morphological and molecular data reveal three rather than...

14
Morphological and molecular data reveal three rather than one species of Sicyos (Cucurbitaceae) in Australia, New Zealand and Islands of the South West Pacic Ian R. H. Telford A,D , Patrizia Sebastian B , Peter J. de Lange C , Jeremy J. Bruhl A and Susanne S. Renner B A N.C.W. Beadle Herbarium, University of New England, Armidale, NSW 2351, Australia. B Systematic Botany and Mycology, University of Munich, Menzinger Strasse 67, 80638, Munich, Germany. C Ecosystems and Species Unit, Department of Conservation, Private Bag 68908, Newton, Auckland 1145, New Zealand. D Corresponding author. Email: [email protected] Abstract. Morphometric data have long suggested the existence of three species of Sicyos in Australia and New Zealand. Molecular data now corroborate this and place the three species in the context of the remaining 72 species of Sicyos, most of them in the New World. We here describe and illustrate Sicyos undara I.Telford & P.Sebastian from northern Queensland and S. mawhai I.Telford & P.Sebastian from New Zealand, and emend the circumscription of S. australis to account for the removal of these two extraneous elements. We also provide a key to the three species, map their ranges, and note their habitat and conservation status. Received 26 August 2011, accepted 7 December 2011, published online 6 June 2012 Introduction Sicyos L. is a genus of 75 species, most of them occurring in Central and South America, although the genus also underwent radiation on Hawaii and has two endemic species on the Galapagos Islands (Sebastian et al. 2012). In the South West Pacic region, Sicyos has been recorded from eastern Australia, Tasmania, New Zealand and smaller islands associated to these major landmasses, as well as Lord Howe Island, Norfolk Island and the Kermadec Islands. Molecular phylogenies for the Cucurbitaceae show that Sicyos and 11 other genera (with some 265 species altogether) form a clade in which the monotypic Queensland genus Nothoalsomitra I.Telford is sister to the remaining genera. This newly circumscribed tribe, named Sicyoeae, includes several New World genera, and also the large Old World genus Trichosanthes L. and the pantropical genus Luffa Mill. (Schaefer and Renner 2011). For the Australian material of Sicyos, two names are available, namely S. australis Endl. (Endlicher 1833) and S. fretensis Hook. f. (Hooker 1847). Both appear on specimens housed at K and BM, but have rarely been used in Australian herbaria. Hooker (1860) later subsumed his Australian species, together with New Zealand material, into a broadly dened S. angulatus L., a species based on material from the eastern United States. Probably because of Hookers eminent status and the scarcity of material in European herbaria, this broad concept became widely accepted (Bentham 1866; Bailey 1900; Allan 1961; Beadle et al. 1962; Curtis 1963; Willis 1972; Beadle 1976). Ferdinand von Mueller had annotated specimens from Gippsland, Victoria, and held in MEL as S. rivularis MS, a manuscript name never validated. Doubts about Hookers broad species concept were rst expressed by Allan (1961) in the Flora of New Zealandwho noted that the South West Pacic collections needed critical comparison with American material; still, he decided to maintain the name S. angulatus for New Zealand indigenous plants. Telford (1982), in his treatment of the Cucurbitaceae for the Flora of Australia,took up Endlichers name S. australis for Australian specimens and also stressed the heterogeneity of the South West Pacic material and the need for a revision. Comparing S. angulatus material from New Zealand and America, Sykes (in Connor and Edgar 1987) found clear differences, but still referred the New Zealand material to S. australis, a decision followed by other New Zealand authors (Webb et al. 1988; de Lange et al. 2004, 2006, 2009b). Chromosome numbers (de Lange and Murray 2002) and observations of morphological differences (Brandon et al. 2004; Forester and Townsend 2004) in the meantime have made it clear that more than one species of Sicyos occurs in New Zealand. Post-1982 collections from Undara Lava Tubes in northern Queensland also suggested the existence of a new species in Australia, prompting a phenetic analysis of 34 herbarium sheets, covering the morphological and geographic range of material available by 1997 (I. R. H. Telford, unpubl. data). The results strongly supported the existence of three species in the South West Pacic region, a nding supported Journal compilation Ó CSIRO 2012 www.publish.csiro.au/journals/asb CSIRO PUBLISHING Australian Systematic Botany, 2012, 25, 188201 http://dx.doi.org/10.1071/SB11032

Upload: others

Post on 07-Jun-2020

1 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Morphological and molecular data reveal three rather than ...renners/Telford_Sicyos_AustrSystBot2012.pdf · Morphological and molecular data reveal three rather than one species of

Morphological and molecular data reveal three ratherthan one species of Sicyos (Cucurbitaceae) in Australia,New Zealand and Islands of the South West Pacific

Ian R. H. TelfordA,D, Patrizia SebastianB, Peter J. de LangeC, Jeremy J. BruhlA

and Susanne S. RennerB

AN.C.W. Beadle Herbarium, University of New England, Armidale, NSW 2351, Australia.BSystematic Botany and Mycology, University of Munich, Menzinger Strasse 67, 80638, Munich, Germany.CEcosystems and Species Unit, Department of Conservation, Private Bag 68908, Newton, Auckland 1145,New Zealand.

DCorresponding author. Email: [email protected]

Abstract. Morphometric data have long suggested the existence of three species of Sicyos in Australia and New Zealand.Molecular data now corroborate this and place the three species in the context of the remaining 72 species of Sicyos, most ofthem in theNewWorld.Wehere describe and illustrateSicyos undara I.Telford&P.Sebastian fromnorthernQueensland andS. mawhai I.Telford & P.Sebastian from New Zealand, and emend the circumscription of S. australis to account for theremoval of these two extraneous elements.We also provide a key to the three species, map their ranges, and note their habitatand conservation status.

Received 26 August 2011, accepted 7 December 2011, published online 6 June 2012

Introduction

Sicyos L. is a genus of 75 species, most of them occurring inCentral and South America, although the genus also underwentradiation on Hawaii and has two endemic species on theGalapagos Islands (Sebastian et al. 2012). In the South WestPacific region, Sicyos has been recorded from eastern Australia,Tasmania, New Zealand and smaller islands associated to thesemajor landmasses, as well as Lord Howe Island, Norfolk Islandand the Kermadec Islands. Molecular phylogenies for theCucurbitaceae show that Sicyos and 11 other genera (withsome 265 species altogether) form a clade in which themonotypic Queensland genus Nothoalsomitra I.Telford issister to the remaining genera. This newly circumscribed tribe,named Sicyoeae, includes several New World genera, and alsothe large Old World genus Trichosanthes L. and the pantropicalgenus Luffa Mill. (Schaefer and Renner 2011).

For theAustralianmaterial of Sicyos, two names are available,namely S. australisEndl. (Endlicher 1833) and S. fretensisHook.f. (Hooker 1847). Both appear on specimens housed atK andBM,but have rarely been used in Australian herbaria. Hooker (1860)later subsumedhisAustralian species, togetherwithNewZealandmaterial, into abroadlydefinedS. angulatusL., a species basedonmaterial from the eastern United States. Probably because ofHooker’s eminent status and the scarcity of material in Europeanherbaria, this broad concept became widely accepted (Bentham1866; Bailey 1900; Allan 1961; Beadle et al. 1962; Curtis 1963;Willis 1972; Beadle 1976). Ferdinand vonMueller had annotated

specimens from Gippsland, Victoria, and held in MEL asS. rivularis MS, a manuscript name never validated.

Doubts about Hooker’s broad species concept were firstexpressed by Allan (1961) in the ‘Flora of New Zealand’ whonoted that the South West Pacific collections needed criticalcomparison with American material; still, he decided tomaintain the name S. angulatus for New Zealand indigenousplants. Telford (1982), in his treatment of the Cucurbitaceae forthe ‘Flora of Australia,’ took up Endlicher’s name S. australisfor Australian specimens and also stressed the heterogeneity ofthe South West Pacific material and the need for a revision.Comparing S. angulatus material from New Zealand andAmerica, Sykes (in Connor and Edgar 1987) found cleardifferences, but still referred the New Zealand material toS. australis, a decision followed by other New Zealand authors(Webb et al. 1988; de Lange et al. 2004, 2006, 2009b).

Chromosome numbers (de Lange and Murray 2002) andobservations of morphological differences (Brandon et al.2004; Forester and Townsend 2004) in the meantime havemade it clear that more than one species of Sicyos occurs inNew Zealand. Post-1982 collections from Undara Lava Tubesin northern Queensland also suggested the existence of a newspecies in Australia, prompting a phenetic analysis of 34herbarium sheets, covering the morphological and geographicrange of material available by 1997 (I. R. H. Telford, unpubl.data). The results strongly supported the existence of threespecies in the South West Pacific region, a finding supported

Journal compilation � CSIRO 2012 www.publish.csiro.au/journals/asb

CSIRO PUBLISHING

Australian Systematic Botany, 2012, 25, 188–201http://dx.doi.org/10.1071/SB11032

Page 2: Morphological and molecular data reveal three rather than ...renners/Telford_Sicyos_AustrSystBot2012.pdf · Morphological and molecular data reveal three rather than one species of

by a genus-wide phylogeny (Sebastian et al. 2012). Here, wedescribe, map and illustrate the two new species, S. undara andS. mawhai (names used hereafter in the present paper),and provide an emended circumscription of S. australis sensustricto. We also provide a key to all three species.

Materials and methods

Phenetic analysis

We here use a morphological species concept based on suitesof covarying attributes (Stuessy 1990), and we tested thedelimitation of our new species using graphical and statisticalanalyses ofmorphological datasets (Sneath and Sokal 1973). Themorphological dataset was derived from herbarium specimensheld at AK, BM, BRI, CANB, CHR, HO, JCU, K, MEL, NE,NSW and W. Each operational taxonomic unit (OTU) consistedof a single gathering, whether mounted on a single sheet ormultiple sheets. Thirty-four OTUs were selected to cover therange of the genus in Australia and New Zealand (Table 1), andthe characters used are listed in Table 2. States were assignedusing dried herbarium specimens for inflorescence, fruit and seed

characters, and rehydrated material for floral characters(Appendix 1); vegetative characters, such as leaf shape andpubescence, were avoided because of their well knownplasticity in the family.

The data matrix was analysed using the pattern analysissoftware package PATN version 3.03 (Belbin and Collins2006), with all characters given equal weight and range-standardised by using the Gower metric association measure.Two options of PATN (Belbin 1990a, 1990b) were used to definepopulation groupings, namely, cluster analysis, employing theflexible unweighted pair group arithmetic averaging (UPGMA),presented graphically as a phenogram, and ordination analysis,using semi-strong hybrid multidimensional scaling (SSH MDS)in a three-dimensional scatter plot.

Molecular analysisFor the molecular phylogeny shown here, we used a modifiedversion of the matrix used in Sebastian et al. (2012). That studyhad 112 accessions, representing 87 species of Sicyoeae,including the type species of all relevant generic names. Here,

Table 1. Operational taxonomic unit (OTU) numbers, codes used in analyses and vouchers for phenetic analysisVoucher data present collector’s name, number and herbarium where housed. Herbarium accession number is used whencollector’s number is absent. Where more than one collector involved, only the first collector is given. Herbarium codes follow

Index Herbariorum, available at http://sweetgum.nybg.org/ih/ (accessed 12 October 2011)

OTU Code Locality Collector

1 NZ1 Cuvier Island, New Zealand R. Moynihan AK119648 (AK)2 NZ2 Hen Island, New Zealand D. Merton CHR183501(CHR)3 NZ3 Little Barrier Island, New Zealand W.R. Sykes 147/85 (CHR)4 NZ4 Auckland, New Zealand J.J. Bruhl 2915 (NE)5 NZ5 Coromandel Island, New Zealand A.E. Esler 3584 (CHR)6 NZ6 Tahuna Road, Te Teko, New Zealand W. Stahel CHR566019 (CHR)7 NZ7 White Pine Bush, New Zealand M. Steverinson CHR551181(CHR)8 NZ8 Poor Knights Island, New Zealand G.N. Park CHR276071 (CHR)9 NZ9 Pakaraka, New Zealand K. Riddell AK238904 (AK)10 NZ10 Whenuakite, New Zealand C. Wallace AK199785 (AK)11 NZ11 Three Kings Islands, New Zealand A.E. Wright 5266 (AK)12 NZ12 Three Kings Islands, New Zealand A.E. Wright 6061 (AK)13 KI1 Kermadec Islands T.F. Cheeseman MEL591390 (MEL)14 KI2 Kermadec Islands C.J. West CHR518213 (CHR)15 KI3 Kermadec Islands W. Sykes 1523/K (CHR)16 NI1 Norfolk Island W. Robinson (K)17 NI2 Norfolk Island F. Bauer 110 (W)18 LH1 Lord Howe Island leg. ign. MEL593267–8 (MEL)19 LH2 Lord Howe Island J. Fullagar 118 (MEL)20 LH3 Lord Howe Island leg. ign. (K)21 TS1 Sisters Island, Tasmania J. Whinray 57 (CANB)22 TS2 Sisters Island, Tasmania J. Whinray 492 (CANB)23 VG1 Buchan River, Victoria F. Mueller MEL593278(MEL)24 VG2 Tambo River, Victoria F. Mueller MEL593283 (MEL)25 NS1 Tuross River, New South Wales J. Whaite 478 (NE)26 NS2 Cambewarra Mountain, New South Wales I.R. Telford 11793 (CANB)27 NS3 Nymboida River, New South Wales J.J. Bruhl 2102 (NE)28 QL1 Killarney, Queensland I.R. Telford 12909 (NE)29 QL2 Isla Gorge, Queensland P. Sharpe 660 (BRI)30 QL3 Eungella, Queensland N. Byrnes 3699 (BRI)31 QL4 Fanning River Caves, Queensland B. Jackes S8643 (JCU)32 QL5 Wind Tunnel, Undara, Queensland I.R. Telford 13319(NE)33 QL6 Pinwill Cave, Undara, Queensland V.J. Neldner 2780 (BRI)34 QL7 Pinwill Cave, Undara, Queensland I.R. Telford 13317 (NE)

New species of Sicyos in the South West Pacific Australian Systematic Botany 189

Page 3: Morphological and molecular data reveal three rather than ...renners/Telford_Sicyos_AustrSystBot2012.pdf · Morphological and molecular data reveal three rather than one species of

we focus the sample to 34 of these to provide two accessions foreach Australian and New Zealand species of Sicyos, and soundsampling across other relevant species.

Thematrix comprised six chloroplast regions, the trnL intron,the trnL–F, rpl20–rps12, trnS–G and psbA–trnH intergenicspacers, and the rbcL gene, and the nuclear rDNA internaltranscribed spacers ITS1 and ITS2, with the intervening 5.8 Sgene, for a length of 5399 aligned positions (4527 chloroplastand 872 nuclear). GenBank accession numbers and voucherinformation are listed in table S1 of Sebastian et al. (2012).DNA extraction, polymerase chain reactions, sequencing anddata preparation were carried out using standard approaches(Sebastian et al. 2010a, 2012). Maximum likelihood (ML)analyses and bootstrap searches using 500 replicates werecarried out using RAxML version 7.2.8 (Stamatakis 2006),relying on the GTR+G model, with model parametersestimated over the duration of specified runs.

Results

Phenetic analysis

The 34 specimens from New Zealand, Tasmania, Lord HoweIsland, Kermadec Islands, Norfolk Island, and the Australianstates Victoria, New South Wales and Queensland used in thephenetic analyses clearly fall into three groups on the basis ofUPGMA clustering (Fig. 1) and SSH MDS ordination (Fig. 2).Table 3 shows the correlation of characters with ordinationpatterns: Fruit and seed attributes contributed most to phenetic

Table 2. Morphological characters used for ordination and clusteranalysis of Sicyos in the South West Pacific

Number Character

Female inflorescence1 Peduncle length (mm)2 Flowers per inflorescence3 Corolla diameter (mm)

Fruit4 Fruit length (mm)5 Fruit diameter (mm)6 Aculei length (mm)7 Aculei per fruit8 Seed length (mm)

Male inflorescence9 Inflorescence length (mm)10 Peduncle length (mm)11 Flower number per inflorescence12 Pedicel length (mm)13 Corolla diameter (mm)

Staminal column length (mm)15 Staminal head diameter (mm)

NZ1NZ2KI2

Sicyos mawhai

Sicyos australis

Sicyos undara

NZ3

NZ5KI1NZ4NZ8

NZ11NZ12NZ6NZ9NZ7NS1NI2VG2LH3LH2VG1QL2LH1TS2TS1

NZ10NS3QL1

QL3QL4QL5QL6QL7

NI1NS2

KI3

0.06

65

0.23

23

0.39

82

0.56

40

0.72

99

Fig. 1. Phenogram based on Gower association and flexible unweighted pair group arithmetic averaging(UPGMA) of the operational taxonomic units (OTUs) of Sicyos in the South West Pacific. See Table 1 for OTUcoding and vouchers.

190 Australian Systematic Botany I. R. H. Telford et al.

Page 4: Morphological and molecular data reveal three rather than ...renners/Telford_Sicyos_AustrSystBot2012.pdf · Morphological and molecular data reveal three rather than one species of

differences. The results indicate that two species should berecognised in Australia, the widespread S. australis (OTUs16–31) and an entity from Undara (OTUs 32–34), which wedescribe below. S. fretensis (OTUs 21, 22) and Mueller’sunpublished S. rivularis (OTUs 23, 24) are grouped withinS. australis. New Zealand material groups into two species,the putative new species (OTUs 1–5, 8, 9, 11–16), namedS. mawhai below, and those (OTUs 6, 7, 10) grouping withS. australis, confirming their placement by Brandon et al.(2004).

Molecular analysis

The interpretation of the phenetic analyses, as indicating theexistence of three biological species, matches the molecularphylogeny (Fig. 3), which recovers three species-level groups

(judging from branch lengths of other species-level groups inSicyos) and, moreover, shows Australian S. undara as sister toS. australis and theNewZealand endemicS.mawhai. These threespecies stem from a common ancestor (Fig. 3) and are sister to aclade from the SouthWesternUnited States,Mexico andBolivia.The American species S. angulatus is placed distantly fromthe Australian–New Zealand clade (Fig. 3), matching themorphological differences observed by Sykes in Connor andEdgar (1987; see also Webb et al. 1988).

Discussion

Many Cucurbitaceae are successful transoceanic dispersers(Schaefer et al. 2009), and Sicyos, with four trans-Pacificrange disjunctions (Sebastian et al. 2012), constitutes a primeexample of this. The latter study dated the trans-Pacific dispersalof the ancestor of the Australian–New Zealand clade of Sicyos to5.6–1.7million years (Ma) ago, the divergence of theQueenslandspecies S. undara to 3.6–0.5Ma, and the arrival of the ancestor ofS. mawhai in New Zealand to 1.6–0.1 Ma. S. australis arrived inNew Zealand more recently.

Dispersal across the ~2000 km of ocean between Australiaand New Zealand has been inferred for many other plant groups(Perrie et al. 2003;Chandler et al. 2007; de Lange et al. 2010; Tayet al. 2010; Prebble et al. 2011) and numerous species, includingapparently recent arrivals (Sykes and de Lange 1993; de Langeet al. 2009a), show trans-Tasman disjunctions.

As is true of their closest New World relatives, the small,one-seeded fruits of the three South West Pacific species bearretrorsely barbed spines, which may facilitate transport in birds’plumage, and both S. australis and S. mawhai sometimes occur incoastal vegetationwhere theymay come in contact with petrels orshearwaters, which regularly travel between the southern Pacificlandmasses.

Sicyos mawhai is the first endemic New Zealand species ofthe Cucurbitaceae family known. Both Sicyos species nowknown from New Zealand are threatened by cucurbit diseases,such as cucumber mosaic virus (CMV) and zucchini yellowmosaic virus (ZYMV) (Delmiglio and Pearson 2006), whichmay also be responsible for the decline of the two Sicyosspecies on the Galápagos archipelago (Sebastian et al. 2010b).Australian populations, by contrast, appear healthy, with noobvious decline through disease (I. R. H. Telford, pers. obs.).

Taxonomy

Sicyos australis Endl., Prodr. Fl. Norfolk. 67 (1833) emend.I.Telford

Type: NORFOLK ISLAND, western mountain, F. Bauer 110;holo: W!.

S. fretensis Hook.f., Hooker’s London J. Bot. 6: 473 bis (1847).

Type: Sisters’ Island, E coast of Flinders Island,R.Gunn [Oct.,1844, J. Milligan 563]; holo: K!; iso: BM, HO, K!, MEL!.

S. angulatus auct. non L.: Hooker, 143 (1856); Bentham, 3: 322(1866); Bailey, 701 (1900); Beadle et al., 199 (1963); Curtis, 237(1963); Willis, 2: 264 (1972); Beadle, 271 (1976).

S. aff. australis (b) (AK 289786; Mangere Stonefields) de Lange et al.(2009b).

1.543

–1.194

–0.9250.781

0.875

Sicyos mawhai

Sicyos acstralis

Sicyos undara

1.416

Fig. 2. Three-dimensional ordination of semi-strong hybridmultidimensional scaling (SSH MDS) for organisational taxonomic units(OTUs) of Sicyos in the South West Pacific. Stress = 0.0847.

Table 3. Correlation of character values with correlation of attributesfor semi-strong hybrid multidimensional scaling (SSHMDS) ordination

of Sicyos in the South West Pacific

Character X Y Z R2

5 0.798 0.355 –0.488 0.9238 0.298 0.666 –0.684 0.9187 0.255 0.377 –0.890 0.82115 –0.176 0.310 –0.934 0.81013 0.662 0.616 –0.428 0.7964 0.467 0.623 –0.627 0.78212 0.788 0.616 –0.019 0.7349 0.557 –0.171 –0.812 0.7103 0.800 0.009 –0.601 0.70714 –0.628 0.123 –0.769 0.6256 –0.649 –0.186 –0.738 0.62010 0.292 –0.457 –0.840 0.53411 0.966 –0.023 –0.257 0.5081 0.778 0.187 0.600 0.4462 0.463 0.797 –0.388 0.265

New species of Sicyos in the South West Pacific Australian Systematic Botany 191

Page 5: Morphological and molecular data reveal three rather than ...renners/Telford_Sicyos_AustrSystBot2012.pdf · Morphological and molecular data reveal three rather than one species of

Monoecious herbaceous climber; stems annual, up to 10mlong, 5-mm diam., sparsely hirsute with simple multicellularhairs and glandular hairs, glabrescent. Tendrils 3–5-branched.Leaves: petiole 10–65mm long, scaberulous; lamina ovate orbroadly ovate in outline, 35–195mm long, 45–200mm wide,cordate with the basal sinus broad, the lobes not overlapping,acuminate, shallowly to deeply palmately 5- or 7-lobed, thelobes broadly triangular, acute or acuminate, margins dentatewith apiculate teeth, sparsely scaberulous adaxially andabaxially with simple hairs, more densely and coarselyalong veins. Male inflorescence an 8–19-flowered raceme25–110(–155)mm long; peduncle 15(–125)mm long, sparselyscaberulous; rhachis glandular hairy. Male flowers: pedicels3–11mm long; hypanthium broadly campanulate, 2.4–2.7-mmdiam.; calyx lobes linear, 0.4–0.7mm long; corolla rotate,4.5–7.8mm in diam., mostly glabrous abaxially, the lobeapices puberulous, glabrous adaxially, white, 5-lobed, thelobes broadly triangular–ovate, obtuse; 2–2.6mm long; disc~1.2mm in diam.; staminal column 1.5–1.8mm long; staminalhead 1.7–2.2-mm diam. Female inflorescence an 8–11-floweredhead; peduncle 8–23(–33)mm long. Female flowers: subsessile;ovary ovate, attenuate, ~3mm long, 1.4-mm diam., minutely and

densely echinulate with barbed aculei; hypanthium above theconstriction broadly campanulate, ~1.5mm in diam., minutelyglandular hairy surrounding the disc; calyx lobes linear,0.4–0.7mm long; corolla 2.5–3.5-mm diam., mostly glabrousabaxially, the lobe apices puberulous, glabrous adaxially, white,5-lobed, the lobes triangular–ovate, obtuse, ~1.6mm long; disc~1mm in diam.; style ~1.4mm long; stigma 2-lobed, thelobes ~0.3mm long, recurved. Fruit ovate, rarely fusiform,6.4–9.5mm long, 2.8–4mm in diam., apically attenuate, thesurface �glabrous or scabridulous with short hairs, echinate;aculei dense, 1.6–3.6mm long, retrorsely barbed. Seedsellipsoidal, 4–5.5mm long, 2.8–3.4mm wide, brown.

Distribution

In eastern Australia, S. australis occurs from Fanning River,~60 km south-west of Townsville, Queensland, south to Orbost,Victoria, inland to the Great Dividing Range, and on Sisters’Island, off Flinders Island, Tasmania (Fig. 4).

In New Zealand, the first collection of which we are awarewas from Thames, North Island, by T. F. Cheeseman of theAuckland Museum. The collection date was not recorded;

Fig. 3. Maximum likelihood phylogram for 33 taxa of Sicyoeae based on combined sequences from chloroplastand nuclear data (5399 aligned nucleotides), analysed under the GTR+G substitution model. The tree is rooted onLinnaeosicyos. The Australasian clade is highlighted. Likelihood bootstrap values �75% are given at the nodes;geographic origin of the sequenced specimens follows the species name and asterisks indicate generic type species.

192 Australian Systematic Botany I. R. H. Telford et al.

Page 6: Morphological and molecular data reveal three rather than ...renners/Telford_Sicyos_AustrSystBot2012.pdf · Morphological and molecular data reveal three rather than one species of

however, consultation of the Cheeseman notebooks suggestssome time in the 1870s and 1880s (E. K. Cameron, pers.comm.). Recent collections (1988+) on the North Island havebeen made from Northland, Auckland, Waikato and the Bay ofPlenty.

Previously recorded from the Lord Howe and Norfolk Islandgroups, the latter being the type locality of the species, S. australishas not been collected from either of these island groups for morethan a century and is presumed extinct in these archipelagos(Green 1994).

Habitat and ecology

In mainland Australia, S. australis grows in dry rainforest(deciduous vine thicket) with Brachychiton australis on basaltor limestone in northern Queensland, and with B. rupestrisand Acacia harpophylla on cracking clay derived from basaltin central Queensland. In coastal and near-coastal montanehabitats from Eungella, Queensland, south to Gippsland,Victoria, S. australis grows in wetter eucalypt forest or onrainforest margins, at disturbed sites in closed forest, often atsheltered sites in gorges and on a wide variety of substrates.

In New Zealand, S. australis is a species of disturbedforest, forest margins, agricultural land, associated wastelandand roadsides. In forested habitats, it often associates withmembers of the Kunzea ericoides agg., Melicytus ramiflorus,Pseudopanax crassifolius, Vitex lucens, Corynocarpuslaevigatus, Dysoxylum spectabile and Metrosideros excelsa.Otherwise, it has been collected threaded through the exoticgrass Cortaderia selloana, or climbing over such naturalisedshrubs and trees as Ligustrum sinense, L. vulgare, Robiniapseudacacia, Populus nigra and Solanum mauritianum inwasteland, on occasion as a weed in maize (Zea mays) crops,or growing on ancient basalt stonewalls within the site of aformer, major pre-European settlement Maori agricultural site.At many of these habitats, S. australis exhibits a ‘weedy’ andinvasive nature, which has lead people to regard it as a recentnaturalisation rather than an indigenous species.

Although the species was first collected from New Zealandby Cheeseman in the late 19th century, it was not then recordedagain from that country until 1988 (when it was gatheredfrom a remote area of coastal forest on the eastern side of theCoromandel Peninsula). On this evidence, we think thatS. australis has naturally colonised New Zealand, probably via

10°0′0′′S

20°0′0′′S

30°0′0′′S

40°0′0′′S

150°0′0′′E 160°0′0′′E 170°0′0′′E

40°0′0′′S

30°0′0′′S

20°0′0′′S

150°0′0′′E 160°0′0′′E 170°0′0′′E

km

Fig. 4. Distribution of Sicyos in South-West Pacific. * S. australis; $ S. mawhai; ~ S. undara. Note that this representsthe historical range of species; localities where S. australis and S. mawhai are known to be extinct are shown.

New species of Sicyos in the South West Pacific Australian Systematic Botany 193

Page 7: Morphological and molecular data reveal three rather than ...renners/Telford_Sicyos_AustrSystBot2012.pdf · Morphological and molecular data reveal three rather than one species of

trans-Tasman avian dispersal, possibly on several occasions, orit has persisted undetected in the less well explored parts of thenorthern North Island of New Zealand until recently. Someevidence suggests that S. australis is much more likely to haverecently colonised New Zealand rather than recently spreadfrom long-established indigenous populations. Notably, thereis its specific association at two sites, ‘Jacks Bush’ (firstcollected 1993) and Whenuakite (first collected 1988), withanother naturally recently established trans-Tasman species,Plectranthus parviflorus, also first recognised from NewZealand in 1988 (Sykes and de Lange 1993), whereas severalother Northland occurrences of Sicyos are in the generalvicinity of another trans-Tasman colonist recently acceptedas indigenous to New Zealand, namely, Rorippa laciniata(de Lange et al. 2009a). Further, on the basis of herbariumevidence, it is significant that the first gatherings ofS. australis were all made from remote indigenous forestremnants in Northland and Coromandel at sites located wellaway from the main ports of entry to New Zealand, butcorresponding to well known sites of trans-Tasman dispersalby vagrants and at times subsequent colonisation and spread(de Lange 1997; Heenan and Forester 1997; de Lange andNorton 1998; de Lange et al. 2011). Once established at thesesites, it is only to be expected that Sicyos, as an undeniably‘weedy’ species with fruits well suited to avian and mammaliandispersal, would be rapidly spread by a host of indigenousand exotic birds, and widespread mammals such as opossum(Trichosurus vulpecula), let alone by humans. This wouldexplain its more recent spread throughout wasteland androadsides as well as its ‘spot’ occurrences as a volunteer inagricultural crops. In this regard, S. australis is following apattern already typical of many other undeniably long-established indigenous New Zealand plants with the samedispersal tendencies and which are now common in suchhabitats, e.g. Alternanthera nahui, Epilobium cinereum,Lachnagrostis filiformis (Heenan et al. 2009).

Phenology

Flowers and fruits mainly November–July.

Chromosome number

2n =~24, from New Zealand material (de Lange and Murray2002).

Conservation status

In mainland Australia, S. australis is widespread and commonthroughout its range and is not considered at risk. It is alsoconserved in many national parks, including Eungella, IslaGorge, Main Range, Lamington and Springbrook inQueensland, Border Ranges, Namoi-Binderay, Dorrigo, OxleyWild Rivers and Morton in New South Wales.

In New Zealand, S. australis is given a threat status (under theinformal tag or phrase name S. aff. australis (b) (AK 289786;Mangere Stonefields) of ‘Non-resident native/coloniser’(Townsend et al. 2008) in de Lange et al. (2009b). Althoughthis category predates recognition of the collection byCheesemancited above, it probably remains valid because extinction andrecolonisation appear more likely than a long-lived seed bank.

Sicyos australis sensu stricto in de Lange et al. (2009b) refersto plants on the Kermadec Islands, then thought to be conspecificwith that species, but herein shown to be S.mawhai, and the threatassessment therein must lapse.

Notes

The protologue of S. fretensis attributes the type collection toRonald Gunn. Gunn’s private herbarium included specimenscollected by others, sometimes with label data transcribedwithout collector, and these were often assumed to have beencollected by Gunn himself (Buchanan 1988).

Illustrations

D. L. Jones & B. Gray, Australian Climbing Plants, fig. 227(1977); N. G. Walsh & T. J. Entwisle (Eds), Fl. Victoria 3: 384,fig. 79h (1996); L. Forester&A.Townsend,ThreatenedPlants ofNorthland Conservancy 61 (2004); upper image.

Additional specimens examined (selection)

NORFOLK ISLAND: 1898, W. Robinson s.n. (K, NSW). NEWZEALAND: North Island: Pakaraka, end of Ludbrooke Road,Pouerua Bush, 28 Apr. 1999, K. Riddell (AK); SW of Pakaraka, edgeof ‘Jack’s Bush’, S slopes of Pouerua, P.J. Bellingham 591 (AK);Mangere, Ihumatao, Otuataua Stonefield, E.K. Cameron 11066 (AK);Coromandel, Whenuakite, off Boat Harbour Road, 10 June 2002,C. Delmiglio (AK); Te Teko, Tahuna Road, Urupa, P.J. de Lange6441 & P.B. Cashmore (AK, NSW); White Pine Bush near Awakeri,maize paddock at end of Ernest Road, 26 May 2006, M. Steverinson &P.B. Cashmore (AK, CHR, NZFRI). AUSTRALIA: Queensland: NorthKennedyDistrict: FanningRiver,NofMingela,B.R. Jackes S8643 (JCU,NSW); South KennedyDistrict: Eungella,N. Byrnes 3699& J. Clarkson(BRI); Leichhardt District: Ka Ka Mundi section, Carnarvon NationalPark, on fire trail from Salvator Rosa, M.B.Thomas 3781 & C.Elmes(BRI); Darling Downs District: Great Dividing Range, ~15 km ENE ofKillarney, I.R. Telford 12909, J.J. Bruhl & L.M. Copeland (CANB, M,NE). New South Wales: Lord Howe Island: no precise loc., J. Fullagar118 (MEL); North Coast: Nymboi–Binderay National Park, PlatypusFlat, J.J. Bruhl 2102 & F.C. Quinn (BRI, CANB, L, MO, NE); CentralCoast: Cambewarra Mountain, Beaumont, I.R. Telford 11793 (AK,CANB, NSW); South Coast: Morton National Park, SentryboxCanyon, I.R. Telford 10805 (CANB, K, NSW); Tuross River,J. Whaite 478 (NE). Victoria: Buchan River, F. Mueller s.n. MEL593278 (MEL). Tasmania: Sisters’ Island, J. Whinray 500 (CANB,MEL). CULTIVATED: ex New Zealand, Northland, WaiomioStream, Waiomio Caves, Kawiti property, P.J. de Lange 5262 (AK).

Sicyos undara I.Telford & P.Sebastian, sp. nov.

Ab affini Sicyo australi inflorescentiis foemineis pauciflorislongius et tenuius pedunculatis nec non fructibus minoribusparcius et minutius aculeatis differt.

Type: AUSTRALIA: Queensland: Cook District: UndaraVolcanic National Park, 50m W of Pinwill Cave entrance, 27March 2009, I.R. Telford 13317 & P. Sebastian; holo: BRI; iso:BISH, CANB, CHR, CNS, K, M, MO, NE, US (Fig. 5: isotypeNE).

Monoecious herbaceous climber; stems annual, up to 5mlong, up to 2.5-mm diam., sparsely hirsute with simplemulticellular hairs and glandular hairs, densely so at nodes,glabrescent. Tendrils 3- or 4-branched. Leaves: petiole10–35mm long, sparsely pilose with multicellular and glandular

194 Australian Systematic Botany I. R. H. Telford et al.

Page 8: Morphological and molecular data reveal three rather than ...renners/Telford_Sicyos_AustrSystBot2012.pdf · Morphological and molecular data reveal three rather than one species of

hairs; lamina subtriangular or broadly ovate in outline,20–117mm long, 20–94mm wide, cordate with the basalsinus broad, the lobes not overlapping, acuminate, palmately

shallowly 3- or 5-lobed, the lobes triangular, acuminate,margins dentate with apiculate teeth, hispid on both surfaces,more coarsely so on veins abaxially. Male inflorescence a

Fig. 5. Isotype of Sicyos undara I.Telford & P.Sebastian.

New species of Sicyos in the South West Pacific Australian Systematic Botany 195

Page 9: Morphological and molecular data reveal three rather than ...renners/Telford_Sicyos_AustrSystBot2012.pdf · Morphological and molecular data reveal three rather than one species of

7–25-flowered raceme, 35–155(–180)mm long; peduncle24–108mm long, glabrous; rhachis glandular hairy. Maleflowers: pedicel 5–21mm long; hypanthium broadlycampanulate, ~1.8mm in diam.; calyx lobes linear, ~0.7mmlong; corolla rotate, 3–5.7mm in diam., abaxially andadaxially glabrous, the margins glandular hairy, greenishwhite, 5-lobed, the lobes triangular; 1–1.8mm long; disc~1mm in diam.; staminal column 0.9–1.1mm long; staminalhead 0.7–0.8mm in diam. Female inflorescence a 4–8-floweredhead; peduncle (20–)32–74mmlong. Femaleflowers: subsessile;ovary ovate, ~1.2mm long, ~0.7mm in diam., echinulate withbarbed aculei; hypanthium above the constriction broadlycampanulate, ~1.3mm in diam.; calyx lobes linear, ~0.7mmlong; corolla 2.6–3mm in diam., glabrous abaxially andadaxially, the margins minutely glandular papillose, greenish-white, 5-lobed, lobes triangular, ~1mm long; disc ~0.5mm indiam; style ~1.5mm long; stigma 2-branched, the branchesrecurved. Fruit ovate or subglobose, 4–5.8mm long,2.2–2.6mm in diam., apically rounded or obtuse, the surfaceglabrous, echinate; aculei scattered, 1.4–2.8mm long, retrorselybarbed. Seeds�ovate, ~3.5mm long, 2mmwide, cream (Fig. 5).

Distribution

Sicyos undara is known only from the Undara Lava Tubes east ofMount Surprise, Queensland (Fig. 4).

Habitat and ecology

Sicyos undara grows in skeletal clay loam in boulder gulliesformed by collapsed lava tubes. The vegetation is deciduousvine thicket (dry rainforest) dominated by Brachychitonaustralis, Ficus virens and Melia azederach, with Pittosporumspinescens and Plumbago zeylanica. Undara Lava Tubes formpart of the Cainozoic McBride Volcanic Province (Griffin andMcDougall 1975), the basalts of which provide habitat for othernarrowly endemic plant species includingAuranticarpa edentataand Ipomoea saintronanensis.

Phenology

Flowers and fruits March–April.

Conservation status

Sicyos undara is known from two subpopulations at Undara, onewith a single plant observed in March 2009, and the other withtwo.A survey is necessary to establish the range of the species andsize of populations. Until assessed, a threat category of ‘CriticallyEndangered’ is suggested under the Environment Protectionand Biodiversity Conservation Act 1999 (Commonwealth).Too frequent wildfires would constitute a major threat. Thespecies is conserved in Undara Volcanic National Park.

Etymology

Named for the only known locality of the species, from the localaboriginal ‘a long way’.

Additional specimens examined

QUEENSLAND: Cook District: Yaramula Station [now UndaraVolcanic National Park], adjacent to Pinwill Cave, 80 km SW ofMount Garnet, V.J. Neldner 2780 (BRI, CANB, CNS); Undara, Wind

Tunnel, 23 March 1994, D. Hansman s.n. (CANB); Undara VolcanicNational Park,Wind Tunnel complex, I.R. Telford 13319&P. Sebastian(BRI, CANB, CNS, M, NE).

Sicyos mawhai I.Telford & P.Sebastian, sp. nov.

Ab affini Sicyo australi inflorescentiis floribusque masculismajoribus, floribus foemineis majoribus in quoque capitulonumerosioribus nec non aculeis fructuum longioribus discrepat.

Type: NEW ZEALAND: North Island: Cuvier Island, largepatch scrambling over bracken on ridgetop, West Point, 25 May1980, A.E. Wright 3583; holo:. AK!; iso: A, AD!, AKU, BISH!,CM, LTU, NA, WELT.

S. angulatus auct. nonL.: Hooker, 72 (1852); Laing, 38 (1915); Allan,319 (1961).

S. australis auct. nonEndl.: Sykes inWebb et al. (1988); deLange et al.(2009b).

S. aff. australis (a) (AK 252822): de Lange et al. (2009b).

Monoecious herbaceous climber; stems annual, up to 10mlong, up to 2.5mm in diam., sparsely hirsute with simplemulticellular celled hairs and glandular hairs, glabrescent.Tendrils 3–5-branched. Leaves: petiole 25–75(–135)mm long,minutely glandular hairy; lamina broadly ovate in outline,45–135mm long, 60–165mm wide, cordate with the basalsinus �closed by overlapping lobes, acuminate, shallowlypalmately 5- or 7-lobed, the lobes rounded–triangular, marginscoarsely dentate with apiculate teeth. Male inflorescence a14–35-flowered raceme (45–)70–85(–205)mm long; peduncle(35–)55–90(–125)mm long. Male flowers: pedicels(3.5–)6–12(–27)mm long; hypanthium broadly campanulate,3.8–4.2-mm diam.; calyx lobes linear, ~1.2mm long; corollarotate, (7–)10–13mm in diam., 5-lobed, the lobesrounded–triangular, 2–3.5mm long; disc ~2mm in diam.;staminal column 1.8–2.2mm long; staminal head 1.9–2.4mmin diam. Female inflorescence a (6–)9–14(–20)-flowered head;peduncle 12–35mm long. Female flowers: subsessile; ovaryovate, 3.2–4mm long, 2–2.6 in diam., echinulate with barbedaculei; hypanthium above the constriction broadly campanulate,~2.4mm in diam., abaxially scabridulous; calyx lobes linear,~0.8mm long; corolla 4–6.8mm in diam., glabrous abaxially andadaxially, glandular papillose on margins, white, 5-lobed; lobestriangular–ovate, 2–2.3mm long; disc ~1.3mm in diam.; style~1.8mm long; stigma 2-branched, the branches recurved. Fruitovate, 8–13mm long, 4–5.8mm in diam., obtuse or subacute,the surface glabrous or sparsely and minutely hairy, echinate;aculei dense, (2.2–)3.6–6mm long, retrorsely barbed. Seeds�ellipsoidal, 5.6–6mm long, 3.8–4.5mm wide, brown. (Fig. 6).

Distribution

Sicyos mawhai occurred formerly on mainland North Islandand the Marlborough Sounds area of South Island. Solander(1782) recorded collections made in 1769–1770 on CaptainJames Cook’s first voyage to the Pacific from North Island,including Tigadu (Anaura Bay), Tolaga (Tolaga Bay),Opuragai (Mercury Bay), Oohuragi (Firth of Thames, WaihouRiver), Motuaro (Bay of Islands, Motuarohia Island), and SouthIsland, including Totaranui (Queen Charlotte Sound). Thespecies is now confined to offshore islands, islets and rock

196 Australian Systematic Botany I. R. H. Telford et al.

Page 10: Morphological and molecular data reveal three rather than ...renners/Telford_Sicyos_AustrSystBot2012.pdf · Morphological and molecular data reveal three rather than one species of

stacks off New Zealand’s North Island from the Bay of Plentynorthwards, and on the more remote Three Kings and KermadecIslands groups (Fig. 4).

Habitat and ecology

Sicyos mawhai is a coastal species that has been collected frombeach strands, including boulder beaches and banks, from coastal

forest (especially forest margins), coastal shrubland and onoffshore islands within sea-bird colonies, most notably, thenesting grounds of petrels (Pterodroma spp., Puffinus spp.).Within its beach habitat, the species has been collected trailingthrough and over Coprosma repens, Myoporum laetum, Ficinianodosa and Phormium tenax, and in forest it often associateswith Corynocarpus laevigatus, Entelea arborescens,Myoporum

Fig. 6. Holotype of Sicyos mawhai I.Telford & P.Sebastian.

New species of Sicyos in the South West Pacific Australian Systematic Botany 197

Page 11: Morphological and molecular data reveal three rather than ...renners/Telford_Sicyos_AustrSystBot2012.pdf · Morphological and molecular data reveal three rather than one species of

laetum, Melicytus ramiflorus, Macropiper excelsum subsp.excelsum, Pseudopanax lessonii and Metrosideros excelsa. Onthe Three Kings Islands, it often grows with Brachyglottisaborescens, Coprosma repens, Metrosideros excelsa andKunzea aff. ericoides (e), and at open sites with Ipomoeacairica. On the Kermadec Islands, it is a common species onRaoul Island where it mostly grows in canopy gaps and alongforest–shrubland margins with Coprosma petiolata, Myrsinekermadecensis, Myoporum rapense subsp. kermadecense,Macropiper excelsum subsp. psittacorum, Metrosideroskermadecensis and Melicytus aff. ramiflorus. On the smallerislands of the Kermadec group, it is also to be found trailingthrough petrel colonies, where it associates with Cyperusinsularis, Scaevola gracilis, Solanum nodiflorum, Ipomoeacairica, I. pescaprae subsp. brasiliensis and Tetragoniatetragonioides. Less commonly, like S. australis occurrencesin New Zealand proper, S. mawhai has been collected frommodified islands such as the Mokohinau Island group growingin induced pasture and wasteland, growing with grasses such asBromus willdenowii, Dactylis glomerata, bracken (Pteridiumesculentum) and black berries (Rubus fruticosus agg.)

Phenology

Sicyos mawhai flowers and fruits mainly in December–May, andalso throughout the year in suitable conditions.

Chromosome number

2n = 26 (de Lange and Murray 2002).

Conservation status

Cameron (1992) recorded the shrinking distribution ofS. mawhai, with extinction at the southern end of its historicalrange. No populations survive on mainland North or SouthIslands (P. J. de Lange, pers. comm.); note that the 1992Auckland collection cited by Cameron (1992) refers toS. australis sensu stricto. Plants are susceptible to CMV andZYMV (Delmiglio and Pearson 2006), and this appears toweaken populations and has been used to explain the rapiddecline of the species. Fruit and seedlings are eaten byrodents; removal of these vermin allows recolonisation(P. J. de Lange, pers. obs.). The species is rated ‘At Risk/Relict’ in de Lange et al. (2009b) as S. aff. australis (a) (AK252822).

Notes

Sicyos mawhai and S. australis occur almost sympatrically, butcan readily be distinguished vegetatively; the former has leaves ofa thicker texture, the lobes more rounded, with the basal pairsometimes overlapping across the sinus, as well as by the floraland fruit differences presented in the key below.

Unfortunately, the Banks and Solander specimens housed intheMuseum ofNewZealand Te Papa Tongarewa (WELT; http://collections.tepapa.govt.nz/theme.aspx?irn = 2152, accessed 12October 2011) and the Natural History Museum, London(BM), have no precise locality on the sheets. The single BMspecimen located has a quote from Solander (1782) attachedlisting the collection sites cited above.

Etymology

Named for the Maori name for the plant, m�awhai, which means‘wavy’ or ‘curly’, possibly in reference to the plant’s habit and/ortendrils. M�awhai is also said to mean ‘useless’ (de Lange 2011),perhaps in reference to a Tainui legend. The namem�awhai is alsowidely applied to Cassytha paniculata R.Br. (de Lange 2011).

Illustrations

F.B. Sampson, Early New Zealand Botanical Art t. 28 (1985).L. Forester & A. Townsend, Threatened Plants of NorthlandConservancy 61 (2004); two lower images.

Additional specimens examined (selection)

NEW ZEALAND: Kermadec Islands: Raoul Island orchard,W.R. Sykes188/K (CHR, K); Raoul Island, Fishing Rock,P.J. de Lange K490 (AK);Macauley Island,MountHaszard,G.Taylor 13 (CHR);Macauley Island,Macauley Plateau, P.J. de Lange K817 (AK). Three Kings Islands:West Island, A.E. Wright 6061 (AK); Great Island, J.S. Edwards s.n.AK 44826 (AK). North Island: Te Paki, Kapowairua/Spirits Bay,Maungapiko Hill, 29 September 2006, L. Bridson & W. Holland(AK); Poor Knights Islands, Aorangi Island, Crater Bay, 2 January1978, G.N. Park (CHR); Hen Island, February 1968, D. Merton(CHR); Little Barrier Island, near West Landing, W.R. Sykes 147/85(CHR); Fanal Island, A.E. Wright 3146 (AK, HO, WELT); CoromandelIslands,Motukarikitahi, A.E. Esler 3584 (CHR). CULTIVATED: OratiaNative Plant Nursery, Auckland, J.J. Bruhl 2915 (AK, CHR, M, NE).

Key to Sicyos in the South West Pacific region

Male flowers mostly >10-mm diam..............................................S. mawhaiMale flowers mostly <10-mm diam.

Fruit <6-mm long, attenuate; peduncle of female inflorescence mostly<25-mm long................................................................................S. australis

Fruit >6mm long, apically rounded; peduncle of female inflorescencemostly >25mm long............................................................. S. undara

Acknowledgements

We thank the directors of the herbariaAK,BM,BRI, CANB,CHR,HO, JCU,K, MEL, NSW and W for the loan of specimens and use of facilities, and theQueensland Department of Environment and Resource Management forpermission to collect. We thank Werner Greuter for the impeccable Latin.Thanks go to JohnHunnex of theNatural HistoryMuseum, London (BM), foran image of the Banks and Solander specimen. Ian R. H. Telford expressessincere thanks to Mike Crisp, Research School of Biology, The AustralianNational University, for his inspiration through time shared at the AustralianNational Botanic Gardens and as supervisor for the initial Sicyosmorphometric analysis.

References

Allan HH (1961) ‘Flora of New Zealand.’ (Government Printer: Wellington,New Zealand)

Bailey FM (1900) ‘The Queensland flora.’ (Queensland Government:Brisbane)

Beadle NCW (1976) ‘Students flora of north eastern New South Wales.’(Botany Department, University of New England: Armidale, NSW)

BeadleNCW,EvansOD,CarolinRC(1962) ‘Handbookof thevascular plantsof the Sydney District and the Blue Mountains.’ (Brown Gem Print:Armidale, NSW)

Belbin L (1990a) ‘PATN: a user’s guide.’ (Division ofWildlife and Ecology,CSIRO: Canberra)

Belbin L (1990b) ‘PATN: a technical reference.’ (Division of Wildlife andEcology, CSIRO: Canberra)

198 Australian Systematic Botany I. R. H. Telford et al.

Page 12: Morphological and molecular data reveal three rather than ...renners/Telford_Sicyos_AustrSystBot2012.pdf · Morphological and molecular data reveal three rather than one species of

Belbin L, Collins A (2006) ‘PATN. V.3.1.’ (Blatant Fabrications Pty Ltd:Brisbane)

Bentham G (1866) ‘Flora Australiensis.’ (Lovell Reeve & Co.: London)Brandon A, de Lange P, Townsend A (2004) ‘Threatened plants of the

Waikato Conservancy.’ (Department of Conservation: Wellington,New Zealand)

Buchanan AM (1988) The Tasmanian collecting localities of Ronald Gunnand Joseph Milligan. Tasmanian Herbarium Occasional Publication 1,1–56.

Cameron EK (1992) Decline of mawhai (Sicyos australis). New ZealandBotanical Society Newsletter 28, 11–12.

Chandler GT, Plunkett GM, Pinney SM, Cayzer LW, Gemmill CEC (2007)Molecular and morphological agreement in Pittosporaceae: phylogeneticanalysis with nuclear ITS and plastid trnL–tmF sequence data. AustralianSystematic Botany 20, 390–401. doi:10.1071/SB07004

Connor HE, Edgar E (1987) Name changes in the indigenous New ZealandFlora, 1960–1986, and nomina nova IV, 1983–1986. New ZealandJournal of Botany 25, 115–171. doi:10.1080/0028825X.1987.10409961

Curtis WM (1963) ‘The Students flora of Tasmania.’ (Government Printer:Hobart)

de Lange PJ (1997)Gratiola pedunculata (Scrophulariaceae); a new additionto the New Zealand flora. New Zealand Journal of Botany 35, 317–322.doi:10.1080/0028825X.1997.10410157

de Lange PJ (2011)Cassytha paniculata a possible indigenous biocontrol forgorse (Ulex europeus)? Trilepidea 93, 9–10.

de Lange PJ, Murray BG (2002) Contributions to a chromosome atlasof the New Zealand flora – 37. Miscellaneous families. New ZealandJournal of Botany 40, 1–23. doi:10.1080/0028825X.2002.9512767

de Lange PJ, Norton DA (1998) Revisiting rarity: a botanical perspective onthemeanings of rarity and the classification ofNewZealand’s uncommonplants. Royal Society of New Zealand Miscellaneous Series 48, 145–159.

de Lange PJ, Norton DA, Heenan PB, Courtney SP, Molloy BPJ, Ogle CC,Rance BD, Johnson PN, Hitchmough R (2004) Threatened anduncommon plants of New Zealand. New Zealand Journal of Botany42, 45–76. doi:10.1080/0028825X.2004.9512890

deLangePJ, SawyerWD,Rolfe JR (2006) ‘NewZealand indigenousvascularplant checklist.’ (New Zealand Plant Conservation Network:Wellington,New Zealand)

de Lange PJ, Heenan PB, Townsend AJ (2009a) Rorippa laciniata(Brassicaceae), a new addition to the flora of New Zealand. NewZealand Journal of Botany 47, 133–137.doi:10.1080/00288250909509800

deLangePJ,NortonDA,CourtneySP,HeenanPB,Barklla JW,CameronEK,Hitchmough R, Townsend AJ (2009b) Threatened and uncommon plantsof New Zealand (2008 revision). New Zealand Journal of Botany 47,61–96. doi:10.1080/00288250909509794

de Lange PJ, SmissenRD,Wagstaff SJ, KeelingDJ,MurrayBG, ToelkenHR(2010) Amolecular phylogeny and infrageneric classification forKunzea(Myrtaceae) inferred from rDNA ITS and ETS sequences. AustralianSystematic Botany 23, 309–319. doi:10.1071/SB10019

de Lange PJ, Townsend AJ, Rolfe JR (2011) Crassula nutans var. minus(Crassulaceae) a new trans-Tasman natural weed arrival to northernNew Zealand. New Zealand Journal of Botany 49, 361–366.doi:10.1080/0028825X.2011.574708

Delmiglio C, Pearson MN (2006) Effects and incidence of cucumber mosaicvirus, watermelon mosaic virus and zucchini yellow mosaic virus inNew Zealand’s only native cucurbit, Sicyos australis. Australasian PlantPathology 35, 29–35. doi:10.1071/AP05098

Endlicher SFL (1833) ‘Prodromus flora Norfolkicae.’ (F. Beck: Vienna)Forester L, Townsend A (2004) ‘Threatened plants of Northland

Conservancy.’ (Department of Conservation: Wellington, New Zealand)Green PS (1994) Oceanic Islands 1. In ‘Flora of Australia 49’. (Ed. AJG

Wilson). (Australian Government Publishing Service: Canberra)

Griffin TJ, McDougall I (1975) Geochronology of the Cainozoic McBrideVolcanic Provinve, northern Queensland. Journal of the GeologicalSociety of Australia 22, 387–396. doi:10.1080/00167617508728905

HeenanPB,ForesterLJ (1997)Mazuspumilio (Scrophulariaceae), an additionto the indigenous flora of New Zealand. New Zealand Journal of Botany35, 437–440. doi:10.1080/0028825X.1987.10410168

Heenan PB, de Lange PJ, Keeling JK (2009) Alternanthera nahui – a newspecies of Amaranthaceae indigenous to New Zealand. New ZealandJournal of Botany 47, 97–105. doi:10.1080/00288250909509795

Hooker JD (1847) FloraTasmaniae spicelegium.Hooker’s London Journal ofBotany 6, 473–474.

Hooker JD (1856) ‘The botany of the Antarctic voyages of H.M. discoveryships Erebus and Terror. 3. Flora Tasmaniae.’ (Lovell Reeve & Co.:London)

Hooker JD (1860) ‘The botany of the Antarctic voyage. 3. Flora Tasmaniae.’(Lovell Reeve & Co.: London)

Perrie LR, Brownsey PJ, Lockhart PJ, Brown EA, Large MF (2003)Biogeography of temperate Australasian Polystichum ferns inferredfrom chloroplast sequence and AFLP. Journal of Biogeography 30,1729–1736. doi:10.1046/j.1365-2699.2003.00938.x

Prebble JM, Cupido CN, Meudt HM, Garnock-Jones PJ (2011) Firstphylogenetic and biogeographical study of the southern bluebells(Wahlenbergia, Campanulaceae). Molecular Phylogenetics andEvolution 59, 636–648. doi:10.1016/j.ympev.2011.03.013

Schaefer H, Renner SS (2011) Phylogenetic relationships in the orderCucurbitales and a new classification of the gourd family(Cucurbitaceae). Taxon 60, 122–138.

Schaefer H, Heibl C, Renner SS (2009) Gourds afloat: a dated phylogenyreveals anAsian origin of the gourd family (Cucurbitaceae) and numerousoversea dispersal events.Proceedings. Biological Sciences276, 843–851.doi:10.1098/rspb.2008.1447

Sebastian P, Schaefer H, Telford IRH, Renner SS (2010a) Cucumber(Cucumis sativus) and melon (C. melo) have numerous wild relativesin Asia and Australia, and the sister species of melon is from Australia.Proceedings of the National Academy of Sciences of the United States ofAmerica 107, 14 269–14 273. doi:10.1073/pnas.1005338107

Sebastian PM, Schaefer H, Renner SS (2010b) Darwin’s Galapagos gourd:providing new insights 175 years after his visit. Journal of Biogeography37, 975–978. doi:10.1111/j.1365-2699.2010.02270.x

Sebastian P, Schaefer H, Lira R, Telford IRH, Renner SS (2012) Radiationfollowing long distance dispersal: the contributions of time, opportunity,and diaspore morphology in Sicyos (Cucurbitaceae). Journal ofBiogeography doi:10.1111/j.1365-2699.2012.02695.x

Sneath PHA, Sokal RR (1973) ‘Numerical taxonomy: the principles andpractice of numerical classification.’ (W.H. Freeman and Co.: SanFrancisco)

Solander DC (1782) Primitiae florae Novae Zelandiae. In ‘Floras of thecountries visited by Captain James Cook’s first voyage’. pp. 381–620.(Botany Library, Natural History Museum: London)

Stamatakis A (2006) RAxML-VI-HPC: maximum likelihood-basedphylogenetic analysis with thousands of taxa and mixed models.Bioinformatics 22, 2688–2690. doi:10.1093/bioinformatics/btl446

StuessyTF (1990) ‘Plant taxonomy: the systematic evaluation of comparativedata.’ (Columbia University Press: New York)

Sykes WR, de Lange PJ (1993) Plectranthus parviflorusWilld. (Lamiaceae)in New Zealand. New Zealand Journal of Botany 31, 11–14.doi:10.1080/0028825X.1993.10419528

Tay ML, Meudt HM, Garnock-Jones PJ, Ritchie PA (2010) DNA sequencesfrom three genomes reveal multiple long-distance dispersals and non-monophyly of sections in Australasian Plantago (Plantaginaceae).Australian Systematic Botany 23, 47–68. doi:10.1071/SB09040

Telford IRH (1982) Cucurbitaceae. In ‘Flora of Australia 8’. (Ed. ASGeorge)pp. 158–198. (Australian Government Publishing Service: Canberra)

New species of Sicyos in the South West Pacific Australian Systematic Botany 199

Page 13: Morphological and molecular data reveal three rather than ...renners/Telford_Sicyos_AustrSystBot2012.pdf · Morphological and molecular data reveal three rather than one species of

TownsendAJ, deLangePJ,NortonDA,Molloy J,MiskellyC,DuffyC (2008)‘The New Zealand threat classification system manual.’ (Department ofConservation: Wellington, New Zealand)

Webb CJ, Sykes WR, Garnock-Jones PJ (1988) ‘Flora of New Zealand.’(Botany Division, DSIR: Christchurch, New Zealand)

Willis JH (1972) ‘A handbook to plants in Victoria.’ (Melbourne UniversityPress: Melbourne)

200 Australian Systematic Botany I. R. H. Telford et al.

Page 14: Morphological and molecular data reveal three rather than ...renners/Telford_Sicyos_AustrSystBot2012.pdf · Morphological and molecular data reveal three rather than one species of

Appendix 1. Dataset used in the phenetic analysis of Sicyos in Australia and the South West PacificSee Table 1 for voucher data. See Table 2 for characters relating to numbers in the header row. OTU, organisational taxonomic unit

OTU Character1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

NZ1 24 9 4.5 10.2 5.3 5.8 42 5.8 93 66 14 16.0 12.8 1.5 2.3NZ2 23 8 4.2 10.0 5.4 5.4 50 6.0 135 95 25 10.9 8.7 1.5 2.4NZ3 25 8 9.8 6.2 4.5 46 6.2 124 72 17 14.5 10.6 1.3 2.4NZ4 26 8 4.3 8.5 6.0 4.2 53 6.4 152 85 35 11.0 10.4 1.8 2.2NZ5 25 8 4.8 9.2 6.2 5.3 48 5.5 110 72 18 12.4 9.2 1.5 2.0NZ6 14 10 7.8 3.8 4.8 26 4.8 37 22 8 4.8NZ7 12 7 3.0 7.2 3.2 4.2 27 4.2 53 42 6 5.0NZ8 25 9 10.5 6.5 4.6 52 6.5 140 88 19 12.0 10.0 2.2 2.8NZ9 12 10 4.3 8.0 3.8 4.2 20 4.0 55 22 11 4.2 8.2 1.4 1.7NZ10 16 8 8.8 3.6 3.8 17 4.8 106 68 21 4.5NZ11 24 14 4.0 11.3 6.5 4.5 38 6.5 116 65 22 11.2 11.8NZ12 34 15 4.8 11.8 6.7 5.5 27 104 78 15 12.5 10.0KI1 24 8 4.5 11.8 6.0 5.5 36 6.8 110 79 16 12.3 10.5KI2 18 8 9.0 5.8 4.6 40 6.0 92 84 23 9.5 10.6 1.3 2.4KI3 23 9 9.2 5.2 4.8 56 6.5 126 52 22 10.8 10.2 2.0 1.9NI1 23 9 2.6 7.8 3.7 7.0 25 4.8 124 93 13 9.3 4.5NI2 16 8 2.5 7.2 3.2 5.5 24 61 49 9 6.5 5.0LH1 22 10 3.0 7.8 3.5 6.5 24 4.7 62 47 8 6.2 6.8LH2 28 8 7.2 3.3 5.5 28 78 61 10 6.0 6.2LH3 17 7 7.0 3.0 6.9 26 4.4 80 36 14 5.8 5.8TS1 10 8 3.4 7.8 3.7 6.0 23 57 35 11 5.2 6.3 1.8 2.2TS2 11 10 2.8 7.3 3.5 7.5 22 4.6 68 40 12 3.5 6.0VG1 21 7 2.7 7.2 3.5 6.0 23 4.5 79 67 12 5.0 6.5VG2 14 8 2.5 6.5 3.0 5.8 21 67 50 15 3.5 5.8 1.5 1.8NS1 13 6 5.5 2.7 4.5 19 4.0 60 53 7 4.8 7.4 1.8 1.7NS2 23 7 2.8 7.5 3.6 7.0 28 4.8 104 93 12 8.5 6.4 1.5 1.7NS3 15 8 2.4 8.2 3.8 3.5 22 5.0 87 65 11 6.8 6.0QL1 13 10 2.4 7.0 3.7 3.5 28 4.2 90 78 13 6.5 5.3QL2 15 6 3.0 7.5 3.8 6.5 21 79 55 9 4.5 5.8QL3 17 7 6.5 3.5 6.8 24 4.0 114 89 11 6.8 5.3QL4 11 7 2.5 8.3 4.0 7.0 22 93 75 12 4.0 4.2 1.6 1.7QL5 36 5 2.0 4.8 2.8 2.6 13 3.5 60 37 14 9.6 4.2 1.1 0.8QL6 27 4 5.3 3.0 2.8 11 34 27 12 6.0 3.7 0.9 0.7QL7 57 6 2.4 4.7 2.9 2.8 12 3.2 95 22 11.3 4.8 0.9 0.7

New species of Sicyos in the South West Pacific Australian Systematic Botany 201

www.publish.csiro.au/journals/asb