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.

[MALVALES + BRASSICALES]: ?

Evolution. Divergence & Distribution. Argout et al. (2011) suggested a date for this clade of a mere 59 million years.

BRASSICALES Bromhead  Main Tree, Synapomorphies.

Idioblastic and stomatal myrosin cells +, glucosinolates from phenyalanine and/or tyrosine [aromatic] and valine/isoleucine/leucine [Branched Chain Amino Acids], idioblastic and stomatal myrosin cells +, little oxalate accumulation; endoplasmic reticulum with dilated cisternae; myricetin, other methylated flavonols, tannins 0; vasicentric axial parenchyma +; tension wood?; mucilage cells in leaf 0; leaves spiral, stipules small; inflorescence racemose; (petals clawed); G [3], ovules in one or two rows; seed coat?; embryo often green. - 17 families, 398 genera, 4450 species.

Evolution. Divergence & Distribution. The earliest fossil known assignable to this clade is from the Turonian, ca 89.5 million years before present (ref?). Wikström et al. (2001: relationships are [Brassicales - Tropaeolum, etc., not included [Malvales + Sapindales]]]) date the origin of stem Brassicales to (89-)85(-81) million years before present, diversification beginning (75-)71(-67) million years before present. The age of crown group Brassicales was estimated as (76-)73(-70) and (63-)69(-57) million years (two penalized likelihood dates), the stem group age being (94-)89(-85) or (80-)74(-68) million years; Bayesian relaxed clock estimates were slightly different, to 46 and 96 million years respectively (Wang et al. 2009), while Magallón and Castillo (2009) estimated ages of ca 91.9 and 92.1 million years for relaxed and constrained penalized likelihood datings for the age of stem Brassicales, and ages of 65.8 and 66 million years (both relaxed and constrained ages again) for the crown group.

Brassicales contain ca 2.2% eudicot diversity (Magallón et al. 1999) and show quite high diversification rates (Magallón & Catillo 2009).

Rodman et al. (1996) suggest a number of apomorphies for the order (and some nodes within it), but where some characters are to be placed on the tree is currently unclear.

Ecology & Physiology. Nearly all the glucosinolate-producing families of flowering plants are in this clade (c.f. Kjær 1974; Dahlgren 1975). Glucosinolates are mustard oil glycosides, and mustard oils themselves are esters of isothiocyanic acid, which contain a R—N=C=S group and have a pungent smell and sharp taste. Specialised, protein-rich myrosin cells contain enzymes like thioglucoside glucohydrolase (a ß-thioglucohydrolase, or myrosinase), that breaks down glucosinolates into glucose and aglucones when plant tissue is damaged by a herbivore and enzyme and substrate are brought into contact. The latter then automatically rearrange and form toxic isothiocyanates, or are converted into thiocyanates (mustard oils), nitriles (not necessarily toxic) and other compounds. This is the "Senfölbombe" or "mustard oil bomb" (see Rask et al. 2000, Wittstock et al. 2003; Grubb & Abel 2006; Halkier & Gershenzon 2006; Agerbirk et al. 2008; Burow et al. 2009 for details, most of which have been worked out in Brassicaceae; Bones & Rossiter 2006 particularly for glucosinolate degradation). Note that not all taxa producing mustard oils have myrosin cells. The ß-glucosidase involved is similar to those that break down cyanogenic glycosides, and there is en evolutionary link between the synthesis of glucosinolates and that of cyanogenic compounds (Halkier & Gershenzon 2006; Morant et al. 2008).

Glucosinolates themselves are synthesised from valine/isoleucine and/or leucine (BCAA - Branched Chain Amino Acids) in several families of Brassicales, i.e. they are aliphatic glucosinolates, but there is much more diversity in families in the terminal polytomy (Rodman 1991a; esp. Mithen et al. 2010 for data). The activity of glucosinolates depends on the structure of the side chain (Hopkins et al. 2009; Sønderby et al. 2010). Note that not all taxa producing mustard oils have myrosin cells, and even in Brassicaceae, which do, such cells may be absent in the young plant (Maile 1980). Myrosin cells may occur in general leaf tissue, and guard cells may also be myrosin cells. In many Brassicales with stomatal myrosin cells, myrosinases occur in large quantities in the guard cells (Jørgensen 1995). They also occur in large amounts (thioglucoside glucohydrolase 1 - TGG1) in the guard cells of Arabidopsis thaliana, at least, even although Brassicaceae lack stomatal myrosin cells; in such cases myrosinases may have become involved in the signaling mechanisms of stomatal opening and closure, or the products of hydrolysed glucosinolates may evalporate through the stomatal pores, deterring herbivores and/or attracting their parasites (Zhao et al. 2008). Attempts are being made to engineer glucosinolates into non-glucosinolate-containing plants such as tobacco (Geu-Flores et al. 2009); this has now succeeded in Nicotiana benthamiana (Sønderby et al. 2010), so anybody for cabbage-tasting tobacco?

Plant-Animal Interactions. Caterpillars of the 780-840 species of Pieridae-Pierinae (foodplants of ca 360 species in 33+ genera recorded, ca 1/3 of all records) are commonly found on members of this order (Fraenkel 1959; Ehrlich & Raven 1964; Braby & Trueman 2006; Braby et al. 2006; Beilstein et al. 2010), including Bretschneidera; however, they are abundant only in the [Capparaceae [Cleomaceae + Brassicaceae]] clade. Pierinae may have moved to Brassicales from an original host in Fabaceae (Braby & Trueman 2006) some (90-)85(-60) million years ago and within ten million years or so of the origin of Brassicales (Wheat et al. 2007), and this shift seems to have been accompanied by a burst in pierine diversification (Fordyce 2010). The ability of pierine caterpillars to live on Brassicales is associated with the evolution of a novel glucosinolate detoxifying mechanism, and these butterflies are more diverse than more basal pierine clades that lack this mechanism (Wheat et al. 2007). After a gene duplication, one of the orthologs produced an enzyme that detoxified glucosinolates by producing nitriles rather than toxic isothiocyanates on their hydrolysis (Fischer et al. 2008); Winde and Wittstock (2011) discuss various other ways in which the herbivore can avoid brassicalean plant defences. Relationships between herbivores and plants - again, nearly all information comes from Brassicaceae - are complex, but specialised feeders, as well as their hymenopteran parasites, may be attracted by isothiocyanates (Hopkins et al. 2009 and references).

Some chrysomelid beetles also favor Brassicales, for example, Phyllotreta (Alticinae - see Jolivet & Hawkeswood 1995), while the dipteran leaf miner Liriomyza brassicae is found on Resedaceae, Cleomaceae, Tropaeolaceae and Brassicaceae (Spencer 1990).

Chemistry, Morphology, etc. For unrelated glucosinolate-containing families, see Putranjivaceae (Malpighiales) and perhaps also - but probably not - Phytolaccaceae (Caryophyllales) and Pittosporaceae (Apiales: Fahey et al. 2001 for a summary). It has also been suggested that Oceanopapaver, a genus of uncertain affinities but now pretty firmly associated with Malvaceae (= Corchorus), has myrosin cells; this, too, is unlikely (Whitlock et al. 2003). A report of ellagic acid in "Capparidaceae" (Bate-Smith 1962) needs to be confirmed. Zindler-Frank (1976) lists seven families scattered throughout the order as having little oxalate accumulation.

Many taxa seem to have diarch roots, although not some Cleomaceae; sampling in the basal pectinations is poor. Note that most families have stipules, albeit small. Strongly developed and fused ventral carpellary bundles may be another synapomorphy for the order (Ronse de Craene & Haston 2006).

For general information, see Villers (1973), Mehta and Moseley (1981), Jørgensen (1981, 1995), Carlquist (1985a), Tobe and Peng (1990), Fisch and Weberling (1990), Rodman (1991a, b), Link (1992a), Rodman et al. (1993, 1994), Tobe and Takahashi (1995), Hufford (1996), Doweld (1996a, b), Ronse Decraene and Smets (1997a) and Kubitzki (2002a, b: as Capparales).

Phylogeny. Relationships within Brassicales show a fair bit of structure (see tree), as Rodman et al. (1997, 1998), Carol et al. (1999), Chandler and Bayer (2000), Kubitzki (2002a), Olson (2002) and Hall et al. (2004) have found. Ronse de Craene and Haston (2006) examine morphological evolution in the clade in the context of a combined molecular (four genes, but some taxa included in the analysis lacking up to three of them) and morphological study; it is unclear what significance to attach to differences in details of the topology of the tree presented there and that used here. Analysis of morphological data alone yielded only one clade (Polanisia + Cleome!) in Brassicales with even weak support, and Bretschneidera and Akania were associated with Sapindaceae (Ronse de Craene & Haston 2006). However, that paper should be consulted for details of floral anatomical/morphological evolution in Brassicales. For the relationships of Emblingiaceae, sometimes associated with Gentianales, see below.

Classification. For a good general account of the families recognised, and the history of the classification of the group, see Fay and Christenhusz (2010).

Previous Relationships. Some Brassicales, Brassicaceae and their immediately related families, have always been placed together based on morphological similarity and chemistry (smell!), but until quite recently the others have been widely separated. Many brassicalean families are included in Violanae (Dilleniidae) by Takhtajan (1997), but Gyrostemonales are in Gyrostemonanae (Caryophyllidae) and Limnanthales in Solananae (Lamiidae). Cronquist (1981) placed Brassicales in Violales (Caricaceae), Capparales (several families), Batales (Gyrostemonaceae, Bataceae), all scattered through Dilleniidae, also in Geraniales (Limnanthaceae) and Sapindales (Akaniaceae), both in Rosidae, etc. Rolf Dahlgren began the process of pulling the order together (e.g. R. Dahlgren 1975a; G. Dahlgren 1989, and references; summary in Jørgensen 1995).

Includes Akaniaceae, Bataceae, Brassicaceae, Capparaceae, Caricaceae, Cleomaceae, Emblingiaceae, Gyrostemonaceae, Koeberliniaceae, Limnanthaceae, Moringaceae, Pentadiplandraceae, Resedaceae, Salvadoraceae, Setchellanthaceae, Stixaceae (to be recognized?), Tovariaceae, Tropaeolaceae.

Synonymy: Akaniales Doweld, Batales Engler, Capparales Berchtold & J. Presl, Caricales L. D. Benson, Gyrostemonales Takhtajan, Limnanthales Martius, Moringales Martius, Resedales Link, Salvadorales Reveal, Tovariales Nakai, Tropaeolales Martius

Akaniaceae + Tropaeolaceae: vessel elements with scalariform perforation plates; axial parenchyma sparse, adjacent to vessels; bracteoles 0; flowers quite large, obliquely monosymmetric; K/C tube +, C clawed; A 8, with short connective prolongations, placentation apical-axile, style long; ovules 1-2/carpel, epitropous; testa vascularized.

Chemistry, Morphology, etc. It is possible that Tropaeolaceae have basically pinnate leaves (Endress 2003c), a matter than can perhaps be cleared up by further developmental studies. This may yield another synapomorphy for the clade. Although both families have a nectary, it is extrastaminal in Tropaeolaceae and intrastaminal in Akaniaceae. Furthermore, exactly which stamens are reduced and details of the plane of asymmetry of the flowers differ between Tropaeolum and Bretschneidera; the former is obliquely asymmetric only in bud (Ronse Decraene et al. 2002a; Ronse Decraene & Smets 2001a). Finally, the hypanthium is described as "lifting sepal lobes and petals high above the stamen insertion" by Ronse Decraene et al. (2002a: p. 44), i.e., it is a calyx/corolla tube in the strict sense; there is also a true hypanthium in Bretschneidera, at least, although it was not evident in the Tropaeolum examined by Ronse Decraene and Smets (2001a). Carlquist and Donald (1996) give additional characters of wood anatomy that may unite these two families.

AKANIACEAE Stapf, nom. cons.   Back to Brassicales

Deciduous or evergreen trees; tannins?; cork subepidermal; young stem with separate bundles; (vessel elements with simple perforation plates); no bordered pits in imperforate tracheary elements; petiole bundle?; cuticle waxes 0, strong cuticular cracks; stomata ?; leaves odd-pinnate, leaflet margins spinulose-toothed or entire, vernation supervolute-curved, petiolules swollen or articulated; (inflorescence branched); K ± connate, C contorted or not; A 8, or 3 (4) abaxial in the whorl opposite petals [Akania]; pollen colpate; nectary + or 0; stigma small, 3-lobed; ovules 2/carpel, (campylotropous - Bretschneidera), micropyle bistomal, outer integument ca 5 cells across, inner integument 2-3 cells across; fruit a septicidal capsule; testa multiplicative, exotestal cells palisade, thick-walled, mesotesta ± thick, cells thick-walled, endotesta thickened; endosperm development?, copious or not, embryo color?, cotyledons large; n = 9 [Bretschneidera].

2[list]/2. S.W. China, adjacent Vietnam, Formosa (Bretschneidera sinensis [Photo - Collection]), E. Australia (Akania bidwillii).

Evolution. Divergence & Distribution. Fossils attributed to Akania are known from Patagonia in Eocene deposits of about 51.9 million years of age (Gandolfo et al. 2011 and references; Wilf et al. 2011).

Chemistry, Morphology, etc. The seeds of Akania have endosperm and the plant may lack myrosin cells, but wood of the two genera is almost identical. In Bretschneidera, glucosinolates are also produced from valine/isoleucine and/or leucine, the leaflets are entire, the flowers are clearly strongly obliquely monosymmetric, the stamens are curved and held under one petal, there is a disc, and the embryo sac is bisporic and 8-nucleate. For ovules, see Mauritzon (1936). I have seen neither fresh vegetative material of Bretschneidera nor flowers of Akania. Seedlings of Akania initially produce at least five simple leaves with pinnate venation.

For general information, see Bayer and Appel (2002).

Classification. Separating Bretschneideraceae from Akaniaceae was considered optional in A.P.G. II (2003), however, there seems nothing lost in combining them (see A.P.G. 2009).

Previous Relationships. A relationship with Sapindales has often been suggested (e.g. Carlquist 1997a), but in this case perhaps largely because Sabiaceae were included in the latter; an analysis of morphological data also suggested this position (Ronse de Craene & Haston 2006). Bretschneideraceae do look rather sapindaceous.

Synonymy: Bretschneideraceae Engler & Gilg, nom. cons.

TROPAEOLACEAE Candolle, nom. cons.   Back to Brassicales

Fleshy vines with twining petioles, (caespitose herbs), often with tuberous roots; glucosinolates also from valine/isoleucine and/or leucine, benzyl- and 4-methoxybenzyl types, only idioblastic myrosin cells, erucic acid [fatty acid] +; cork cambium deep seated? to more superficial; stem with separate bundles; petiole bundles annular; pericyclic fibres 0; cuticle waxes tubular; leaves flat in bud, (palmately compound), lamina palmately lobed, peltate or not, margins toothed to entire, stipules small, in seedling only, to fringed, subfoliaceous and throughout the plant; flowers often axillary, (bracteoles +); ± strongly monosymmetric, hypanthium ?; 3 adaxial K, connate, spurred, (spur ca 1 mm or less), C 2 + 3 (+ 2), clawed, (margin ± deeply lobed/laciniate); A 8; pollen trinucleate; median carpel adaxial, styles short, trifid in appearance, stigma dry; ovules 2/carpel, micropyle endostomal, tenuinucellate; fruit a schizocarp (samara), mericarps drupaceous or nutlike, K deciduous; seed pachychalazal, coat undistinguished, part of mesotesta suberized; amyloid [xyloglucans] in cotyledons, suspensor haustoria penetrate micropyle; n = 12-15.

1/105. New World, esp. Andean (map: see Sparre & Andersson 1991). [Photos - Tropaeolum Flower, Tropaeolum Flower]

• Tropaeolaceae are rather fleshy vines with twining petioles. The broad leaf blades have palmate venation and may be palmately-lobed, palmate or peltate. The flowers are large, strongly monosymmetric, and have an adaxial calycine spur, five clawed petals, and eight stamens. The fruit is usually a three-lobed schizocarp, and each unit contains but a single seed.

Chemistry, Morphology, etc. Carlquist and Donald (1996) report vague storying of the secondary phloem of the root. For an interpretation of the axillary flowers common in Tropaeolum, see Bayer and Appel (2002). The nectary may be hypanthial (Troll 1957) and ends up in the spur; it is outside the stamens (Ronse Decraene et al. 2002). The median carpel is actually slightly off the median (Eichler 1878; Ronse Decraene & Smets 2001a); two antepetalous stamens are suppressed (Ronse Decraene et al. 2002). There are initially two ovules per carpel, but one does not develop very far. The developing seed has a suspensor haustorial system (Walker 1947).

Phylogeny. For phylogenetic relationships in the family, see Andersson and Andersson (2000). Distinctive reproductive morphologies - Magallana, with its winged fruits, and Trophaeastrum, which lacks a nectary spur - are derived from within the general Tropaeolum morphology of spurred flowers and simple schizocarpic fruits.

Classification. There is only a single genus in the family since the recognition of Magallana and Trophaeastrum would make Tropaeolum paraphyletic (Andersson & Andersson 2000). For an account of the southerly and temperate section Chilensia, see Watson and Flores (2010a, b).

[[Moringaceae + Caricaceae] [Limnanthaceae [Setchellanthaceae [[Koeberliniaceae [Bataceae + Salvadoraceae]] [Emblingiaceae [Pentadiplandraceae, [Gyrostemonaceae + Resedaceae], Tovariaceae, [Cleomaceae [Capparaceae + Brassicaceae]]]]]]][: several [³6] ovules/carpel.

Evolution. Divergence & Distribution. The age of this clade is estimated as (90.5-)72-(-47.9) million years (Couvreur et al. 2010).

[Moringaceae + Caricaceae]: woody, stems stout [pachycaulous or cauduciformous]; endoplasmic reticulum-dependent vacuoles; cambium storied; nodes also multilacunar; cuticle wax platelets as rosettes; lamina venation palmate, colleters on petiole/lamina, stipules as glands; inflorescences thyrses; flowers whitish; G opposite sepals, ovary longitudinally sulcate, placental strands opposite the ventral bundles +, placentation parietal, style hollow; ovules many/carpel, micropyle bistomal, outer integument 4-6 cells across; testa multiplicative, mesotesta ± lignified.

Chemistry, Morphology, etc. Ronse de Craene and Haston (2006) suggested that nodes were unilacunar in this clade; they are tri or multilacunar. The sulci in the ovary are in the interplacental position. Whether or not the thickened mesotesta of the two families is comparable needs to be confirmed, certainly there are substantial anatomical differences in the seed coat (Olson 2002a).

Phylogeny. For relationships in this clade, see Olson (2002a).

MORINGACEAE Martynov, nom. cons.   Back to Brassicales

Deciduous trees or shrubs (stem succulents); glucosinolates also from valine/isoleucine and/or leucine; hairs unicellular; schizogenous gum canals +; leaves odd-pinnate, 1- or 3-compound; flowers obliquely monosymmetric; hypanthium short (long), lined with nectary, K petaloid, median [abaxial] C usu. larger than others; stamens = and opposite petals, declinate, monothecal, staminodes +, opposite sepals; gynophore +; G ([2-4]), style slender, stigma truncate-porate; micropyle zig-zag, outer integument vascularized, inner integument ca 3 cells across, parietal tissue ca 3 cells across, endothelium +; fruit a 3-angled explosively-dehiscent effectively loculicidal capsule; seeds 3-angled, winged (not); mesotesta thick, outer and inner parts with helical thickenings, middle part much thickened, tegmen thin, (multiplicative); n = 11, 14.

1[list]/12. India to Africa, Madagascar, Moringa oleifera is quite widely cultivated (map: from Olson 2001). [Photo - Flower, Collection]

• Moringaceae are woody, often quite stout-stemmed shrubs or trees whose foliage often smells unpleasantly when crushed. They are recognisable by their spiral, deciduous, odd-pinnate, 1- to 3-compound leaves that have conspicuous glands at the articulations and at the base; often clearly obliquely monosymmetric and papilionoid flowers (the median petal is adaxial) with five stamens; and long, slender, tricarpellate, three-angled, capsular fruits.

Evolution. Vegetative Variation. Olson and Rosell (2006) suggest that heterochrony is involved in the evolution of the various life forms in the family; the bottle-tree growth form is probably plesiomorphic, the tuberous shrub growth form is probably derived (see also Olson 2006 for wood anatomy). All species have rather fleshy roots/rootstock and are usually to be found in more or less arid habitats.

Chemistry, Morphology, etc. For wood anatomy, see Olson and Carlquist (2001); there are reports of vestured pits from the family (Jansen et al. 2001b). Even although flowers of all species are slightly monosymmetric early in development (Olson 2002b), flowers at anthesis may be polysymmetric or strongly monosymmetric. When flowers are monosymmetric, they are borne with the median petal adaxial, and when they are polysymmetric the median petal is in the normal abaxial position (Olson 2003). Carpel orientation is in the plane of symmetry of the flower (Ronse Decraene et al. 1998a). Corner (1976) suggests that the micropyle is exostomal. Seeds are borne along the middle of the valves which means that dehiscence is effectively loculical given that the placentation is parietal. The seedlings have either palmately compound leaves or simple leaves with palmate venation (M. Olson, pers. comm.).

Some general information is taken from Ernst (1963), who described the ovules as being apotropous. Kubitzki (2002d) and the Moringa website (Olson, 1999) summarize information on the family.

Phylogeny. For relationships, see Olson (2000b).

Synonymy: Hyperantheraceae Link

CARICACEAE Dumortier, nom. cons.   Back to Brassicales

(Viny, but with stout tuber), usu. prickly; benzylglucosinolates +, only stomatal myrosin cells; articulated laticifers +, anastomosing; lamina palmately-veined or strongly lobed (palmate), vernation flat-curved to involute, glands on adaxial surface at base; plants di(mon)oecious; inflorescences axillary, cymose; staminate flower: K connate, small, C connate, contorted or valvate; stamens adnate to corolla, connective often developed, 10, of two lengths, whorled, the longer opposite the sepals, or = and opposite sepals, nectary on pistillode; carpellate flowers: as above, but C often free; A 0; nectary 0; G [5], (placentation axile), styles ± separate, stigmas flabellate or almost petaloid (capitate), dry; inner integument 4-6 cells across; fruit a berry; sarcotesta +, mucilaginous, mesotesta tanniniferous, with lignified ribs, endotesta crystalliferous (lignified), exotegmen fibrous [?sclereidal?]; embryo white; n = 9.

4(-6)[list]/34: Carica 23. Mostly tropical America (three genera in Mexico); Africa (Cylicomorpha only) (map: from Badillo 1971). [Photo - Plant, Flower, Fruit]

• Caricaceae are stout-stemmed often prickly trees, rarely vines, with flowing, latex-like exudate and palmately-compound or -lobed leaves. The inflorescences are axillary, cymose, and the flowers - always imperfect - are moderate in size. The fruit is a rather large berry and contains numerous seeds borne on five parietal placentae and each surrounded by mucilage.

Evolution. Floral Biology & Seed Dispersal. Jacaratia has carpellate flowers with white, spreading stigmas perhaps mimicking the androecium of staminate flowers, but nectar is produced only in the staminate flowers (Bawa 1980). Carica seems to have a similar floral syndrome.

Despite its fleshy fruits, myrmecochory is reported from Carica (Lengyel et al. 2010).

Chemistry, Morphology, etc. Reports of cyclopentenoid cyanogenic glycosides in the family have been questioned (Jørgenson 1995). Some general information is taken from Badillo (1971), Miller (1982) and Kubitzki (2002d); for embryology, see Singh (1970), wood anatomy, see Carlquist (1998c); and for floral development, see Ronse Decraene and Smets (1998b).

Phylogeny. For phylogenetic relationships, see Kyndt et al. (2005). Geographical relationships are interesting, since the African Cylicomorpha is sister to the rest of the family, which are American.

Previous Relationships. Cronquist (1981) includind Caricaceae in his Violales sandwiched between Achariaceae in the restricted sense (more or less herbaceous, South African) and the North American Fouquieraceae; Takhtajan (1997) placed his monofamilial Caricales in a generally similar position, i.e. along with other families that have parietal placentation.

[Setchellanthaceae [Limnanthaceae [[Koeberliniaceae [Bataceae + Salvadoraceae]] [Emblingiaceae [Pentadiplandraceae, [Gyrostemonaceae + Resedaceae], Tovariaceae, [Cleomaceae [Capparaceae + Brassicaceae]]]]]]]: nodes 1:1; extended 3' terminus of rbcL gene.

Chemistry, Morphology, etc. For the extended 3' terminus of rbcL gene, see Karol et al. (1999).

Phylogeny. Setchellanthus comes out just basal to Limnanthaceae in molecular phylogenies (Karol et al. 1999, support weak, see also Rodman et al. 1997; Chandler & Bayer 2000). Its inclusion in this clade is very strongly supported (e.g. Karol et al. 1999).

SETCHELLANTHACEAE Iltis   Back to Brassicales

Shrub; hairs unicellular, T-shaped, on multicellular podium; myrosin cells 0; young stem with vascular cylinder; ?stomata; lamina 2ndary veins subbasal, stipules 0; flowers axillary, large, (5)6(7)-merous; K connate, splitting irregularly, C clawed; A many, centrifugal, in 5-7 groups on elongated axis; pollen tricolpate; nectary 0; gynophore short; placentation axile, style short, stigmas subcapitate; ovules 10-14/carpel, in two ranks; fruit a septifragal capsule, central columella persistent; testa soft, endosperm development?, scanty; n =?

1/1: Setchellanthus caeruleus. Mexico (map: see Iltis 1999). [Photos - Flower, Flower, Fruit]

• Setchellanthaceae are pungent-smelling shrubs that may be recognised by their large, blue, axillary flowers with parts usually in sixes and an irregularly-splitting calyx. Vegetatively the plant is rather undistinguished, although it has T-shaped hairs and the rather small, entire, exstipulate leaves have secondary veins arising from near the base and lack much in the way of a petiole. The three-carpellate fruit is septifragal and there is a persistent columella.

Chemistry, Morphology, etc. The fusion of the marginal ventral carpellary bundles is commissural. Some information is taken from Carlquist and Miller (1999: anatomy), Iltis (1999: general), Tobe et al. (1999: flowers), Tomb (1999: pollen), and Kubitzki (2002d: general).

Previous Relationships. Setchellanthus used to be included in Capparaceae.

[Limnanthaceae [[Koeberliniaceae [Bataceae + Salvadoraceae]] [Emblingiaceae [Pentadiplandraceae, [Gyrostemonaceae + Resedaceae], Tovariaceae, [Cleomaceae [Capparaceae + Brassicaceae]]]]]]: (indole glucosinolates present in appreciable amounts); root hairs in vertical files.

Chemistry, Morphology, etc. Root trichoblasts have been sampled in rather few families. The distinctive vertical files of root hairs are known from Limnanthaceae and some terminal members of Brassicales, and although they have not been observed in Tropaeolaceae other Brassicales basal to Limnanthaceae have not been studied (Dolan & Costa 2001).

LIMNANTHACEAE R. Brown, nom. cons.   Back to Brassicales

Herbs; erucic acid, ellagic acid, myricetin, non-hydrolysable tannins +, isokestose oligosaccharides as storage, only idioblastic myrosin cells; cork?; leaf pinnate, or lamina pinatelly lobed, vernation conduplicate, margins with teeth, stipules 0; (flowers single, axillary, bracteoles 0); flowers 3-5-merous; K valvate, C contorted (open); nectaries on abaxial bases of antesepalous A; A 2x K, of two lengths, largest opposite sepals; pollen with encircling colpus; G [2-5], opposite sepals, when 3 median member abaxial, no vascular bundles in carpel wall, placentation basal-parietal, style gynobasic, hollow, branches ± well developed, stigma punctate to minutely capitate, dry; ovule 1/carpel, apotropous, unitegmic, integument 14-16 cells across, parietal tissue 0; megaspore mother cells several, embryo sac tetrasporic, 4- or 6-nucleate; fruit a schizocarp, mericarps muriculate, K persistent, green; seed coat pachychalazal, thick, with vascular bundles, otherwise undistinguished; endosperm haustorium +, embryo color?, cotyledons with backwardly-directed lobes, with amyloid [xyloglucans]; n = 5.

1(2)[list]/8. Temperate North America (map: from Culham 2007). [Photo - Flower] [Photo - Flower (close-up)]

• Limnanthaceae are rather soft-stemmed herbs with spiral, exstipulate, deeply-lobed or compound leaves and flowers that are rather widely spreading and with a large, valvate calyx, contorted corolla, and gynobasic style or styles. The fruit is a schizocarp separating into shortly spiny, one-seeded portions (mericarps).

Chemistry, Morphology, etc. Details of the development of the embryo sac are unclear. According to van Tieghem (1898), the ovules are epitropous, while Maheshwari and Johri (1956) and Johri (1970) described an endosperm pouch or haustorium on the funicular side of the micropyle region in Floerkea.

Some information is taken from Link (1992a) and that on embryology from Fagerlind (1939b), Mathur (1956) and Maheshwari and Johri (1956), on wood anatomy from Carlquist and Donald (1996); for general information, see Bayer and Appel (2002).

Previous Relationships. The ovules of Limnanthaceae, tenuinucellate, unitegmic, and with tetrasporic embryo sacs, differ from those of other Brassicales, which are usuaully crassinucellate, bitegmic, and with a monosporic, 8-nucleate embryo sac (but see Akaniaceae), but chemistry and molecular data place them here.

Limnanthaceae were often included in Geraniales (e.g. Cronquist 1981), but their androecium is diplostemonous (that of Geraniaceae, at least, is obdiplostemonous), the ovules are apotropous (epitropous), and carpel orientation differs (Eckert 1966). Limnanthaceae were placed in Solananae by Takhtajan (1997).

[[Koeberliniaceae [Bataceae + Salvadoraceae]] [Emblingiaceae [Pentadiplandraceae, [Gyrostemonaceae + Resedaceae], Tovariaceae, [Cleomaceae [Capparaceae + Brassicaceae]]]]]: glucosinolates from chain-elongated branched-chain amino acids; styles short to absent; ovules campylotropous; seeds exotegmic, exotegmen fibrous; embryo strongly curved.

Evolution. Divergence & Distribution. The fossil Dressiantha, from some 90 million years ago in the Cretaceous-Turonian of East North America, may be assignable to a node somewhere around here, Gandolfo et al. (1998b) in a morphological analysis placing it in a clade that included Koeberliniaceae, Bataceae, Brassicaceae, etc., although excluding Gyrosteomonaceae. However, with a floral formula of K4, C5, A5, G [2], decussate calyx, epipetalous stamens, staminodes and ?nectariferous disc internal to the staminal whorl, anthers with prolonged connectives, and obliquely-oriented gynoecium, the relationships of this fossil are unclear to me.

Chemistry, Morphology, etc. Amino acids like isoleucine with branched chains may have additional carbons along the chain; the occurrence of such chain-elongated branched-chain amino acids is to be pegged here on the tree (J. E. Rodman, pers. comm.). There is extensive variation of floral merism in the Emblingiaceae-Brassicaceae group in particular, but 4-merous flowers could be an apomorphy at this level - but with plenty of reversals. A fibrous, if unlignified, exotegmen may well be another apomorphy; it has, for example, been reported from Koeberliniaceae, Salvadoraceae, Resedaceae, Cleomaceae, etc. As Tobe and Raven (2008) suggest, optimisation of this and other embryological features on the tree is unclear; if ovule and seed characters are placed at this node, they reverse in [Bataceae + Salvadoraceae].

[Koeberliniaceae [Bataceae + Salvadoraceae]]: idioblastic myrosin cells 0; flowers 4-merous; pollen 3-colporoidate; G [2]; fruit indehiscent; exotestal cells well developed; n = 11.

Chemistry, Morphology, etc. For variation in and possible synapomorophies of this group, see Ronse Decraene and Wanntorp (2009).

KOEBERLINIACEAE Engler, nom. cons.   Back to Brassicales

Woody, thorny; ellagic acid?, tannins?; no glucosinolates; cork pericyclic; perforation plates bordered; pits vestured; intercellular canals +; leaves minute, fugacious, stipules 0; inflorescences axillary, (flowers 5-merous); A (10); tapetal cells multinucleate; nectaries at the base of A; G with gynophore, oblique, placentation axile, style +, stigma ± minutely expanded; ovules ca 10/carpel, tenuinucellate, apotropous and epitropous, micropyle zig-zag, outer integument 2 cells across, non-multiplicative, nucellar epidermal cells radially enlarged; fruit a berry; exotesta with massive cuticle, then tanniniferous cells, exotegmen walls very thick, lignified, cells moderately elongated [fibrous]; embryo green, endosperm type?, moderate, cotyledons incumbent.

1/2. C. and S.W. North America, Bolivia (map: from Holmes et al. 2009). [Photo - Habit] [Photo - Flower]

• Koeberliniaceae are thorny, much-branched shrubs of dry regions; the prophylls are basal on the branches/thorns. The flowers are small and undistinguished, and the fruit is a few(1-2)-seeded berry terminated by the persistent style.

Chemistry, Morphology, etc. The ovules look as if they may be campylotropous (see also Tobe & Raven 2008). Nodal anatomy is taken from that of the bracts (Mehta & Moseley 1981). For anatomy, see Gibson (1979), for floral anatomy, see Mehta and Moseley (1981), for embryology, see Tobe and Raven (2008), and for general information, see Kubitzki (2002d); von Schrenk, Aug. 8, Texas - seed anatomy.

Classification. See Holmes et al. (2009) for a monograph.

Previous Relationships. Canotia, sometimes placed here (e.g. Hutchinson 1973), is included in Celastraceae. Both are thorny shrubs, but that is the main extent of their similarity. Koeberlinia itself has been included in Capparaceae (Cronquist 1981).

[Bataceae + Salvadoraceae]: wood ± storied; perforation plates not bordered; rays wide, multiseriate; nodes 1:2; stomata paracytic; leaves opposite, with 2ndary veins ascending from at or near base; bracts with colleters on their tips; G (2); ovules 2/carpel, basal; exotegmen not fibrous; endosperm 0, embryo ± straight, color?

Chemistry, Morphology, etc. The two families are very similar morphologically (Rodman et al. 1996) and anatomically (Carlquist 2002a), while Ronse de Craene and Haston (2006, see also Ronse de Craene & Wanntorp 2009) list a number of other features the two families share including colleter-like stipules, flowers that are slightly monosymmetric and horizontally oriented relative to the inflorescence axis, a sepal tube, endostomal micropyle, etc. The flowers of neither family are easy to interpret.

BATACEAE Perleb, nom. cons.   Back to Brassicales

Fleshy shrublets; (hydroxy)proline betaines +, tannins?; cork pericyclic; perforation plate borders vestigial; pits vestured; leaves fleshy, stipules unvascularized, intrapetiolar or cauline; plant monoecious or dioecious, inflorescences usually axillary, densely spicate; flowers small, bracteoles 0, nectary 0; staminate flowers: K 2, median, enveloping flower, or K 4, connate; P clawed; A = and alternate with P; pollen surface psilate, ektexine spongy, undifferentiated; carpellate flowers: P 0; G 4-locular [carpels subdivided], stigmas sessile, capitate-penicillate; ovules collateral, epitropous, micropyle ± zig-zag, nucellar cap +; fruit multiple, or a drupe with four pyrenes; seed coat membranous.

1[list]/2. N. Australia and S. New Guinea, tropical America, and the Galapagos (introduced into the Hawaiian Is.) (map: from van Steenis & van Balgooy 1966; Heywood 1978; George 1982). [Photo - Flowers]

• Bataceae are halophytic subshrubs with opposite, small, entire, fleshy leaves that have minute "stipules". The flowers are staminate or carpellate and aggregated into dense, spicate, more or less strobilate, inflorescences. The carpellate flowers lack bracts or perianth and are fused together to form a compound fruit.

Chemistry, Morphology, etc. For the nodal anatomy of Batis maritima, with what is presumably the foliar trace disappearing as it approaches the node, the leaf being supplied by two bundles from the angles of the stem, see van Tieghem (1893), but cf. Johnson (1935) and R. A. Howard (pers. comm.). The stipules need study: van Tieghem reports them to be absent, Johnson (1935) that they are between the broad leaf base and the stem, Rogers (1982b) that they are cauline, while Ronse De Craene (2005) in a floral study describes the fairly massive structures in this position in the flowers as being colleters (note that these descriptions are not all mutually exclusive). The morphological nature of the structure enveloping the staminate flowers is most obvious in B. argillicola, but there is controversy over the nature of this structure, too. Ronse De Craene (2005) described it as being derived from four sepals in Batis maritima, although he noted that it had only a single vascular trace; some of the lobing of the tubular structure may in fact be caused by pressure from other parts of the developing flower rather than reflecting an inherently four-merous tube.

Batygina et al. (1985) provide information on the ovules, for pollen, see Tobe and Takahashi (1995), for floral development of Batis maritima, see Ronse De Craene (2005), and for general information, see Bayer and Appel (2002).

SALVADORACEAE Lindley, nom. cons.   Back to Brassicales

Woody; tannins 0; cisternae of endoplasmic reticulum dilated, but no myrosin cells at all; cork superficial; wood storied; (vestured pits + - Salvadora); included phloem + (0 - Azima); cuticle waxes with platelets; lamina vernation flat-curved [Salvadora]; plant dioecious or polygamous, or flowers bisexual; (bracteoles 0); floral orientation oblique; K 2-4(-5), (one-trace), contorted, basally connate, C (5), contorted or imbricate, (connate); (C/A tube +, short [Salvadora]); A = and opposite sepals, free, basally connate, or adnate to C; pollen surface reticulate; nectar glands alternating with or abaxial to A or 0; G (with gynophore), 1-2-4-locular [?false septae], (style short), stigma at most slightly lobed; ovules 1-2/carpel, apotropous, micropyle exo- or endostomal, outer integument 10-15 cells across, (with a vascular bundle - Azima), inner integument 3-5 cells across, (obturator +); fruit a berry or drupe; exotestal cells palisade, slightly thickened, inner walls mucilaginous, crystalliferous, tegmen becoming crushed, exotegmic cells fibrous, not lignified; cotyledons thick; n also = 12.

3[list]/11: Salvadora (5). Africa (inc. Madagascar) to South East Asia and West Malesia, often in drier regions (map: from Aubréville 1974 - Malesian distribution rather optimistic, perhaps only in Java). [Photo - Habit, Fruits]

• Salvadoraceae are rather small-leaved shrubs, sometimes scrambling, or small treees with opposite, rather leathery leaves; small, 4-merous flowers with connate sepals and two carpels with a short style; and a fleshy fruit.

Chemistry, Morphology, etc. The stipules of Salvadora persica are described as being colleter-tipped. Azima has two trace-one gap nodes to the bracts and bracteoles, Salvadora has one trace, one gap nodes to the bracts (Kshetrapal 1970). R. A. Howard (pers. comm.) reported two trace, one gap nodes from both genera.

The flowers may be slightly monosymmetric and there may be a poorly developed petal-stamen cup (Ronse de Craene & Wanntorp 2009). Ronse de Craene and Wanntorp (2009) descibe the gynoecium as probably originally being bicarpellate and with parietal placentation; if the ovary appears to have two loculi it is because of the development of a structure perhaps comparable to the false septum of Batis.

For floral vascularization, see Kshetrapal (1970), for pollen, see Lobreau-Callen (1977) and Perveen and Qaiser (1996), but not much is known, for wood anatomy, see Carlquist (2002a) and Saxena and Gupta (2011), and for general information, see Kubitzki (2002d).

Synonymy: Azimaceae Wight & Gardner

[Emblingiaceae [Pentadiplandraceae, [Gyrostemonaceae + Resedaceae], Tovariaceae, [Cleomaceae [Capparaceae + Brassicaceae]]]]: Glucosinolates also from tryptophane [= indole glucosinolates] +; cisternae of endoplasmic reticulum dilated and vacuole-like; cuticle wax crystalloids 0; inflorescence terminal, bracteoles 0; floral development open, C clawed; disc/nectary outside A; ovules in two rows; endotesta crystalliferous; 3' rbcL extension.

Evolution. Divergence & Distribution. Ronse de Craene and Haston (2006) suggest a number of other characters that may be synapomorphies for this clade, including floral symmetry and embryo development. Note that nectary morphology and absence/presence/position is very variable in Brassicales outside this core group.

Genes & Genomes. It is possible that the At-ß duplication is to be placed at this node; it is absent from Carica (Barker et al. 2009: but sampling).

Chemistry, Morphology, etc. Whether or not Emblingiaceae have indole glucosinolates is unknown; only small quantities have been reported from Pentadiplandra (references in de Nicola et al. 2012).

In Pentadiplandraceae, Brassicaceae and Tovariaceae the lateral sepals are initiated before the median sepal (Ronse Decraene 2002). For ovule type and the different mechanisms by which the ovule becomes campylotropous, see Boesewinkel (1990) and Bouman and Boesewinkel (1991), for bracteoles, see Ronse Decraene (1992).

Phylogeny. Phylogenetic relationships in this core group of Brassicales have been partly resolved in a three-gene study by Hall et al. (2004). Ronse de Craene and Haston (2006) found that Emblingiaceae moved outside the group in a combined morphological-molecular study, but many data were missing for Emblingia in particular and its floral morphology is odd and poorly understood (see below). Ronse de Craene and Haston (2006) suggested that it might be sister to [Bataceae + Salvadoraceae], but noted that there was little support for this position.

EMBLINGIACEAE Airy Shaw   Back to Brassicales

Subshrub; plant hispid; mustard oils?; cork cambium deep-seated; cambium storying?; sclereids +; leaves ± opposite; flowers axillary, monosymmetric, resupinate, K connate, lobed, deeply divided adaxially, C 2, ?not clawed, abaxial, connate by epidermis, slipper-shaped, nectary abaxial, androgynophore curved abaxially; A 8, median members absent, 4 abaxial fertile, 4 adaxial staminodial, forming a torus, pollen with short colpi with rounded ends and bulging apertures, the adjacent exine being thickened; G [2-3], placentation axile, stigma shortly lobed; ovule 1/carpel, basal; fruit indehiscent, pericarp thin; seed arillate, testa thick; endosperm ?type, scanty, embryo color?, hypocotyl short; n = ?

1[list]/1: Emblingia calceoliflora. W. Australia (map: from FloraBase 2004).

• Emblingiaceae are low-growing herbs with axillary, ebracteolate and strongly monosymmetric flowers. The flowers are held upside down, the large, connate calyx is slit adaxially, there are two, abaxial, slipper-shaped petals (hence the specific epithet), and there is a curved androgynophore with an abaxially-positioned nectary at the base.

Chemistry, Morphology, etc. The plant dries yellowish. There has been considerable disagreement over the floral structure: is the flower resupinate or not? Is the hood petal-like or not? Are there one or three carpels? I largely follow Melville's interpretation (in Erdtman et al. 1969), see also Mueller (1860). Detailed studies of all aspects of this plant are needed, nevertheless, its embryo is curved, its flowers monosymmetrical, and its nectary is between the petals and stamens, all features appropriate for a position around here.

For general information, see Kubitzki (2002d) and Ronse de Craene and Wanntorp (2009: stipules present, reduced).

Previous Relationships. Emblingia was included in Polygalaceae by Cronquist (1981) and Polygalales by Takhtajan (1997) based on what seemed to be floral similarities. Savolainen et al. (2000b) placed Emblingiaceae in Gentianales, a position that is not currently supported.

[Pentadiplandraceae, [Gyrostemonaceae + Resedaceae], Tovariaceae, [Cleomaceae [Capparaceae + Brassicaceae]]]: ?

Evolution. Animal-Plant Interactions. The larvae of Chrysomelidae-Alticinae beetles are quite commonly to be found on members of this clade (Jolivet 1988).

Classification. This is the core Brassicales of Ronse de Craene and Haston (2006).

PENTADIPLANDRACEAE Hutchinson & Dalziel   Back to Brassicales

Shrubs or lianes; benzyl- and 4-methoxybenzyl glucosinolates +, ellagic acid?, tannins?; cork?; vessel elements with ? perforations; wood ?storied; nodes 3:3; mucilage cells +; stipules +, minute; inflorescence axillary, subcorymbose; flowers polygamous; K 5, valvate, C 5, connivent at enlarged, concave base, limb flat; short andogynophore/disc; staminate flowers: A 9-13, connective shortly produced; pistillode +; carpellate flowers: staminodes +; gynophore short, G [3-5];, opposite sepals, placentation axile, style long, stigma shortly lobed; ovules ca 10/carpel, in two ranks, type?; fruit a berry; 1 seed/loculus, coat with layer of white, wooly, elongated cells towards outside ["seed pubescent"]; embryo white; n = ?

1/?1: Pentadiplandra brazzeana. Tropical W. Africa (map: from Hall et al. 2004).

Chemistry, Morphology, etc. The fruit contains the sweet-tasting protein, brazzein. The plant does not dry dark.

Are there supernumerary buds? There are bordered pits in wood fibres and mucilage cells in the leaf epidermis (Boodle, K, ms.). Ronse Decraene (2002) suggests that the stipules are large, but they are certainly not particularly big on the vegtative part of the plant. Embryologically - and in many other respects - Pentadiplandra is poorly known, although Ronse Decraene (2002) has described its floral anatomy; Ronse de Craene and Haston (2006) suggest that its floral morphology is close to the ancestral form of the core Brassicales. There are no marginal or placental strands in the ovary.

For general information, see Bayer and Appel (2002), for glucosinolates, see de Nicola et al. (2012), and for embryo colour, Martin Cheek (pers. comm.)

Previous Relationships. Cronquist (1981) included Pentadiplandra in his rather broadly circumscribed Capparaceae, Takhtajan (1997) segregated it as a family.

[[Gyrostemonaceae + Resedaceae], Tovariaceae, [Cleomaceae [Capparaceae + Brassicaceae]]]: (glucosinolates chain-elongated BCAAs)

Evolution. Divergence & Distribution. The stem group age for this group is estimated to be 47-45 million years before present, the crown group age 42-33 million years before present (Wikström et al. 2001)

Chemistry, Morphology, etc. For details on glucosinolate variation - quite extensive - within this clade, see Mithen et al. (2010). Nucellar tracheids have been reported in Capparaceae and Resedaceae, at least (Werker 1997). The wood anatomy of Brassicaceae and Resedaceae is rather similar (Schweingruber 2006). For stipules, see Weberling (2006).

[Gyrostemonaceae + Resedaceae]: hairs unicellular; styluli +; calyx persistent; seeds arillate.

Evolution. Floral Biology & Seed Dispersal. Possession of myrmecochorous seeds may be an apomorphy at this level.

Phylogeny. The composition of this clade and relationships within it are currently uncertain. Of the sampled Stixeae (ex Capparaceae) that come out around here, the Asian Tirania may be close to Gyrostemonaceae and the New World Forchhammeria may be closer to Resedaceae (Hall & Sytsma 2000, 2002; Hall et al. 2002), or both may be associated with Resedaceae (Hall et al. 2004: details of the relationship depend on the gene sequenced). Stixis and Neothorelia are the other genera involved. Their flowers tend to be 3-, 5- or 6-merous, there is no differentiation between the two perianth whorls, they have large stigmas and the style is absent or branched, and the placentation is axile (see also Kers 2002). Tirania has six sepals and petals and axile placentation. Forchhammeria has two carpels, as well as an irregular number of sepals (see Gyrostemonaceae), no petals, parietal placentation and one ovule/carpel; only one one ovule/fruit usually develops and it also has methyl glucosinolate (Mithen et al. 2010). Stixeae (= Stixaceae Doweld) are Indo-Malesian (red in map, from Jacobs 1960), and some are climbers with successive cambia (Carlquist 1988), while Forchhammeria is Central American (green in map, partly from Tropicos xii.2010); all are certainly out of place in Capparaceae (Kers 2002), but whether they will need to be placed in a separate family awaits further work.

GYROSTEMONACEAE A. Jussieu, nom. cons.   Back to Brassicales

Trees to shrubs; myrosin cells 0, tannins?; cork subepidermal; wood storied; petiole bundle arcuate; leaf vernation flat, (stipules 0); plants usu. dioecious, inflorescence various; flowers small; P uniseriate, connate, 4-8-lobed or not; axis flattened, disc-like; staminate flowers: A 6-many, in 1 or more whorls around axis, centripetal, filaments ± 0; pollen tricolpate, ektexine spongy, undifferentiated; carpellate flowers: G (1 [2-)many], borne around axis in 1 (2) whorls, connate or not, when G 2, transverse, placentation axile-apical, (styluli marginal), stigmas decurrent, large and spreading or not; ovule 1/carpel, apotropous; fruit a dry or succulent schizocarp (achene; syncarp); endosperm copious, embryo color?; n = 14.

5[list]/18+: Gyrostemon (12). Australia, not in the north (map: see George 1982).

• Gyrostemonaceae are yellowish-drying (including the twigs) shrubs with rather small, often quite narrow but otherwise featureless leaves. The rather small flowers with a persistent, connate calyx (presumably; there is only a single perianth whorl) and many stamens and/or carpels clearly arranged in a whorl are quite distinctive.

Evolution. Divergence & Distribution. Fossils of Gyrostemonaceae have been reported from New Zealand (Lee et al. 2001).

Floral Biology & Seed Dispersal. Gyrostemonaceae are wind-pollinated; the seeds are myrmecochorous (Lengyel et al. 2010).

Chemistry, Morphology, etc. For gynoecial orientation, see Friedrich (1956), for pollen, see Tobe and Takahashi(1995), for additional information, see Goldblatt et al. (1976: general), Hufford (1996: floral development) and George (2002d: general).

RESEDACEAE Martinov, nom. cons.   Back to Brassicales

Herbs (shrubby); idioblastic myrosin cells 0, BCAA glucosinolates, tannins 0; cork?; no bordered pits in imperforate tracheary elements; lamina margins entire to pinnatifid, (stipules 0); flowers vertically monosymmetric, hypanthium short or 0, K ± valvate, (4-)6(-8), C valvate, (0, 2, 4-)6(-8), unequal, the adaxial largest, ligulate at junction of claw and limb, limb usu. ± fringed; disc esp. pronounced adaxially, (bipartite), almost petaloid; A 3-many from ring primordium and centrifugal, basally connate or not; (short gynophore +), G [(2) 3-6(-8)] (± free), opposite sepals or when 3, median member often adaxial, often open apical-adaxially, placentation parietal (axile); ovules (1-)several/carpel, (tenuinucellate), (bistomal), outer integument ca 2 cells across, inner integument 3-4 cells across, hypostase +; fruit with apical opening persisting between the styles (follicle; berry); (aril 0), endotestal cells cuboid, ± thickened, unlignified, crytsalliferous, exotegmic cells fibrous, lignified, (thickening U-shaped; overgrown), endotegmen crystalliferous; n = 5-15.

3[list]/75: Reseda (68). Warm temperate and dry subtropical, esp. Mediterranean-Middle East-North African, also southern Africa, S.W. North America and S.W. China (all Oligomeris) (map: see Meusel et al. 1965; Hultén & Fries 1986; Martín-Bravo et al. 2009). [Photo - Flower]

• Resedaceae are yellowish-drying sometimes somewhat shrubby herbs with terminal, racemose inflorescences and rather small, monosymmetric flowers. The leaves are quite variable, although there are often small stipules. The petals are free and clawed; the adaxial petals are usually largest and the limb is often fringed. The extrastaminal disc/nectary is often conspicuous. The dry fruits often have an apical opening persisting between the styles (a quasi follicle) and the seeds are often arillate.

Evolution. Divergence & Distribution. Martín-Bravo et al. (2007) discuss the phylogeny and biogeography of the family. Oligomeris, native to both the Old and New Worlds, shows a considerable disjunction (Martín-Bravo et al. 2009).

Floral Biology & Seed Dispersal.The seeds of Reseda are myrmecochorous (Lengyel et al. 2010).

Chemistry, Morphology, etc. Ochradenus has C 0, A many; G 3, the carpels are ultimately closed, and the fruit is berry-like - cf. Gyrostemonaceae (Hufford 1996); however, this genus seems to be derived from within Reseda (Martín-Bravo et al. 2007). The androecium of Reseda luteola may be in 3-4 whorls (for references, see Abdallah 1978), cf. again also Gyrostemonaceae. The appendages on the fruit are also described as being caruncles.

For some seed/ovule anatomy, see Guignard (1893) and Singh and Gupta (1968), for stipules, see Weberling (1968), for general information, see Abdallah (1967), Abdallah and de Wit (1979) and Kubitzki (2002d), for floral development of Reseda lutea, see Leins and Sobick (1977), and for wood anatomy, see Carlquist (1998a) and Schweingruber (2006).

Phylogeny. From the topology of the tree presented by Martín-Bravo et al. (2007), it seems that just three genera could be recognised - [Caylusea [Sesamoides + Reseda]].

Synonymy: Astrocarpaceae A. Kerner

TOVARIACEAE Pax, nom. cons.   Back to Brassicales

Herbs to shrubs; glucosinolates not from phenylanaline or tyrosine, tannins?; cork?; no bordered pits in imperforate tracheary elements; leaves trifoliolate, stipules cauline or on leaf base; flowers (6-)8(-9)-merous; K free; stamens = and opposite sepals; gynophore short, G [(5) 6(-8)], alternating with K, placentation ± axile, stigmas spreading; ovules many/carpel, in several ranks, micropyle zig-zag, outer integument ca 2 cells across, inner integument ca 3 cells across, funicle long; fruit a berry; exotestal cells ± enlarged, tanniniferous, walls thickened, endotestal cells small, exotegmic cells fibrous, walls reticulately thickened; endosperm thin, embryo color?; n = 14.

1[list]/2. Tropical America (map: see Hall et al. 2004). [Photo - Flower, Fruit]

• Tovariaceae are herbs that can be recognised by their trifoliate, stipulate leaves, terminal, racemose inflorescence, and 6-9-merous, polypetalous flowers with a short style and spreading stigmas.

Chemistry, Morphology, etc. The ovules are ± anatropous, but become campylotropous by the post-fertilisation development of the exotegmen (Boesewinkel 1990). For general information, see Appel and Bayer (2002).

[Capparaceae [Cleomaceae + Brassicaceae]]: sinapine, methyl glucosinolates, erucic acid [fatty acid] +, glucosinolates also from methionine [aliphatic glucosinolates]; stomatal myrosin cells 0, cisternae of endoplasmic reticulum organelle-like, etc.; cork also cortical; lateral wall pits of vessels vestured; nodes also 3<:3<; eglandular hairs simple, unicellular [?level]; leaves simple to palmately compound, blades usu. conduplicate, margins pinnately lobed to entire; flowers 4-merous, (vertically monosymmetric); K (2-)4(-6); A from 4 primordia, centrifugal, longer than the petals, filaments articulated; gynophore +, carpels when 2, collateral, placentation parietal, placental strands well developed, stigma lobed, subcapitate or not; ovules many/carpel, (tenuinucellate), micropyle zig-zag (endostomal); K deciduous; exotesta palisade or not, endotesta with inner walls ± thickened, endotegmen lignified (or not); cotyledons accumbent or incumbent.   Back to Brassicales

Evolution. Divergence & Distribution. The divergence between Capparaceae and the [Cleomaceae + Brassicaceae] clade has been dated to ca 23 million years before present (Wikström et al. 2001), but this is in conflict with other estimates for ages in this clade, e.g. divergence between Capparaceae and Brassicaceae at ca 41 million years before present (Schranz & Mitchell-Olds 2006), while Beilstein et al. (2010) suggest a still older age of (83.2-)71.3(-59.7) million years for this node.

Plant-Animal Interactions. Pierid caterpillars (Pieridae-Pierinae - the whites - there are ca 840 species) are notably common on members of this group (see also Beilstein et al. 2010, also discussion above). For details of the interactions of butterflies and plants, see Courtney (1986) and Chew (1988 and references); species of Appias subgenus Catophaga (albatrosses) are found more or less indiscriminately on Drypetes (Putranjivaceae) and Capparaceae (Yata et al. 2010), and of course both groups contain glucosinolates.

Ca 1,000 species of these plants are susceptible to pseudoflower-forming rust fungi (Puccinia spp.) (Roy 1993, 2001); see especially Brassicaceae. The oomycete Albugo, the white blister rust, parasitizes members of this clade - and also Convolvulaceae - Albugo candida occuring on members of all three families (Choi et al. 2009; Thines & Voglmayr 2009).

Genes & Genomes. The rate of molecular evolution of two herbaceous groups of this clade studied is notably higher than that of woody rosids (Barker et al. 2009).

Chemistry, Morphology, etc. Rodman et al. (1996) list 11 possible apomorphies for this node; Iltis et al. (2011) also suggest apomorphies in the group. The distribution of root hairs and of methyl glucosinolates, and variation patterns in seed coat anatomy are a little odd.

Both Brassicaceae and Resedaceae have glucosinolates derived from elongated amino acid chains, as do Tovariaceae, Gyrostemonaceae, and Resedaceae (Kjær 1973; esp. Fahey et al. 2001; Mithen et al. 2010). Only Wasabia japonica in Brassicaceae has a glucosinolate similar to those in Cleomaceae and Capparaceae, while one aromatic glucosinolate of Cleomaceae and Capparaceae is also found in Resedaceae. Forchhammeria has methyl glucosinolate like Capparaceae and Cleomaceae (Mithen et al. 2010). Quaternary ammonium compounds, including betaines, are common in both Capparaceae and Cleomaceae, and also Boscia (see the [Resedaceae + Gyrostemonaceae] clade), but have not been detected in Pentadiplandra or Emblingia - or Buhsia (McLean et al. 1996).

True blue or red flowers are very rare in the whole group. Campylotropy is by the inpushing of the chalazal bundle. The ventral carpellary bundles are fused and weakly developed (Ronse de Craene & Haston 2006). Rodman et al. (1996) list 11 possible apomorphies for this node. Guignard (1893) provides details of ovule and especially seed anatomy.

Phylogeny. Capparaceae are sister to Cleomaceae + Brassicaceae; for further details of relationships see Hall and Sytsma (2000) and Hall et al. (2002). Vaughan and Whitehouse (1971) suggest that Brassicaceae differ from Capparaceae (inc. Cleomaceae) in that the latter have a testa that is only two (not three) cell layers thick, a persistent tegmen (rare) and a thicker endosperm (1 cell layer thick). Judd et al. (1994) provide a morphological phylogeny for Brassicaceae and Capparaceae sensu latissimo.

Classification. Although Cruciferae/Brassicaceae s. str., cabbage and mustard, have always been considered as one of the most natural plant families, their recognition makes Capparaceae s.l. (= Capparaceae s. str. + Cleomaceae) paraphyletic. So the alternatives are to have one family (Brassicaceae s.l.); three families; or two families, a Brassicaceae including Cleomoideae and a Capparaceae. The second option is followed here (see also A.P.G. 2009; Iltis et al. 2011).

CAPPARACEAE Jussieu, nom. cons.   Back to Brassicales

Trees and shrubs (herbs, climbers); root hairs 0; pyrrolidine alkaloids +; petiole bundle annular or arcuate; sclereids +; stipules usu. minute; (inflorescence fasciculate); flowers often monosymmetric; (K + C tube +), (K connate), C (0); (nectary elongated - Cadaba); A (1-)4-8(-many and centrifugal), pollen surface variously sculpted; G [2-12], (when 2, superposed-oblique), (placentation axile; secondary septae +; style +); outer integument ca 2 cells across, inner integument 3-4 cells across; fruit a berry (transversely schizocarpic; septicidal); seeds 5-30 mm long, (with a little to much invaginated coat); (endotesta crystaliferous), tegmen multiplicative, to 6 layers thick, exotegmen radially enlarged, sclerified, endotegmen with lignified bands on anticlinal walls, or most cells lignified; radicle-hypocotyl short, cotyledon texture and folding variable; n = (7-)10(-15+); 6 bp insertion in ndhF gene.

16[list]/480: Capparis (250), Maerua (100), Boscia (37), Cadaba (30). Largely tropical (map: from Jacobs 1960; Wickens 1976; George 1982; Jalas & Suominen 1991; Culham 2007 [New World]). [Photo - Flower, Fruit.]

• Capparaceae are woody, the inflorescence is racemose, the flowers have a gynophore and often many stamens with long filaments, and the fruit is indehiscent.

Evolution. Ecology & Physiology. Capparaceae are notably prominent in seasonally dry tropical forest (Pennington et al. 2009).

Chemistry, Morphology, etc. Crateva has glands (?colleters) at the base of the lamina. Some Capparaceae have supernumerary buds and dry yellowish.

For embryology, see Narayana (1962b), for some information on floral development, see Leins and Metzenauer (1979) and Ronse Decraene and Smets (1997a, b), and for general information, see Kers (2002).

Phylogeny. A paraphyletic Crateva is strongly supported as being sister to the rest of the subfamily; Capparis is probably diphyletic (Hall et al. 2002; Hall 2008). Hall (2008) discusses generic limits in Capparaceae, which are in need of substantial work; for example, New World Capparis will need a new name. New World Capparaceae have several distinctive and perhaps unique glucosinolates (Mithen et al. 2010).

Previous Relationships. Forchhammeria and perhaps other Stixeae that used to be placed here are properly to be placed in or near Resedaceae (Hall et al. 2002, 2004)). Setchellanthus (see Setchellanthaceae) and Koeberlinia (Koeberliniaceae, both still Brassicales) also used to be included in Capparaceae.

[[Cleomaceae + Brassicaceae]: herbaceous annuals (shrubs); inflorescence ± corymbose, (bracts foliaceous); C 4; A 6; G [2]; fruit septicidal, persistent placental strands + (0); seeds 0.5-4 mm long.

Evolution. Beilstein et al. (2010) suggest that this node is (76.5-)64.5(-54.4) million years old.Chemistry, Morphology, etc. Some Cleomaceae and Brassicaceae have similar acylated anthocyanins (Jordheim et al. 2009). For foliaceous bracts, see Eichler (1878) and Prenner et al. (2009). Since Aethionema, with a more or less sessile gynoecium, is sister to all other Brassicaceae, similarities of Stanleya, etc., to Cleomaceae (e.g. herbaceous habit; long gynophore) are presumably parallelisms (Galloway et al. 1998); Stanleya, etc., form a clade well embedded in Brassicaceae. For floral development, see Leins (2000, and references).

CLEOMACEAE Berchtold & J. Presl   Back to Brassicales

Root hairs 0; petiole bundle(s) arcuate; leaves often palmate, (stipules +); bracts foliaceous (not); plant monoecious (Podandrogyne); flowers monosymmetric, (androgynophore +; gynophore 0); (C toothed); anthers linear, coiled at dehiscence; pollen surface variously sculpted, often spinulose; (G [4], orthogonal); outer integument 2-3 cells across, inner integument 2-10 cells across, parietal tissue 3-5 cells across, nucellar cap ca 2 cells across, (endothelium +); (fruit indehiscent); seeds (arillate), coat invaginated between radicle and and cotyledons; exotegmen cells radially enlarged, sclerified, endotegmen cells with lignified bands on periclinal walls; (suspensor massive, haustorial), radicle-hypocotyl long, cotyledons incumbent; n = ³9.

10[list]/300: Cleome (275: including Podandrogyne). Tropical and warm temperate, esp. America (map: see Wickens 1976; George 1982; Jalas & Suominen 1991; Culham 2007). [Photo - Inflorescence, [Flower.]

• Cleomaceae are herbaceous or shrubby plants with palmately compound leaves, the inflorescence is racemose, the flowers have four clawed petals, six stamens, and two carpels; the gynoecium has a gynophore, the stamens have long filaments, and the dehiscent fruit has a htin, persistent, loop-like woody placenta that remains on the plant after the fruit valves have fallen off.

Evolution. Divergence & Distribution. The age of stem-group Cleomaceae is uncertain, given the comparable uncertainty of that of Brassicaceae, their sister group - ages range from 22-18 million years (Wikström et al. 2001), some 41 million years (Schranz & Mitchell-Olds 2006), or some 50 million years (Al-Shehbaz et al. 2006), and there are other estimates more extreme in both directions. A duplication (hexaploid) of the genome in the family occurred ca 20 million years before present (it was detected in Cleome spinosa, Csa: Schranz & Mitchell-Olds 2006; Barker et al. 2009). Soltis et al. (2009: duplication unlocalized) suggest that diversification in Cleomaceae may be connected to this genome duplication.

Ecology & Physiology. There are a few species of Cleome with C₄ photosynthesis, e.g. C. gynandra, and photosynthesis with carbon-concentrating mechanisms may have evolved four times in the genus, at least three of which are C₄ types (Feodorova et al. 2010; see also Voznesenskaya et el. 2007 for details of the photosynthetic characterizations; Koteyeva et al. 2011a for the C₄ morphologies involved; Christin et al. 2011b for some dates).

Chemistry, Morphology, etc. Note that in Cleomaceae the inflorescence may be a corymb, there are usually 6 stamens, etc., just like Brassicaceae - e.g. as in Podandrogyne (= Cleome) - also with orange flowers, 3-foliolate leaves and a gynophore - however, the stamens are rarely tetradynamous (but see Cleome africana). The flowers are also sometimes initially disymmetric, as in Brassicaceae, so this might be another synapomorphy at a higher level. However, the basal condition for Cleomaceae may be monosymmetry even early in development, with the abaxial sepal much enlarged and more or less covering the rest of the developing flower - almost cochleate aestivation, but there is in fact a variety of floral morphs in Brassicaceae and the phylogenetic structure at the base of Cleomaceae is not well supported (Patchell et al. 2010, esp. 2011).

For general information, see Kers (2002: in Capparaceae), for embryology, see Sachar (1956b), anther dehiscence, see Mitchell-Olds et al. (2005).

Phylogeny. For a phylogeny of Cleome and its immediate relatives, see Catalan et al. (2007) and Inda et al. (2008b). Relationships within Cleomaceae have rather little support, even if there are suggestions that Cleome itself is widely scattered on the tree (Hall 2008). A recent ITS study with quite broad sampling again recovered a highly paraphyletic Cleome; rooting was somewhat of a problem and there was little support for the backbone of the tree, but guite good support for much of the finer detail (Feodorova et al. 2010).

The small-flowered Dipterygium, placed in Capparaceae in a subfamily by itself by Kers (2002), is possibly to be included here. It has six stamens that are all equal in length, a filiform style, a winged, 1-seeded nut, and incumbent cotyledons; methylglucosinolates are recorded from the plant (Hedge et al. 1980).

Classification. Generic limits are unsatisfactory and will need attention, although I am sure that the wholesale dismemberment of Cleome that is under way (see Feodorova et al. 2010 for what has to be done and what still needs to be done) is a course that I would have taken...

Synonymy: Oxystylidaceae Hutchinson

BRASSICACEAE Burnett, nom. cons.//CRUCIFERAE Jussieu, nom. cons. et nom. alt.   Back to Brassicales

(Nortropane alkaloids +), methyl glucosinolates 0; roots lacking mycorrhizae; cork ?always deep-seated; (included phloem +); stomata anisocytic; hairs variously furcate; (leaves pinnately lobed), stipules 0; bracts 0 (foliaceous); floral development closed, (flowers disymmetric; monosymmetric); (C fringed or lobed); A (2, 4, to 24), the two outer shorter than the four inner [tetradynamous], about as long as petals; lateral nectary lobes outside inner A, etc.; pollen trinucleate, surface often reticulate; gynophore 0 (+); ovary with commissural septum (0 - Pringlea), style often +, short, stigma commissural; ovules slender, outer integument 2-4(-5) cells across, inner integument (2-)3-8(-15) cells across, endothelium +, parietal tissue ca 1 cell across, hypostase +; (megaspore mother cells several); seed/embryo folded, (spiral), coat not invaginated, testa 3-layered, exotestal cells reticulately thickened on radial walls, often mucilaginous, endotesta lignified, thickenings U-shaped or on anticlinal walls alone, (unthickened), not crystalliferous, tegmen multiplicative or not, not persistent; chalazal endosperm cyst +, endosperm 1-layered, radicle-hypocotyl short to long; n = (4-)8(-13); duplication of PHYB -> PHYD gene; sporophytic self-incompatibility system present.

338[list]/3710 - two groups below. World-wide, esp. N. temperate (map: from Vester 1940; Hultén 1971). [Photos - Collection].

Aethionemeae Al-Shehbaz, Beilstein & E. A. Kellogg

Plant glabrous; nortropane alkaloids +; 3 veins on petal claws, median nectaries 0, (1-)2-4(-8) ovules/carpel; fruit angustiseptate; n = 7, 8, 11, 12, 14...

1-2/70. The Mediterranean and Europe to Afghanistan (map: from Mark Menke, pers. comm.). [Photo - Flowers.]

The Rest.

(Eglandular hairs branched, stellate, T-shaped); (styles long); cotyledons also variants of conduplicate-incumbent, etc.; genome duplication.

300/3710: Draba (365), Cardamine (200), Erysimum (225), Lepidium (230), Alyssum (195), Arabis (120), Boechera (ex Arabis - 110), Physaria (105: inc. Lesquerella), Rorippa (85), Heliophila (80), Isatis (80), Noccaea (80), Thlaspi (55), Biscutella (55), Matthiola (50), Descurainia (50), Hesperis (45), Sisymbrium s. str. (45: only Old World). World-wide, esp. N. temperate (but less E. North America, and even more so humid lowland tropics). [Photo - Flowers, Flowers, Fruit.]

Synonymy: Arabidaceae Döll, Drabaceae Martynov, Erysimaceae Martynov, Isatidaceae Döll, Raphanaceae Horaninow, Schizopetalaceae A. Jussieu, Sisymbriaceae Martynov, Stanleyaceae Nuttall, Thlaspiaceae Martinov

• Most Brassicaceae are herbs, quite often with a cabbagey smell and a pleasant, biting taste, and a number of taxa have distinctive small, branched, eglandular hairs. The ebracteate inflorescence is corymbose, the flowers have four, clawed petals (the alternate name 'Cruciferae' for this family is taken from the cross-like arrangement of these four petals), six stamens with relatively short filaments (so the anthers are included), the gynoecium has two carpels and a short style, and the dehiscent fruit has both persistent, almost woody placental strands and a membranous false septum.

Evolution. Divergence & Distribution. Stem group Brassicaceae have been estimated to be 22-18 million years old (Wikström et al. 2001), some 41 million years old (Schranz & Mitchell-Olds 2006), or some 50 million years old, with divergence of Aethionemeae from the rest ca 40 million years before present (Al-Shehbaz et al. 2006, see also Koch 2011). Recent estimates suggest an age of (45-)19(-1) million years before present for stem group Brassicaceae, (35-)15(-1) million years for crown group diversification, and (28-)11(-1) million years for diversification of the family minus Aethionema (Franzke et al. 2009: the figure are 95% HPD [high probability distribution]); the ages quite similar similar to those of Al-Shehbaz et al. (2006) - (49.4-)37.6(-24.2) million years - have recently been suggested by Couvreur et al. (2010 and summary; note that it is not certain that the fossil Dressiantha is a member of Brassicales). Couvreur et al. (2010: Table 3) give times of diversification of various clades within the family. However, Beilstein et al. (2010) suggest an age of (64.2-)54.3(-45.2) million years for crown group diversification, (54.3-)46.9(-39.4) million years for diversification of the family minus Aethionema, and comparably older ages for other nodes in the family (crown group Arabidopsis itself may be (17.9-)13.0(-8.0) million years old). Koch (2012) suggests ages for the rather isolated Cochlearieae that span (22.3-)13.8, 10.3(-3.3) million years, depending on the gene used. Clade ages seem particularly uncertain in Brassicaceae.

Thlaspi primaevum, some 30.8-29.2 million years old, is known from Oligocene deposits in Montana (Manchester & O'Leary 2010); there are even impressions of the distinctive seeds on the capsule wall (Beilstein et al. 2010). However, the attribution of this fossil to the family, let alone the genus, has been questioned (Franzke et al. 2011).

The stem clade may have been adapted to warm and humid conditions in the east Mediterranean area, then moving into more open and dry environments (Franzke et al. 2009). It has also been suggested that Brassicaceae originated in a tropical environment, subsequently radiating with the onset of aridification and global cooling in in the mid-Tertiary, a diversification perhaps associated with a genome duplication after the divergence of Aethionemeae (Couvreur et al. 2010: see below). Initial diversification was in the Old World, perhaps in the Irano-Turanian region (Franzke et al. 2009, 2011 and references), and there there seems to have been a rather rapid radiation of the lineages that encompass most of the diversity of the family today, there still being little resolution along the backbone of the family (Beilstein et al. 2006; Al-Shehbaz et al. 2006; Bailey et al. 2006a, b; Couvreur et al. 2010).

Biogeographical histories within the family can be complex. Thus chloroplast and nuclear genomes of Californian and African ancestry variously combined in Antipodean Lepidium (Dierschke et al. 2009). Kiefer et al. (2009) discuss the phylogeographic structure of the speciose largely North American Boechera; divergence seems to be a Pleistocene phenomenon with considerable hybridization and apomixis. Draba is a young polyploid complex in which there is considerable geographical structure in the distribution of diploids and polyploids (Jordon-Thaden & Koch 2008; see also Jordon-Thaden et al. 2010).

Ecology & Physiology. As is well known, glucosinolate diversity in the family is considerable, as is variation in content between different species in the same community. The presence of particular glucosinolates may induce oviposition by Pieris butterflies, whether or not the crucifer with that glucosinolate is edible or kills the caterpillar (Chew 1979, see also 1988); only after mating will the butterfly respond to the odour (Ikeura et al. 2010). Parasitoids are also involved, and the complexity of the interactions increases exponentially (Fatouros et al. 2008). Caterpillars of cabbage white butterflies (Pieris spp.) and the diamond-back moth (Plutella xylostella) are able to convert the glucosinolates that are produced when they damage plant tissue as they eat into non-toxic substances such as nitriles (Wittstock et al. 2004; Ratzka et al. 2002); Winde and Wittstock (2011) discuss these and other ways in which insects can avoid the deleterious effects of glucosinolates. Insects such as the cabbage aphid, Brevicoryne brassicae, sequester the glucosinolates, the aphid even producing its own myrosinases that break the glucosinolates down and so helping to deter predators (Kazana et al. 2007).

According to Medve (1983), any mycorrhizal associations in the roots of Brassicaceae are at best weak and facultative. Thus although arbuscular mycorrhizae have recently been reported from Thlaspi, it is doubtful if an effective symbiosis results (Regvar et al. 2003). Indeed, glucosinolates can depress vesicular-arbuscular mycorrhizal activity, and this is a possible element in the invasive capabilities of Alliaria petiolata in parts of North America (Wolfe & Klironomos 2005). For more on glucosinolates, see the introduction to the order.

Arabidopsis thaliana can take up nitrogen in an organic form as amino acids (Hirner et al. 2006), although the general significance of this is unclear. Most of the nickel and zinc hyperaccumulators known in flowering plants occur in Brassicaceae (Baker & Brooks 1989; Grennan 2009 for a review); Freeman et al. (2009) suggests that selenium, accumulated by Stanleya pinnata, protects the plants against herbivory by prairie dogs.

Floral Biology & Seed Dispersal. Floral variation in the family is quite extensive, given a rather sterotypical floral formula - K4, C4, A6, G (2). The flowers of Iberis amara are monosymmetric, although the whole corymbose inflorescence is functionally more like a single polysymmetric flower with radiating petals - the two large petals of the outermost flowers (Busch & Zachgo 2007). Genera like Streptanthus have strongly monosymmetrical flowers borne on a more elongated axis. Several taxa have more or less fimbriate (Ornithocarpus, Schizopetalum) petals, while those of Draba, for example, are bilobed. For details of the action of the sporophytic incompatability system of the family, see Tarutani et al. (2010 and references).

More or less explosively dehiscent capsules are common in the family, but the extotesta is often mucilaginous (e.g. d'Arbaumont 1890), helping either in the establishment of the dispersed seed when it rains, or the further dispersal of the seed if it becomes attached to animals. Brassiceae are distinctive in that many members have heteroarthrocarpic fruits; these consist of an apical, indehiscent portion that does not differentiate into valve tissue, and a basal portion that is typical valve-type tissue and that may or may not be dehiscent (Hall et al. 2011); the two parts of the fruit may separate transversely and there are rhus very different means of dispersal on the one plant. Phylogenetic relationships in the tribe are uncertain, but heteroarthrocarpy is unlikely to be a synapomorphy for it, and the feature has evolved more than once there (Hall et al. 2011).

Plant-Animal Interactions. The association between the pseudoflower-forming Puccinia rust and host has been much studied in Brassicaceae (Roy 1993 [particularly fine photograph], 2001; Ngugi & Scherm 2006). Insects come to the pseudoflowers, attracted both by the colouration of the leaves and floral fragrances. These latter differ from those of both of the crucifer host and the other plants in the general area, containing i.a. constitutively-released isothiocyanates (Raguso & Roy 1998). They pick up the rich fructose nectar secreted by the fungus along with the fungal spermatia in the nectar; they then fly to another "flower", deposit the spermatia and pick up more. On combination of spermatia of the appropriate mating types, diploid aecia are produced, the "flower" stops producing nectar, and the petals, i.e. the leaves, become green. Ruxton and Schaefer (2011) discuss the evolution of such associations. The oomycete Albugo, the white blister rust, parasitizes a number of species of Brassicaceae (Ploch et al. 2010), although the genus quite commonly also persists as a symptomless endophyte (Ploch & Thines 2011).

Genes & Genomes. There has been extensive hybridization and genome duplication in the family at various times; for example, there is extensive polyploidy and hybridization in Cardamine (Lihová & Marhold 2006). Duplication of the whole genome seems also to have occurred fairly soon after the split from the Capparaceae and has been dated to 34-25 million years before present, the Ata palaeopolyploidization (Vision et al. 2000; Blanc & Wolfe 2004; Barker et al. 2009), which perhaps occurred after the divergence of the Aethionema clade (Blanc et al. 2003; de Bodt et al. 2005; Schranz & Mitchell-Olds 2006; Franzke et al. 2009); see also the findings of Galloway et al. (1998) on the pattern of duplication of the ADC (arginine decarboxylase) gene. Soltis et al. (2009: duplication unlocalized) and Franzke et al. (2011) suggest that diversification in Brassicaceae may be connected to this genome duplication. There is also further extensive genome duplication within the family (Kellogg & Bennetzen 2004), and this has happened more than once (Vision et al. 2000; Blanc & Wolfe 2004; Blanc et al. 2007; Lysack et al. 2007); "diploid" species like Brassica oleracea, with n = 9, are hypothesised to be ancestral hexaploids (Mitchell-Olds et al. 2005). Jaillon and Eury et al. (2007, and references) suggest that Arabidopsis has had two whole genome duplications, although there is some discussion as to exactly how many bouts of duplication have occurred. There has been extensive transposition of genes in the Arabidopsis genome - between 1/4 and 3/4 of the genes may have moved some time after the origin of the order[?] (Freeling et al. 2008; see also Schranz et al. 2007). The origin of a trnF pseudogene has been associated with a duplication in the common ancestor of the Halimolobus + Boechera + Cardamine clade, some 21-16 million years before present (Koch et al. 2005). Franzke et al. (2011) discuss genome duplication events in detail, suggesting that several remain to be disocovered. See Mathews and McBreen (2008) for the duplication of PHYBgene.

It has been suggested that n = 8 is ancestral in the family, with subsequent extensive genome rearrangement (Lysack et al. 2006). Details of karyotype evolution in the family are of considerable interest. Thus one major clade (lineage I) has x = 8, as does lineage II, while another clade that includes lineage II has x = 7, but with subsequent increase (Mandáková & Lysack 2008; Franzke et al. 2011), indeed, x = 4 may be the pre-duplication genome (Franzke et al. 2011 and references).

Chemistry, Morphology, etc. For tocopherols in Brassicaceae, see Goffman et al. (1999), Badami and Patil (1981) for seed fatty acids, and Harborne (1999, but sampling) for distinctive sulphur-containing phytoalexins. Bowman (2006) discusses morphology in general in the context of comparative developmental genetics. The border cells of the root cap dissociate in rows (Driouich et al. 2006); other Brassicales should be examined for this character. There are rarely glandular stipules in the inflorescence and elsewhere (see Weberling 2006 for a summary, also Bowman 2006).

There is a long-standing controversy in the family concerning the six stamens: did they arise by dédoublement or by reduction? Stamen number may sometimes increase, thus Megacarpaea polyandra may have 24 stamens. Brassicaceae have tryphine, in which some constituents of the disorganised tapetal cells are still visible, covering the pollen grains, not pollenkitt, as in other angiosperms (Pacini & Hesse 2005), although details of the distribution of this feature are unclear. Another controversy concerns carpel number. The commissural stigmas of Brassicaceae have been supposed to indicate that the gynoecium is basically 4-carpellate. Although such stigmas are notably common in groups with parietal placentation, there may be normally oriented bundles outside the inverted placental ventral carpellary bundle in Crataeva religiosa, perhaps indicative of an original 4-carpellate condition with axile placentation (Dickison 2000, but cf. Brückner 2000). Note, however, that taxa with four carpels are uncommon in Brassicales. The chalazal endosperm cyst may be involved in the movement of metabolites into the developing seed, there being transfer cells around it (Brown et al. 2004). Guignard (1893) suggested that the seed coat of Lunaria was endotegmic; there was no lignification in the testa at all. It would be interesting to know if duplication of the PHYB gene occcured before the split of the Aethionema clade from the rest of the family.

See Vaughan and Whitehouse (1971), Prasad (1975) and Bouman (1975) for ovules and seed coat, Erbar and Leins (1997a, b) and Leins and Erbar (2010) for floral development, and Al-Shehbaz (1984), Appel and Al-Shehbaz (2002), Koch et al. (2003), Hurka et al. (2005) and Mitchell-Olds et al. (2005) for general information and references. See also Pammel (1897): testa anatomy), Khoul et al. (2002: testa surface morphology), Khalik et al. (2002 and references) for pollen morphology; Brown et al. (2004) for the endosperm cyst (Aethionema not sampled, cyst probably not in in Cleome, at least). Lysack et al. (2005) discuss chromosome triplication in Brassiceae and Lysack et al. (2009) the evolution of genome size, which is not particularly linked with chromosome number, while Warwick and Al-Shehbaz (2006) summarize chromosome numbers. See Grubb and Abel (2006) for glucosinolate metabolism; Brock et al. (2006) for nortropane alkaloids (in both Aethionema and Cochlearia, i.a.); Beilstein et al. (2006) for trichome evolution; Koch and Kiefer (2006) for a summary of biogeographic studies; and Bernadello (2007) for nectary variation. Schweingruber (2006) gives details of phloem and xylem anatomy.

Phylogeny. Of late, substantial progress has been made in providing a phylogenetic framework for the family (e.g. Koch et al. 2001: Koch 2003; Beilstein et al. 2006) and realigning taxa accordingly. Aethionema, a variable but poorly known genus with angustiseptate fruits, is sister to the rest of the family (e.g. Zunk et al. 1996, 1999; Koch et al. 2001; Beilstein et al. 2006). For the limits of Aethionema, see Khosravi et al. (2008). However, the relationships between many of the tribes - and perhaps even the circumscription of some - are still unclear (Beilstein et al. 2006; Al-Shehbaz et al. 2006; Bailey et al. 2006a, b; Franzke et al. 2009; Mummenhoff et al. 2010; Zhao et al. 2010).

Warwick and Sauder (2005) found a monophyletic Brassiceae that needed but little adjustment from its classical delimitation, but "well-known" genera such as Brassica, Diplotaxis, and Erucastrum were polyphyletic; as they noted, this should affect how breeders went about their business. Relationships there are still somewhat unclear (Hall et al. 2011). For a phylogeny of Draba, with three major clades, mostly perennial and a number Arctic-alpine, see Jordon-Thaden et al. (2010); for the circumscription of Arabis, see Koch et al. (2010). Although Arabidopsis thaliana is perhaps the most important model vascular plant in biology, the limits of the genus Arabidopsis itself are only now being established (see Clauss & Koch 2006 for a discussion of its immediate relatives). Mummenhoff et al. (1997), Koch and Mummenhoff (2001) and Meyer (2003) discuss generic limits surrounding Thlaspi, a polyphyletic genus; the sometimes invasive Alliaria petiolata, with very different fruit morphology, is to be placed around here. Sisymbrium s. str. is restricted to the Old World, the New World taxa being unrelated and mixed in with Thelypodieae (Warwick et al. 2002, 2006a). For a study of Alysseae and related tribes, see Warwick et al. (2008), of Schizopetaleae and Thelypodieae, Warwick et al. 2009), of Cochlearieae, Koch (2012), of Matthiola, Jaén-Molina et al. (2009), of Boechera and relatives, Kiefer et al. (2009a, b), of some Asian taxa, German et al. (2009: ITS), of Iranian Brassicaceae, see Khosravi et al. (2009), in Chorisporeae-Dontostemoneae, see German et al. (2011), and in some Chineae Brassicaceae, see Liu et al. (2011).

See also Price et al. (1994), Galloway et al. (1998), and Warwick et al. (2007) for other phylogenetic studies.

Classification. Earlier classifications of the family are something of a disaster area. Thelypodieae, with their exserted stamens, gynophore, etc., appeared to be similar to Capparaceae and to have a plesiomorphic morphology, but are in fact derived within Brassicaceae. Both generic and tribal limits have tended to be based on single characters like fruit shape (reified as fruit "types") and embryo curvature and have turned out to be very unsatisfactory (e.g. Mummenhoff et al. 1997; Al-Shehbaz et al. 2006; Moazzeni et al. 2010). Some genera such Draba and Lepidium are monophyletic or largely so on both morphological and molecular grounds, but others, such as Brassica, are not (e.g. Mitchell-Olds et al. 2005 and references). The combination of flattened fruits and accumbent cotyledions has arisen perhaps 54 times in the family, and Lepidieae sensu Schultes are now placed in ca 13 tribes (I. Al-Shehbaz, pers. comm.). There has been parallel or convergent evolution of just about all the morphological features used to distinguish genera (Koch 2003; Al-Shehbaz et al. 2006; Bailey et al. 2006a, b; Beilstein et al. 2008; Franzke et al. 2011). Hence the major disagreements over generic limits (see above). Although "intergeneric" hybridisation has been considered to be quite common in Brassicaceae (see summary in Warwick et al. 2006), this is of uncertain significance given these problems with generic circumscriptions.

Al-Shehbaz et al. (2006) recognised 25 monophyletic tribes based on well-supported clades recognised in molecular studies; they included 260 of the 338 genera of the family then recognised in these tribes (see also Beilstein et al. 2008; German & Al-Shehbaz 2008). The number of tribes is now 44+, with 318 genera included (Warwick et al. 2010: see Supplementary Table 1 for contents of these tribes; see also Koch 2012). Things are getting overly splitty, largely because the limits of named tribes are not being expanded as previously unplaced genera end up as sister taxa to these tribes, but it is also the result of starting to construct a formal classification when relationships are imperfectly known.

For Isatideae, see Moazzeni et al. (2010). For the suggested placement of some of the more temperate annual species of Draba in separate genera, see Jordon-Thaden et al. (2010).

Warwick et al. (2006b) provide a species checklist for the family.