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.

[AUSTROBAILEYALES [[CHLORANTHALES + MAGNOLIIDS] [MONOCOTS [CERATOPHYLLALES + EUDICOTS]]]]: ethereal oils in spherical idioblasts [lamina and P ± pellucid-punctate]; tension wood 0; tectum reticulate-perforate [here?], nucellar cap + [character lost where in eudicots?]; 12BP [4 amino acids] deletion in P1 gene.

[[CHLORANTHALES + MAGNOLIIDS] [MONOCOTS [CERATOPHYLLALES + EUDICOTS]]] / MESANGIOSPERMAE: benzylisoquinoline alkaloids +; outer epidermal walls of root elongation zone with cellulose fibrils oriented transverse to root axis; P more or less whorled, 3-merous [possible position]; carpels plicate; embryo sac bipolar, 8 nucleate, antipodal cells persisting; endosperm triploid; ?germination.

I largely follow Ronse De Craene et al. (2003) on the insertion of floral organs. To add where?: A whorled, carpel fusion by congenital occlusion.

Evolution. Divergence & Distribution. The divergence of magnoliids, i.e. the age of this branching point, is estimated at (190-)167(-47) million years (95% HPD, Smith et al. 2010, Table S3), while a fossil-based estimate is ca 100 million years (Crepet et al. 2004).

Phylogeny. Relationships between the lineages immediately above the basal pectinations in the main tree, the ANITA grade - Amborellales, Nymphaeales and Austrobaileyales here - were for some time unclear, although recent work is suggesting resolutions. The positions of Ceratophyllales and Chloranthaceae have been particularly labile, the former having been placed e.g. as sister to the eudicot lineage, or sister to all flowering plants. Indeed, Graham et al. (2005) found that the inclusion of Ceratophyllum and Chloranthaceae in analyses could destabilise relationships among other early branching clades, e.g. the position of the monocots became labile. There has been weak support for an (eu)magnoliid clade being sister to the eudicots (e.g. D. Soltis et al. 2000; P. Soltis et al. 2000; Wu et al. 2007; Qiu & Estabrook 2008 [compatability analysis]), in the former Chloranthales, monocots, and Piperales were successively sister to the remaining taxa, Magnoliales, Canellales and Laurales also belonging, but these relationships also had little support. Hilu et al. (2003: a matK analysis alone), suggested that Ceratophyllaceae were sister to eudicots (see also D. Soltis et al. 2000; Borsch et al. 2005: 62% jacknife, 96% posterior probabilities, the study focussed on three rapidly evolving genes; Müller et al. 2006: 96% posterior probability; some analyses in Saarela et al. 2007; Qiu & Estabrook 2008). However, Zanis et al. (2002) analysed 11 genes from all genomes and found some support for the magnoliids (strong support for this clade) as being sister to eudicots, with Chloranthaceae, and [monocots + Ceratophyllaceae] occurring as successively more basal branches; below these were the ANITA grade (see also Graham & Olmstead 2000; Borsch et al. 2003). Support for some of these nodes depended on the method of analysis (see also Borsch et al. 2000; Graham & Olmstead 2000; Hilu et al. 2001; Whitlock et al. 2001; Zanis et al. 2002) or the particular gene studied (see Duvall & Bricker 2002 - nuclear 18s; Hilu et al. 2003 - matK; Duvall et al. 2006 - nuclear 18S in particular causes problems). Davis et al. (2004) found the magnoliids to be sister to monocots, with Chloranthaceae and Ceratophyllaceae successively sister to a clade including the few eudicots in the analysis, but support for none of these positions was exactly overwhelming (the best supported topology in Duvall et al. 2006 is somewhat similar). There was fairly good support for a grouping [monocots + Ceratophyllaceae] in a compartmentalised 6-gene analysis (Zanis et al. 2003; see also Qiu et al. 1999; some analyses in Saarela et al. 2007). Müller et al. (2006) found very poorly supported relationships between Chloranthaceae and monocots, which together linked with the magnoliids. In work by Whitlock et al. (2002) largely similar groupings were recognised, but support was only moderate; the exact position of Chloranthaceae remained unclear. Some analyses suggested the possibility of a sister taxon relationship between Chloranthaceae and eudicots (Borsch et al. 2003), while Hilu et al. (2003) in a matK analysis, suggested that they were sister to monocots. Although the support was weak in parsimony analyses, there were 100% posterior probabilities in Bayesian analyses, furthermore, in the latter case only, [Chloranthacaeae + monocots] were sister to magnoliids, although the probabilities there were low. Qiu et al. (2005; see also Löhne & Borsch 2005) found initial rather strong bootstrap support for an association between monocots and Ceratophyllaceae in a 9-gene analysis being vitiated by the failure to obtain much support in any of the subanalyses and by details of the topology obtained in the 9-gene analysis itself (e.g. Acorus sister to the Alismatales that were included) that are rather improbable. Other studies with more restricted sampling also show no clear pattern (e.g. Jansen et al. 2006a). And on it goes! - relationships are unclear in Qiu et al. (2006b). Doyle and Endress (2000) presented combined morphological and molecular data that suggested Piperales might be sister to monocots, Parkinson et al. (1999) molecular evidence that they were sister to eudicots, but in both cases the support was weak, and neither position is likely. Interestingly, Piper has the "monocot" pattern of PHY genes (A, B, C genes only), while Ceratophyllum has PHYE (Mathews et al. 1995). A polytomy involving all these taxa is evident in P. Soltis et al. (1999: support low), while Soltis et al. (2005b) very reasonably summarize their discussion on relationhips in this area by showing a pentatomy.

Indeed, morphological analyses have quite often suggested that more or less herbaceous clades like Piperales, Chloranthaceae, and Nymphaeales might be related to monocots, the pal(a)eoherb hypothesis (see e.g. Donoghue & Doyle 1989; Taylor & Hickey 1992; Endress 2000). Although quite popular in the latter years of the last century, its support comes from homoplasies associated with the adoption of the herbaceous habit. Indeed, Piperales are now firmly embedded in the magnoliid clade, Nymphaeales are very near the base of the whole angiosperm clade, etc.

It has been suggested that the discovery of structural changes in the genome might clarify relationships in this area (Löhne & Borsch 2005; Qiu et al. 2005). Indeed, some resolution may be on the way. Graham et al. (2005) found a rather weakly supported (73% bootstrap) [monocot + eudicot] grouping, but this was weaker when Chloranthaceae and Ceratophyllaceae were included. However, evolution of some floral developmental genes, e.g. in the C and D lineages, are also consistent with such a relationship (Kramer et al. 2004). The positional relationships between members of the androecium and the perianth, the stamens being individually opposite perianth members, and perhaps some other characters (trimery of some floral whorls is fairly widespread within Ranunculales) are also consistent with such relationships. (Note that the same perianth member/stamen relationship may be found in some magnoliids, e.g. Lauraceae.) There may also be phylogenetically interesting variation in how calcium oxalate is synthesized, although sampling is very poor; neither members of the ANITA grade nor magnoliids have been examined. Similarly, vacuolar crystal formation associated with membranes and paracrystalline bodies with widely spaced subunits are found in eudicots (crystals seem also to be formed in other ways here), while in monocots there are no membrane complexes and the paracrystalline bodies have closely spaced subunits (Horner & Wagner 1995; Evert 2006). Some patterns of RNA editing may be phylogenetically informative at this level (Logacheva et al. 2008).

Recently, Jansen et al. (2006b: 37 whole chloroplast genomes, see also Zhengqiu et al. 2006), Cai et al. (2006: 35 whole chloroplast genomes, neither Chloranthales, Ceratophyllales, nor any member of Austrobaileyales included), and Duvall et al. (2006: four genes, three compartments, nuclear PHYC gene, 18S being excluded - overall, a relationships between magnoliids and monocots was preferred), Mathews (2006a: three PHY genes, 105 taxa), and Hansen et al. (2007: 61 protein-coding chloroplast genes) have also found support for this [monocot + eudicot] grouping. Jansen et al. (2006b) and Hansen et al. (2007) found support for a sister-group relationship between Chloranthales and the magnoliids, although the analyses in Mathews (2006a) preferred an unresolved position for Chloranthales and Ceratophyllales above the Austrobaileyales, while Duvall et al. (2006) and Qiu et al. (2010: chloroplast data, support weak) found a relationship between Chloranthales and Ceratophyllales. Soltis et al. (2007a) found weak support for [Ceratophyllum + eudicots]; relationships of Chloranthaceae were unclear and there was very weak support for monocots as sister to [magnoliids, Chloranthaceae, eudicots, etc.]. When sequences of complete chloroplast genomes are analysed, an association between monocots and eudicots is more strongly suggested. Thus Jansen et al. (2007) found strong support for this grouping, a number of alternative topologies being excluded, and although support for this grouping in Moore et al. (2007) was somewhat less strong; Ceratophyllum, with quite a long branch, was sister to eudicots (in some reconstructions, Piper, with a very long branch, was also involved). Similarly, Chloranthus was found to be sister to the magnoliids with moderate to strong support (Jansen et al. 2007), a position also found by Saarela et al. (2007), many analyses in Moore et al. (2007, 2010), and Smith et al. (2010).

However, in Moore et al. (2010) there were a variety of other "minority" relationships - Ceratophyllum or [Ceratophyllum + Piper] as sister to monocots, and Chloranthaceae sister to a clade including magnoliids, monocots, and eudicots. Goremykin et al. (2009b) recently found a [Ceratophyllum + magnoliid] clade (Chloranthaceae were not included) after removing 2,500 highly variable positions from the analysis of chloroplast genome data; with the removal of 1,000 positions Ceratophyllum was sister to a [monocot + eudicot] clade. Analyses of morphological data gave a grouping [Ceratophyllum + Chloranthaceae] with quite strong support (Endress & Doyle 2009; see also Huang et al. 2010, weak support using the ycf2 gene; Moore et al. 2011, very weak support). Finally, in the massive parsimony analysis of Goloboff et al. (2009) a set of relationships [Ceratophyllum [[Amborella + Nymphaeales]....[[Chloranthaceae + monocots] eudicots]]] was retrieved, Duarte et al. (2010) found the relationships [monocots [magnoliids + eudicots]], and Tamura et al. (2011: 6 plastid, 7 mitochondrial, and 1 nuclear genes sequenced) found some support for a [Piperales + monocot] clade, as did Barrett and Davis (2011: or [Piperales + Ceratophyllales] the sister group).

A further complication was introduced by Lee et al. (2011) who found that monocots were sister to all angiosperms minus Nymphaeales and Amborellales. Although this may be a sampling problem (no Austrobaileyales, Chloranthales or Ceratophyllales were included), Lee et al. (2011) initially looked at almost 23,000 sets of orthologues from nuclear genomes of 101 genera of land plants (most taxa represented by ESTs, genes included if represented in as few as 4 taxa).

Nevertheless, pending further studies, the relationships [[Chloranthales + magnoliids] [monocots [Ceratophyllales + eudicots]]] are used as a basis for further discussion, rather different from the relationships suggested in the first six editions of this site and also from the tree in A.P.G. II (2003) and Soltis et al. (2005b). Direct links are provided to the pages where the clades just mentioned are discussed further; this discussion will generally be just above or below where the link takes you: Ceratophyllales, Chloranthales, eudicots, and monocots.

[CHLORANTHALES [[MAGNOLIALES + LAURALES] [CANELLALES + PIPERALES]]]: sesquiterpenes +; seed endotestal.

[[MAGNOLIALES + LAURALES] [CANELLALES + PIPERALES]] / MAGNOLIIDS / MAGNOLIANAE Takhtajan: (neolignans +); vessels solitary and in radial multiples; lamina margins entire; A many, spiral [possible position here], extrorse; antipodal cells ephemeral, hypostase +, nucellar cap +, raphal bundle branches at the chalaza.    Back to Main Tree

4 orders, 20 families, 9900 species.

Evolution. Divergence & Distribution. Crown group magnoliids diverged 134.5-125.7 million years ago (Moore et al. 2007), while Bell et al. (2010) suggest ages of (138-)122(-108) or (130-)125(-121) million years depending on the method used. Magallón and Castillo (2009) suggest that crown group divergence was ca 201.7-198.2 and 128-127.7 million years ago - relaxed and constrained penalized likelihood datings respectively, the stem group age being only a little older than the higher figures. An earlier fossil-based estimate for both stem and crown divergence is ca 98 million years (Crepet et al. 2004: magnoliids sister to monocots). Other estimates are somewhat intermediate, ranging from (181-)155(-136) million years old (with eudicot calibration) to (198-)163(-138) million years (without: Smith et al. 2010) or 132-122 million years (Wikstöm et al. 2001).

For summaries of the fossil history of the group, espeacially prominent in the Mid Cretaceous and later, see Friis et al. (1997, 2006).

Plant-Animal Interactions. Caterpillars of Papilionidae-Papilioninae butterflies are notably common (almost 33% of the records) on members of this group; they are, however, apparently so far not known on Myristicaceae, in Laurales they predominate on Lauraceae, and in Piperales on Aristolochiaceae (see Scriber et al. 1995 for references; Zakharov et al. 2004). They are also commonly found on Rutaceae, which have similar alkaloids.

Chemistry, Morphology, etc. Hegnauer (1990) discussed the chemistry of the Polycarpicae, which includes Austrobaileyales and Ranunculales; similar isoflavonoids are found in Magnoliaceae, Lauraceae and Chloranthaceae.

For a convenient summary of a number of features of wood anatomy, see Herendeen et al. (1999b). Magnolioid roots are scattered through the clade, being recorded from a number of woody members of Laurales and Magnoliales. These roots are stout, often >3 mm across, lack root hairs (?always), and are endomycorrhizal - associated conditions (see Baylis 1975; St John 1980). Taxa with braodly lobed but otherwise entire leaf blades are scattered throughout this clade (but they are not known from the small Canellales). See Erbar (1983, inc. Illicium) for carpel morphology and Ronse Decraene and Smets (1992b) for androecial morphology.

Phylogeny. Although the sister group relationship of Piperales with Canellales in particular is at first sight unexpected, the magnoliid clade as a whole and the relationships within the group are turning out to be quite robust. There was not - and still really is not - much if any morphological support for this grouping (Doyle & Endress 2000; see also the characterisation above); features like tectum structure, etc., show considerable variation (J. A. Doyle 2005). However, molecular support for the clade has been increasing in successive studies (e.g. Qiu et al. 1999, 2000, 2005 [but note variability in levels of support depending on the analysis, the node sometimes collapses], 2006b, 2010; Zanis et al. 2002; Jansen et al. 2006b; Zhengqiu et al. 2006; Cai et al. 2006; Müller et al. 2006, support for [Canellales + Piperales] rather poor; Jansen et al. 2007, but little maximum parsimony support; Moore et al. 2007, but not appearing in all analyses; Soltis et al. 2007a; Soltis et al. 2011); this support includes the possession of unique indels (Löhne & Borsch 2005). Soltis et al. (2007a) found weak support for a grouping [Magnoliales + Canellales]; relationships in Bell et al. (2010) are [Piperales [Laurales [Canellales + Magnoliales]]], although they have little support.

[MAGNOLIALES + LAURALES]: cuticle waxes as annularly-ridged rodlets, palmitol the main wax; A whorled; pollen with lamellate endexine; carpel cross-zone initiated late; ovules lateral.   Back to Main Tree

Phylogeny. Divergence & Distribution. Magallón and Castillo (2009) suggest that the two clades diverged ca 198.2 and 127.7 million years ago - relaxed and constrained penalized likelihood datings.

Chemistry, Morphology, etc. There is considerable variation as to how the androecium is initiated; it would not take much for "flowers with inner staminodes" to be an apomorphy at this level.

MAGNOLIALES Bromhead  Main Tree, Synapomorphies.

Vessels in multiples; secondary phloem stratified; pith septate [with sclerenchymatous diaphragms]; nodes 3:3; petiole vasculature an arc with an adaxial plate; branching from the current innovation; leaves two-ranked, lamina vernation conduplicate; "bract" sheathing; P whorled; G occluded by fusion and secretion; ovule with outer integument 5-10 cells across, obturator +; seeds medium-sized, testa vascularized, multiplicative; endosperm ?type, irregularly ruminate. - 5 families, 154 genera, 2929 species.

Evolution. Divergence & Distribution. Magallón and Castillo (2009) suggest that stem group Magnoliales are ca 198.2 and 127.7 million years old - relaxed and constrained penalized likelihood datings - and equivalent ages for the crown group are 171.5 and 116.6 million years.

There is extensive discussion on character evolution in Sauquet et al. (2003), and much of the character hierarchy here is based on this paper. Note, however, that where characters like extrorse/introrse anther dehiscence and ruminate/non-ruminate testa are placed on the tree depends on how the characters are optimised, or even defined (e.g. ruminate endosperm), although detailed anatomical-developmental work might clarify such issues. Furthermore, there is some conflict with the positions of characters as they are optimised on a more extensive tree for basal angiosperms - although less detailed for Magnoliales (cf. Ronse De Craene et al. 2003, also Judd et al. 2003). Doyle and Endress (2000) and Soltis et al. (2005) suggest additional characters for the clade, including reduced fibre pit borders, palisade parenchyma, foliar astrosclereids, and pollen with continuous tectum. Some of these may well need to be added to the apomorphies above.

Chemistry, Morphology, etc. Intra- and interfamilial variation of morphological characters that are frequently used to reconstruct phylogenies is considerable, in particular, Furness et al. (2002) emphasize the variability of microsporogenesis in the order.

The "bracts" with sheathing bases that have been reported for a number of Magnoliales are interesting (Endress & Armstrong 2011: condition in Degeneriaceae?), given that leaf bases in the clade tend to be ordinary, i.e., quite narrow - except for the stipulate Magnoliaceae. Deroin (2010) noted a tendency for the vasculature of the perianth and/or androecium in some Annonaceae and Magnoliaceae to be pentamerous.

For additional information, see Sugiyama (1976a, b, 1979: esp. considerable variation in cotyledonary nodal morphology), Hiepko (1964b: perianth vasculature), Endress (1977b, 1986a, 1994a: floral morphology), van Heel (1981, 1983: carpel development), Erbar and Leins (1983: floral development), Metcalfe (1987: general anatomy), Taylor and Hickey (1995: general), Ronse Decraene and Smets (1996a: androecium), Kimoto and Tobe (2001: embryology), and Kim et al. (2003, 2004: AP3 and P1 genes, note that Myristicaceae are unsampled).

Phylogeny. Molecular data suggested that Myristicaceae are sister to the rest of the order, but support was only moderate (D. Soltis et al. 2000); the addition of morphological data strengthened that position, and also placed Magnoliaceae as sister to the remaining taxa (Doyle & Endress 2000), although the latter position had only moderate support (cf. P. Soltis et al. 2000; see also Sauquet et al. 2001). The family pairs [Annonaceae + Eupomatiaceae] and [Degeneriaceae + Himantandraceae] are both well supported (D. Soltis et al. 2000; P. Soltis et al. 2000; Doyle & Endress 2000). Other studies confirm these relationships (Sauquet et al. 2003; Müller et al. 2006), although Hilu et al. (2003: matK analysis alone) suggested a somewhat different set of relationships, but the sampling was poor and there was scanty support for the critical nodes, and Qiu et al. (2010: mitochondrial genes) also found a somewhat different topology, Degeneriaceae being sister to the rest of the order minus Myristicaceae, although support for this position was weak. Soltis et al. (2011) found some support for the position of Magnoliaceae as sister to the rest of the order, the clade [Degeneriaceae + Myristicaceae] being well supported although otherwise support was weak; recovery of this unexpected topology was ascribed to signal from rDNA sequences, while Morton (2011: nuclear gene Xdh) found a set of relationships [Himantandraceae [Myristicaceae [Degeneriaceae [the rest]]]].



Includes Annonaceae, Degeneriaceae, Eupomatiaceae, Himantandraceae, Magnoliaceae, Myristicaceae.

Synonymy: Annonales Berchtold & Presl, Degeneriales C. Y Wu et al., Eupomatiales Reveal, Himantandrales Doweld & Shevyryova, Myristicales Berchtold & Presl - Magnoliineae, Myristicineae Chatrou - Annonanae Doweld, Magnolianae Takhtajan - Magnoliidae Takhtajan - Magnolipsida Brongniart - Magnoliid I group (Nandi et al. 1998).

MYRISTICACEAE R. Brown   Back to Magnoliales

Tree with whorled branches, branches plagiotropic; exudate red; isoflavonoids, flavonoids diverse [flavones +], polyketides [acetogenins], (tryptamine alkaloids) +, isoquinoline alkaloids 0; cork also in outer cortex; primary stem ± with continuous cylinder; vessel elements with simple or scalariform(-reticulate) perforations; tannin-containing tubes in the xylem; sieve tubes with non-dispersive protein bodies, (plastids with protein crystalloids and starch); petiole bundles bicollateral; (branched) sclereids or fibres +; (acicular) crystals +; hairs branched or stellate (T-shaped); cuticle waxes as platelets; (leaves spiral), (lamina lobed - some Knema); plants dioecious; flowers and receptacle small; P (2-)3(-5), connate; staminate flowers: A whorled (spiral), 2-40, connate (not), (anthers unithecate); (pollen inaperturate, ulceroid, spiraperturate, ektexine granular), pollen aperture membrane sculpted; pistillode 0; carpellate flowers: staminodes 0; G 1; (stylulus ± long); ovule 1/carpel, subbasal, inner integument (3-)7-10 cells across; fruit a follicle, dehiscing abaxially as well, (indehiscent); seed large, pachychalazal, aril +, micropylar-funicular, vascularized, (aril small); testa multiplicative, exotesta thick-walled, endotesta palisade, lignified, crystalliferous, tegmen multiplicative or not, exotegmen with fibres or sclerotic or tracheidal cells, chalaza massive, vascularized, with a lignified counter-palisade; endosperm nuclear, with (starch and) oil; embryo with hypocotyl not developed; n = 19, 21, 22, 25, 26; germination hypogeal.

Myristicaceae

20[list]/475: Myristica (175), Horsfieldia (100), Knema (95), Virola (60). Pantropical. (map: from de Wilde 2000 [Indo-Malesia]; Heywood 2007).[Photo - Carpellate flower, Fruit].

Evolution. Divergence & Distribution. Diversification within the family may be recent, within 21-15 million years before present, although this seems very recent indeed given the age of the family, it distribution throughout the humid tropics, and its apparently low dispersability (J. A. Doyle et al. 2004), unfortunately, the recent discovery of fossil seeds apparently of Myristicaceae from the Eocene (London Clay) does not solve the problem because they cannot be placed accurately on the phylogenetic tree of Myristicaceae (J. A. Doyle et al. 2008a). Note that although there is a long branch leading to the family, there is little molecular divergence between its extant members (Sauquet et al. 2003); this might indeed suggest recent diversification. In this respect Myristicaceae can be compared with Annonaceae, also pantropical if a little younger; there diversification may have occurred very much earlier some 84-57 million years before present (Doyle et al. 2004; Scharaschkin & Doyle 2005; esp. Couvreur et al. 2011a).

Genes & Genomes. Isozyme duplication (in Myristica) suggests ancient polyploidy (Soltis & Soltis 1990).

Chemistry, Morphology, etc. There are tannin-containing tubes in the xylem, and free phloem strands in the center of the midrib bundle. The wood rays are not notably broad.

The inflorescence morphology seems to me a little confusing, with a large, abaxial "bracteole" being drawn by Ronse de Craene (2010; see also Armstrong & Tucker 1986); this is addition to paired lateral prophyllar/bracteolar-like structures that may (staminate inflorescence) or not subtend buds. There are three traces per perianth lobes (Siddiqui & Wilson 1976). The synandrium of Myristica has a sterile apex, probably axial (Armstrong & Tucker 1986), and judging from the vasculature, the unithecal anthers found in some species of Myristica may represent divided bithecaal anthers (Wilson & Maculans 1967; Armstrong & Tucker 1986); anthers in Knema are tetrasporangiate (Xu & Ronse de Craene 2010a). For the complex vascularisation of the carpel in Myristica, see Wilson and Maculans (1967), and for the vascularization of the ovule in Myristica fragrans, see Dickison (2000). A small structure developing between the two integuments is reported by Nair (1972). Corner (1976) commented on the complexity of the seed coat in Myristicaceae which is as elaborate as that of any other angiosperm. Not all taxa appear to have a multiplicative testa, and some have a multiplicative tegmen; the ruminations may develop best from the pachychalazal part of the seed.

For more information, see Hegnauer (1969, 1990: chemistry), Mauritzon (1939a) and Endress (1973), both seed, Kerster and Baas (1981: anatomy), Kühn and Kubitzki (1993: general), de Wilde (2000: Malesian Myristicaceae), Jiménez-Rojas et al. (2002: growth patterns - Massart's model), Sauquet (2003: androecium), Chatrou (2003: apomorphies), and Sauquet and Le Thomas (2003: pollen).

Phylogeny. Relationships within the family are unclear. The African and Madagascan taxa may form a clade, possibly sister to Compsoneura (but perhaps long branch attraction), overall, geography and relationships may be summarized as [[[Asia] + [America]] [Madagascar and Africa]]. Within the Asian/Malesian representatives, Knema and Myristica may be sister taxa. Sauquet et al. (2001, 2003), Sauquet (2003) and Sauquet and Le Thomas (2003) suggest that the free stamens (in some species they are numerous and apparently spirally inserted) and small aril of Mauloutchia, apparently plesiomorphic features, are in fact more likely to be derived.

[Magnoliaceae [[Himantandraceae + Degeneriaceae] [Eupomatiaceae + Annonaceae]]]: primary stem with distinct bundles [eustele]; wood with broad rays; flowers solitary, large, receptacle well-developed, cortical vascular system +; P = K + C; A many, spiral [possible position here], filaments with three veins, anther thecae separate, embedded in the broad filaments, the connective prolonged; G spiral; ovule with funicular obturator; 10-aa deletion in PI-derived motif in AP3 gene.

Evolution. This clade may have diversified 120-100 million years before present (Doyle et al. 2004).

Chemistry, Morphology, etc. There is more than one kind of cortical vascular system in this clade, and what parts of the flower are supplied by which system varies (Ronse De Craene et al. 2003, see also Deroin 1991, 1999a).

MAGNOLIACEAE Jussieu   Back to Magnoliales

(Deciduous); (silicon concentration high), sesquiterpene lactones +; vessel elements with simple and scalariform perforation plates, (walls vestured - Liriodendron); wood fluorescing; (secondary phloem ± stratified, rays broad); (sieve tube plastids with protein crystalloids and starch); nodes 6 or more:6 or more; secretory cells 0; petiole also with medullary bundles (strictly annular - Liriodendron); leaves also spiral, lamina vernation laterally or vertically conduplicate (supervolute-curved), stipules sheathing stem, open opposite petiole; flowers terminal or axillary; P 3 + 3 or 3 + many, K and C ± distinguishable; (A introrse - Magnolia), (filament vein single), tapetal cells multinucleate; pollen grains boat-shaped; G 1-many, fusion complete, but secretory canal +, stigma (terminal), dry, elongate (not); ovules(1-)2-12(-16)/carpel, micropyle exo/bistomal, outer integument 4-10 cells across, inner integument 2-3(-4) cells across, parietal tissue ca 8 cells across; fruit dry; testa vascularized [inconspicuous in Liriodendron], endotesta lignified; endosperm not ruminate; n = 19.

Magnoliaceae

2[list]/227: Magnolia (ca 225). The Americas (but not W. North America), and South East Asia to Malesia (map: from Good 1974; Lozana-Contreras 1994).[Photo - Collection]

Evolution. Divergence & Distribution. Archaeanthus, a Cenomanian fossil of some 98 million years in age from North America, may be assignable to Magnoliaceae (Dilcher & Crane 1984), and it is placed sister to the family in the constrained morphological analysis of Doyle and Endress (2010). It has follicular fruits with 10-18 seeds carpel. Lesqueria, from about the same period, may also belong around here; it has similar fruits (Crane & Dilcher 1984). The young fruits of Protomonimia, with several carpels borne in spirals on a flattened axis, the young seeds having a palisade ?exotesta with inpushings of the anticlinal cell walls and ?mesotestal cells with sinuous anticlinal walls (Nishida & Nishida 1988), may be of a taxon somewhere near Magnoliaceae: If a member of Laurales, it would probably be stem Laurales. Liriodendron was widely distributed in the Northern Hemisphere in the early Tertiary (Ferguson et al. 1997).

Plant/Animal Interactions. Beetles are common pollinators, and the flowers may be thermogenic (Seymour 2001); floral scents of the family have been extensively studied (Azuma et al. 1999).

Genes & Genomes. Both isozyme duplication and stomatal size increase over time suggest ancient polyploidy (Soltis & Soltis 1990; Masterson 1994).

Chemistry, Morphology, etc. Sclerenchymatous diaphragms are particularly conspicuous in the Michelia group of Magnolia. Terminal tracheids are commonly silicified, forming distinctive phytoliths (Piperno 2006).

The micropyle is exostomal in the Michelia group of Magnolia. The endexine is lamellate and the columellae fused-granular (Xu & Kirchoff 2008). Nectar is secreted from the exposed surfaces of the carpels in some species of Magnolia. The seed of Magnolia may dangle from the open fruitlet attached by extended annular thickenings of vessels in the protoxylem.

In Liriodendron the leaves are lobed, with palmate secondary veins, the petiole bundle is strictly annular, the fruit is a samara, the seeds lack a sarcotesta, the mesotesta being more or less sclerotic, and the endosperm is slight and not ruminate. In Magnolia the leaves are entire, the anthers are introrse, the ovule lacks a funicular obturator(?), the dehiscence of the fruit varies, but it is not adaxial (so the fruit is not a follicle in the strict sense), it may dehisce abaxially, there is a sarcotesta and a scleroendotesta with crystals and lignified fibrils in the cells, and there is much endosperm. There is also a simple or tubular pore in the seed coat that marks the passage of the vascular bundle through the sclerotesta (Xu 2003).

Some information is also taken from Hegnauer (1969, 1990: chemistry), Bouman (1977: ovules and seeds of Liriodendron), Nooteboom (1993: general), Yamada et al. (2003b: ovules), Pan et al. (2003: embryology), Xu (2006) and Xu and Rudall (2006: floral development), and Xu and Kirchoff (2008: pollen morphology); see Azuma et al. (2000), Ueda et al. (2000) and Wang et al. (2006) for molecular phlogenies, Figlar (2000) for branching patterns, Charlton (1994) and Liao and Xia (2007) for phyllotaxis and vernation.

Phylogeny. Magnolia s. str. is pretty wildly paraphyletic (Qiu et al. 1995; Azuma et al. 2001 [section Talauma is sister to the rest of Magnolia], Kim et al. 2001a, b; Azuma et al. 2001; Nie et al. 2008 [again section Talauma is sister to the rest, but otherwise support along the backbone is poor]).

Taxonomy. Generic limits around Magnolia s. str. are unclear. Although section Talauma often appears to be sister to the rest of the genus in phylogenetic analyses (see above), it seems best to expand the genus to encompass the whole of the current Magnolieae. The family thus includes only two genera and tribes are superfluous. See also Nooteboom (2000) for suggestions about classification, and for a checklist and bibliography (under old generic names), see Frodin and Govaerts (1996).

Synonymy: Liriodendraceae F. Barkley

[[Himantandraceae + Degeneriaceae] [Eupomatiaceae + Annonaceae]]: anthers valvate, staminodes internal, secretory; pollen atectate, psilate, with granular infratectum, total exine thin, 0.2-1.3 µm thick; style with pollen tube transmission tissue differentiated; fruit indehiscent.

Evolution. Divergence & Distribution. Endressinia brasiliana, from the Brazilian Crato formation of some 113 million years ago has i.a. distinctive glandular staminodia. It is placed sister to this clade by Doyle and Endress (2010), largely confirming the position suggested by Mohr and Bernardes-de-Oliviera (2004).

Chemistry, Morphology, etc. Doyle (2007) notes that the venation of members of this clade is often poorly differentiated and so of low rank (but cf. many Annonaceae, albeit not "basal" members) - I have not placed this character on the tree. For the inner staminodes that are often so conspicuous in the flowers, and their function in pollination - food for pollinators, attractants - see Endress (1984).

[Degeneriaceae + Himantandraceae]: flowers axillary; outer integument with annular opening; x = 12.

DEGENERIACEAE I. W. Bailey & A. C. Smith   Back to Magnoliales

Alkaloids 0; cork ?; vessel elements with scalariform perforation plates; sieve tube plastids with protein crystalloids and starch; nodes 5:5; petiole also with medullary bundles; secretory cells 0; cuticle waxes as platelets; leaves spiral; bracts?; K 3, C many, whorled; pollen boat-shaped; G 1, basally ascidiate, occluded by fusion only; ovules many/carpel, outer integument 7-8 cells across, inner integument ca 3 cells across, funicle long; fruit a follicle; exotestal cells palisade, thin-walled, sarcomesotesta +, endotesta with lignified internal fibrils; suspensor massive, embryo with 3 (4) cotyledons.

Degeneriaceae

1[list]/2. Fiji, Viti Levu (map: original!).

Chemistry, Morphology, etc. Some information is taken from Hegnauer (1973, 1990, as Winteraceae: chemistry) and Kubitzki (1993b: general).

HIMANTANDRACEAE Diels   Back to Magnoliales

Polyketide alkaloids only; vessel elements with simple (and scalariform) perforations; sieve elements with non-dispersive protein bodies; trichomes peltate; cuticle wax crystalloids 0; branching?; P or bract + bracteoles enclosing the flower in 2 series, caducous, or 2 connate + 4 connate; "C" 3-23 [staminodial]; pollen scabrate; additional staminodes with glands; G 6-30, (1) 2 pendulous ovules/carpel; fruit ± syncarpous, a drupe with several stones; seeds flattened, testa not multiplicative, mesotesta fibrous, endotesta aerenchymatous; endosperm development?; seedling?

Himantandraceae

1 (Galbulimima)[list]/2. The Celebes, New Guinea and N.E. Australia (map: from Hoogland 1972; Endress 1983). [Photo - Flower, Fruit, Buds, Habit]

Chemistry, Morphology, etc. For the testa, see Doweld and Shevyryova (1997); the ruminations are at best obscure. The perianth may be staminodial in origin; according to Johri et al. (1992) the outer integument is not vascularized. Some other information is taken from Hegnauer (1966, 1989: chemistry) and Endress (1993: general).

[Eupomatiaceae + Annonaceae]: sieve tube plastids with protein crystalloids and starch; rays 8-15-seriate; petiole bundles arcuate; trunk leaves spiral; prophyll single, adaxial; inflorescence +; fruit ± berry-like; mesotesta fibrous.

Chemistry, Morphology, etc. Optimisation of prophyll position on the tree is uncertain. Adaxial prophylls are common in Annonaceae, including Anaxagorea, although some taxa have paired, lateral prophylls (Fries 1911).

EUPOMATIACEAE Orban, nom. cons.   Back to Magnoliales

Also rhizomatous, with xylopodium or root tubers; pith not septate; vessel elements with scalariform perforation plates; secondary phloem stratification?; sieve tubes with non-dispersive protein bodies [check], plastids also with protein rods; nodes (5-)7(-11):(5-)7(-11); secretory cells +; (stomata anomocytic); branching?; inflorescence fasciculate; receptacle concave, calyptra +, thick, deciduous, with sclereids, P 0; A introrse; pollen with encircling equatorial sulcus, exine homogeneous; petaloid staminodes 15+, connate basally and with fertile stamens, with many glands; G many, ± connate, occluded by fusion only, placentation sublaminar, stigma flat, papillate; ovules 2-11/carpel; aggregate fruit +; testa ?not vascularized, exotestal cells with thickened unlignified walls, endotesta unlignified, exotegmic cells cuboid, slightly lignified, endotegmic cells enlarged, crushed; n = 10; germination epigeal/phanerocotylar.

Eupomatiaceae

1[list]/3. New Guinea and E. Australia (map: from Hoogland 1972; Endress 1983). [Photos - Flower, Flower]

Chemistry, Morphology, etc. The main axes are mixed, initially being orthotropic and bearing spirally-arranged leaves, later being more or less plagiotropic and with distichous leaves. There are about three two-ranked bracts on the pedicel; the calyptra itself is often interpreted as being a modified, amplexicaul bract (e.g. Endress 2003b; esp. Kim et al. 2005b); although it might seem that such a dramatic change in leaf insertion was unlikely, sheathing bracts seem to be common in Magnoliales (Endress & Armstrong 2011). There is no evidence of isozyme duplication (Soltis & Soltis 1990).

Some information is taken from Hegnauer (1966, 1989: chemistry), Endress (1983, 1993: general) and Rix and Endress (2007 - an easy account).

ANNONACEAE Jussieu   Back to Magnoliales

Vessel elements with simple perforation plates; epidermal cells with single crystals [?level]; flowers with open development; P very thick, 3, ± calycine, often? valvate, + 3 + 3, ± corolline, open to valvate; A whorled, filaments with a single vein; funicle short; endotestal plug +, mesotesta with several layers of longitudinal and then of transverse fibres, aerenchyma internal; vascular bundle in antiraphe; seed ruminations irregular.

Anaxagorea

129/2220 [list] - three groups below. Largely tropical.

1. Anaxagorea

Uniseriate hairs terminate in a rounded cell, otherwise hairs ± stellate; trunk leaves distichous; (petiole vascular tissue ± annular; with adaxial bicollateral plate; small arcuate bundles variously arranged); stomata allelocytic; staminode secretions?; pollen with granular infratectum; G 3, receptacular vascular system 0; ovules 2/carpel; fruit an explosively dehiscent follicle; endotesta aerenchymatous, tegmen alone involved in ruminations, with oil globules; n = 8.

1/21. Tropical America, Sri Lanka to West Malesia, also Halamahera and Ceram (map: from Maas & Westra 1984, 1985). [Photo - Flower, Fruit, Fruit.]

Other Annonaceae

[The Ambavia clade + The Rest]

(Outer staminodes +), internal staminodes 0; ovules with outer integument 4-6 cells across, (middle integument 4-6 cells across), inner integument 2-4 cells across, amyloid [xyloglucans] in endosperm +. (map: from Aubréville 1974a.)

2. The Ambavia clade

Anther connective ± tongue-like; carpels 3-8, stipe articulated, (extragynoecial compitum + - Cananga), ovules (1)2-16/carpel, integuments 3; (endosperm ruminations lamelliform); n = 7.

9/50: Drepananthus (26). Tropical, inc. Madagascar.

3. The Rest

(Lianes; shrubs; deciduous); acetogenins [antimicrobial] +, neolignans 0?; terminal cell of uniseriate hairs pointed (hairs stellate); (primary stem with continuous cylinder); cambium storied; (sieve tube plastids also with protein fibres); petiole bundle also annular; branched sclereids or fibres common; secretory cells +; (prophyll single, adaxial); petioles short; inflorescence cymose (leaf opposed), or flowers axillary; flowers (medium in size), (2-)3(-4)-merous; A (³3), (latrorse, introrse), anther connective peltate-truncate, (thecae septate); tapetal cells plasmodial, multinucleate; (microsporogenesis successive); (pollen with circular sulcus), (trinucleate); G (3-)many (syncarpous; parietal placentation), spiral, when 3 opposite outer P, stigma capitate to U-shaped, wet, extragynoecial compitum common; ovules 1-many/carpel, (tritegmic), (nucellar cap 0); fruit berrylets (follicles; carpels coalescent); seeds (with micropylar aril), ruminations mostly regularly spiniform or lamelliform, also involving testa; (sarcotesta +), endotesta crystalliferous, (with thin walled, longitudinal fibres); n = (7) 8-9; germination epigeal/crypto- or phanerocotylar.

3a. The Short Branch Clade/ malmeoids

(Pollen globose, ± spiny, disulcate, ektexine columellate); (tegmen with oil globules); endosperm ruminations spiciform (lamelliform); (mesotesta with transverse fibres only - Polyalthia).

Polyalthia (150), Pseuduvaria (50), Unoniopsis (45). Lowland tropics.

3b. The Long Branch Clade / annonoids

(Climbers); pollen inaperturate (Clade A: Trunk leaves distichous; anthers locellate; pollen in tetrads or polyads); endosperm ruminations lamelliform.

Guatteria (280), Annona (120-175, soursop, sweetsop: inc. Rollinia), Xylopia (100-160), Uvaria (110-150), Artabotrys (100), Goniothalamus (50-120), Duguetia (70-95), Fissistigma (60), Monanthotaxis (55), Friesodielsia (50-60: ?monophyletic). Predominantly lowland tropics, rarely temperate.

Synonymy: Hornschuchiaceae J. Agardh, Monodoraceae J. Agardh [where?]

Evolution. Divergence & Distribution. Estimates of the age of stem Annonaceae vary considerably: (110.4-)106.3(-102.0) million years (Couvreur et al. 2011a: HPD estimates), (110.4-)101.7(-99.4) million years old (Surveswaran et al. 2010: HPD estimates), (101.46-)98.01(-94.91) million years old (Su & Saunders 2009), 91-82 ± 4 million years old (Wikström et al. 2001), or somewhere between 84.7 and 62.6 million years (Erkens et al. 2009: a variety of estimates); see also Pirie and Doyle (2012). Crown group diversification may have occurred 92-89 million years ago (Su & Saunders 2009) or some 82-57 million years before present (Doyle et al. 2004; Scharaschkin & Doyle 2005 [84 million years before present]; Couvreur et al. 2011a: additonal estimates).

Divergence within Anaxagorea may have begun ca 44 million years before present (Scharaschkin & Doyle 2005). Diversification of the bulk of the family, i.e. not including Anaxagorea, may be Tertiary, figures of (76.6-)68, 64.5(-53.0) million years for the divergence between the ambavioids and the rest being suggested by Erkens et al. (2009), or some time between 84.4-63.6 million years ago (Su & Saunders 2009; see also Couvreur et al. 2011a for other estimates). Crown group ambavioids themselves diversified perhaps (80.9-)57.0(-52.3 million years ago (Surveswaran et al. 2010: q.v. for other dates, HPD estimates) or (78-)69.4(-60.2) million years (Couvreur et al. 2011a: HPD estimates). Futabanthus, based on fossil flowers from the late Cretaceous (Early Coniacian, ca 89 million years ago) of Japan, can perhaps also be assigned to this clade (Takahashi et al. 2008b).

J. A. Doyle et al. (2004), Richardson et al. (2004), Couvreur et al. (2011a) and others discuss the historical biogeography and diversification of the family. The stem node of the [long + short branch clades] may date to 70-65 million years before present and the crown node to 66.7-56.6 ± 2.3 million years before present ([78.08-]67.33[-55.22] million years - see Su & Saunders 2009; Couvreur et al. 2011a for estimates of this and other major clade divergence times). In any event, much divergence is largely after the break-up of the continents and Tertiary in age. Couvreur et al. (2011a) contrast the diversification patterns of the long- and short-branch clades, the long-branch clade beginning to diversify ca 66 million years ago and the short branch clade much later, ca 33 million years ago; the former has about twice as many species as the latter, even if rather paradoxically the latter may have a higher diversification rate - once it got going (see Couvreur et al. 2011a for other estimates of the ages of these clades, some of which do not show such a pronouced gap before beginning diversification). Su and Saunders (2009) give dates for additional splits in the family, focusing on Pseuduvaria. The large New World genus Guatteria has undergone much diversification in (Amazonian) South America perhaps 8.8-4.9 million years before present, having moved to South America from Central America, perhaps from Africa via Europe (Erkens 2007; Erkens et al. 2007a, b, 2009).

Ecology. Annonaceae are an important liane group of the tropical South East Asian forests (Gentry 1991), and there have been several origins of this habit there within the long-branch clade (Meinke et al. 2011). Artabotrys, one of these liane clades, has curved hooks which represent modified inflorescences; the orthotropic stems may bear stout, paired and rather vicious thorns. The other clades do not have hooks or thorns, and the development of the liane habit varies.

Floral Biology & Seed Dispersal. Saunders (2010: good photographs) summarizes variation in floral morphology in the context of pollination. Flowers in which the stamens are enclosed in some sort of bowl-shaped structure are common, and floral rewards are often found on the inner surfaces of the inner perianth whorl. In Neotropical Annonaceae pollination of the more or less odoriferous flowers is predominantly by a variety of beetles, although pollination by flies (e.g. Su et al. 2008) and thrips is also known; there is considerable variation in the scents produced (Goodrich & Raguso 2009). Beetle pollination is common throughout the family, and it has been suggested that the tough, apically expanded connective that predominates in stamens of flowers of group 3 above may protect the pollen from the depradations of unwanted visitors (Gottsberger 1999; Silberbauer-Gottsberger et al. 2003). The flowers of some Annonaceae show thermogenesis (Seymour 2001), and some kind of extracarpellary compitum is common (Deroin 1991). I do not know if the staminodes of Anaxagorea produce any secretions/floral rewards.

Dispersal of fruit is predominantly by mammals and birds.

Vegetative Variation. Growth patterns are interesting. The main axes of Annonaceae are often mixed (?Troll's model), initially being orthotropic and with spirally-arranged leaves, later being more or less plagiotropic with two-ranked leaves (see also Eupomatia). In some Xylopia, etc., the plagiotropic branch systems are frondose and very beautiful. Taxa such as Asimina and Annona have orthotropic axes with distichous leaves. In general, axillary branches depart from the stem slightly to one side of the axil.

Chemistry, Morphology, etc. Protoberberidine alkaloids are reported from Annonaceae (Wink 2008). Cork in the roots of Goniothalamus may be superficial (Blunden & Kyi 1974). The wood occasionally fluoresces. Terminal tracheids are fairly commonly silicified (Piperno 2006). For the petiolar anatomy of Anaxagorea, see Jovet-Ast (1942) and Maas and Westra (1984); it sseem unlikely that there are annular petiolar bundles, although the midrib anatomy can be complex. Tetrameranthus sometimes seems to have opposite leaves; according to George Schatz (pers. comm.) it lacks plagiotropic branches and has all branches with spirally-arranged leaves.

There are reports that the pollen grains of both Annona and Asimina may germinate via the proximal pole (Hesse et al. 2009a), although in the former genus and some others rotation of individual grains of the tetrad during development has been noted (Tsou & Fu 2002, 2006; Lora et al. 2009b). Xylopia, like Anaxagorea also with staminodes and follicles, has micropylar-arillate seeds (Corner 1976: other genera may also have reduced arils - see Svoma 1997). Reports that the micropyle is formed by both integuments need to be confirmed (cf. Svoma 1998b). Monodora (derived) has a rather thin outer integument, and the seed coat anatomy of those taxa with an extra integument (Zwischenintegument) can be complex (Christmann 1986); I have not thought of this in the context of the phylogeny above.

Additional information is taken from Fries (1919: inflorescence morphology), Jovet-Ast (1942: indumentum and anatomy), Corner (1949: seed anatomy), van der Wyck and Canright (1956: anatomy), Hegnauer (1964, 1989: chemistry), Okada and Ueda (1984: cytology), Morawaetz (1986, 1988: esp. cytology), Deroin (1991b: stigmatic surface and carpel morphology), Endress (1975: carpel and stamen development), Kessler (1993: general), Steinecke (1993: esp. flower and fruit), Rudall and Furness (1997: tapetum), Svoma (1998a: seeds), Svoma (1998b) and Endress (2011b), both ovules, Deroin (1999a: receptacular vascular system), Doyle et al. (2000: pollen evolution), Bakker (2000a, b: general); Johnson (2003: leaf insertion), Junikka and Koek-Noorman (2007: bark anatomy), Couvreur et al. (2008a: evolution of syncarpy, etc.), Couvreur et al. (2008b: pollen in the Monodora group), Sun et al. (2008: epidermal anatomy), and Lora et al. (2009: infraspecific variation in number of pollen nuclei in Annona cherimola. For floral (developmental) morphology, see van Heusden (1992), Ronse Decraene and Smets (1990b), Leins and Erbar (1980, 1982, 1996, 2010), Xu and Ronse De Craene (2010b) and Endress and Armstrong (2011: Anaxagorea).

Phylogeny. See Doyle and le Thomas (1994, 1996) for morphological phylogenies and character evolution, and for a molecular phylogeny of Anaxagorea that is integrated with morphological variation, see Scharaschkin and Doyle (2005, 2006). The position of Unonopsis on the tree has been somewhat erratic, probably because of ancient paralogy (Pirie et al. 2007). For relationships in the ambavioid group, see Surveswaran et al. (2010).

The main group above, "The rest", is made up of two major clades. One is the Malmea-Piptostima-Miliusa (MPM) clade, including a probably polyphyletic Polyalthia; this is the short branch clade of Richardson et al. (2004). It is not very speciose, mostly lacking large genera aside from Polyalthia, which shows relatively little internal molecular divergence (for its limits, see Mols et al. 2008a; Xue et al. 2011). New World members of the short-branch clade form a monophyletic group (Pirie et al. 2006). For Pseuduvaria, see Su and Saunders (2006) and Su et al. (2008, 2010). Mols et al. (2008b) explore character evolution within the miliusoid clade. The other main clade is the inaperturate pollen clade, in which there is more molecular divergence (hence its other name, the long branch clade). This clade includes Artobotrys, Guatteria, Xylopia, Annona, etc. (the latter is in Clade A above: see also J. A. Doyle et al. 2004; Pirie et al. 2005). Details of the groupings need to be worked up. Uvaria is polyphyletic (Mols et al. 2004), but Zhou et al. (2009a, esp. b, 2010) suggested that its limits be extended; geography and phylogeny correlate quite well. For Guatteria, see Erkens (2007) and Erkens et al. (2007a, b). Richardson et al. (2004) provide many other details of the diversification of the short- and long-branch clades.

Classification. The limits of genera and hence the numbers of their included species in the inaperturate pollen/long branch clade seem particularly uncertain. Zhou et al. (2010) clarify the limits of Uvaria, Xue et al. (2011) those of Polyalthia, while Surveswaran et al. (2010) discuss generic limits in the ambavioid clade. For a checklist of the New World species of the genus, see Maas et al. (2011); Annonbase may also be consulted.