antepetalous stamens arise from the flanks of primordia of antisepalous stamens. Hufford (2003) described staminode evolution in detail. The stigma is at least sometimes very long (Loasa triphylla - see Hanf 1935).

Additional information is taken from Vijayaraghavan and Kaur (1967), Thompson and Ernst (1967), Brown and Kaul (1981: floral morphology), Weigend (1996), Hufford (1988, 1989, 1990) and especially Moody and Hufford (2000) and Weigend (2004: general).

Phylogeny. Strongly supported relationships suggested by Moody and Hufford (2000), Moody et al. (2001), Hufford et al. (2003) and Hufford (2003) are Eucnide [Schismocarpus [Loasoideae [Mentzelioideae + Gronovioideae]]]. Within Loasoideae, the clade [Plakothira + Klaprothia + Kissenia] may be sister to the rest, but that relationship has little support (Hufford et al. 2005), or the clade may be part of a major polychotomy (see also Weigend et al. 2004a). The distribution of [Plakothira + Klaprothia + Kissenia] is remarkable - the Marquesas, South America, Africa...

HYDROSTACHYACEAE Engler, nom. cons.   Back to Cornales

Annual to perennial submerged rosette herbs; primary root 0, adventitous roots +; kaempferol +, iridoids 0; vessels present, ?type; nodes ?; stomata 0; leaves deeply and complexly divided, surface with small enations, stipule single, intrapetiolar (two, lateral); inflorescence spicate, plants di(mon)oecious; P 0, nectary 0; staminate flowers: A 2, extrose, monothecal, pollen in tetrads, inaperturate; carpellate flowers: G [2], transverse, placentation parietal, many tenuinucellate ovules/carpel, styles separate, filiform, impressed in the apex of the ovary; fruit a septicidal capsule; seeds minute, exotestal, outer cell walls much thickened, mucilaginous; endosperm scanty or 0, micropylar haustorium +; n = 10-12.

1[list]/20. C. and S. Africa, Madagascar (map: from Rauh & Jäger-Zürn 1966b).

• Hydrostachyaceae are submerged, rosette-forming, aquatic herbs with complex leaves bearing enations, stipules that are usually intrapetiolar, dense, spicate inflorescence, and rather small flowers with extrorse anthers and free styles.

Chemistry, Morphology, etc. The caffeoyl ester chlorogenic acid is found here and in the Loasaceae-Hydrangeaceae clade (Rønsted et al. 2002). Another interpretation of the androecium is that is consists of one tetrasporangiate stamen. Vessels are reported (Jäger-Zürn 1998), but are not described. The integument is about 5 cells thick; there seems not to be an endothelium. The styles are more or less impressed into the apex of the ovary, a feature that Leins and Erbar (1988) noted was common in Lamiales, although I do not know the general distribution of this feature.

For floral development, see Leins and Erbar (1988), for general information, see Erbar and Leins (2004a), and for chemistry, Rønsted et al. (2002).

Previous Relationships. Hydrostachyaceae have variously been suggested as being sister to Decumaria (Hydrangeaceae) (Albach et al. 2001), or close to Crassulaceae (Saxifragales), or - perhaps - close to Podostemaceae (near Clusiaceae, in Malpighiales). However, Takhtajan (1997) included Hydrostachyales in his Lamiidae, and Cronquist (1981) also put it in that general area.

Grubbiaceae + Curtisiaceae: leaves opposite, bases connected by a ridge; flowers small, one tenuinucellate ovule/carpel, style short, lobed; walls of stone made up of sclereids; endosperm copious.

Evolution. The age of this clade is ca 90 million years, suggesting that it is very much a relict in the Cape flora (Warren & Hawkins 2006), although fossil fruits of Curtisia have recently been identified in the Eocene of southern England (Manchester et al. 2007).

Chemistry, Morphology, etc. For characters holding these two families together, see in part Xiang et al. (2002).

Classification. Xiang et al. (2002) suggested that Grubbiaceae and Curtisiaceae might be combined, but they are kept separate here because they are rather different in appearance (see also A.P.G. III 2009).

GRUBBIACEAE Endlicher, nom. cons.   Back to Cornales

Evergreen ericoid shrubs; iridoids 0?; hairs unicellular; cuticle waxes as long narrow platelets; leaf margins revolute; inflorescences axillary, capitate or cone-like; flowers also 6-merous, K valvate, C 0; A 8, 12, anthers inverted, bisporangiate/monothecal; G [2], transverse, disc hairy, placentation axile at base, becoming free-central, ovule epitropous, integument "thick"; fruit a syncarp, seed [per fruit proper] 1, coat thin; endosperm ?type, micropylar and chalazal endosperm haustoria +, embryo "large"; n = ?

1[list]/3. Cape Province, South Africa (map: from Vester 1940).

• Grubbiaceae are ericoid shrubs with capitate or cone-like inflorescences lacking radiating inflorescence bracts; its fruits are closely aggregated (cf. Bruniaceae).

Chemistry, Morphology, etc. The family is poorly known. The inversion of the anther is very comprehensive in Grubbiaceae, and for some (e.g. Fagerlind 1947b) this has suggested relationships with Ericaceae. Carlquist (1978a) found Grubbiaceae to be anatomically identical to Bruniaceae (Bruniales, in the asterid II clade), cf. also Geissolomataceae (Crossosomatales - rosids).

Some information is taken from Schnizlein (1843-1870: fam. 18 - carpel orientation), Fagerlind (1948b: embryology), Dahlgren and van Wyk (1988: general) and Kubitzki (2004b: general).

Synonymy: Ophiraceae Arnott

CURTISIACEAE Takhtajan   Back to Cornales

Evergreen trees; route I secoiridoids +, ?ellagic acid; ?cork; ?nodes; petiole bundle annular, with medullary strands; leaves ± flat, margins serrate; inflorescence terminal; K small, open, stamens = and opposite sepals, pollen with H-shaped endapertures; G [2-4], ovules epitropous, gynoecium with axial/central vascular bundles; fruit 4-seeded; endotesta tanniniferous, rest ± collapsed; ?endosperm haustoria, embryo minute; n = 13.

1/1: Curtisia dentata. Southern Africa (map: from Palgrave 2002; Yembaturova et al. 2009; fossil [blue] from Manchester et al. 2007a]). [Photo - Fruit]

Evolution. Manchester et al. (2007a) recognised the distinctive fruits of Curtisia from the Eocene of southern England; the fossils were originally described under Epacridaceae (= Ericaceae - Styphelioideae).

Chemistry, Morphology, etc. Takhtajan (1997) described the hairs of the branchlets, petioles and inflorescences of Curtisia as being stellate; they are simple and curled. The "plications" (Cullen 1978) in the young leaves are in fact only prominent veins. Curtisia is embryologically unknown, but it lacks transseptal bundles, having the "normal" central bundles.

For general information, see Kubitzki (2004b).

Classification. See Yembaturova et al. (2009).